Wikipedia:Administrators' noticeboard/Δ

See also the archived discussion here regarding betacommandbot and NFCC10c This a centralized page of all discussion pertaining to User:Betacommand and his bot User:BetacommandBot. These discussions were previously on the Administrators' Incidents Noticeboard, but haved been moved here in the interests of space and navigation. This page is for participating and discussing in a civil manner all matters that relate to Betacommand and his bot.

Please do not use this page to attack these users instead of participating in civil discussion. This is not a complaints department. If you have a private matter that concerns operation of BetacommandBot, please discuss it with Betacommand on his talk page.

11,000 images tagged NFUR in one day

Note: this issue is now at Wikipedia:Bots/BetaCommandBot and NFCC 10 cMickMacNee (talk) 19:13, 17 February 2008 (UTC)

And 6,000 yesterday [[1]]. All to be modified in 7 days or face deletion. How can that possibly be squared with with betacommandbot's stated desire of not wanting to delete images? I have had 7 tags appear in my watchlist today, and looking at comments on the bot talk page from another user with 7 and an 11, that seems about average, with one admin getting an unholy 65 direct talk page tags over the 2 days. I, like other well meaning editors who have not uploaded these images do have a willingness to investigate, fix and educate where necessary, but are simply put off by the sheer number of tags in a short space of time, with no tags applied for the preceeding 5 days. The intransigence/absence of betacommand on the bot talk page is also frustrating many. I have also seen the speedy deletion of images purely on the say so of uploaders getting fed up with the bot and just jacking it in. The majority of uploaders just seem baffled/confused/annoyed, with none really making any headway faced with tag explosions like this. Many image tags wont even be seen by the uploader in 7 days. Something seriously needs to be done about the way this bot is operated/scheduled. MickMacNee (talk) 04:36, 14 February 2008 (UTC)

I am glad that I am not the only one angered by this. I find it incredibly disruptive and destructive to Wiki. While I understand that the policy is clearly written, the policy was recently changed, but in most of these cases, the image does qualify as fair use, but it simply needs to mention the specific article its already used in. Why can the bot just do this automatically? Some of the users it is notifying aren't even active anymore to change the image. --Veritas (talk) 04:41, 14 February 2008 (UTC)

This policy has been effect since mid 2006. and bots cannot write valid rationales. β^_command 04:46, 14 February 2008 (UTC)

I wholeheartedly agree with this. While tagging images is necessary, tagging them at this rate overwhelms the ability of Wikipedians to deal with fixing fair use criteria, the ability of admins to delete the images without a script, and the ability of people to review that the bot did tag the image properly. Please slow down the bot to reasonable levels. Titoxd^{(?!? - cool stuff)} 04:48, 14 February 2008 (UTC)

Even still, why does a bot even need to do this? We're sacrificing reason for efficiency at a destructive and brutal pace. If it is not possible for a bot to operate in a constructive manner it should be shut down. It is clearly not capable of fixing problems, but only pointing them out and often to users who are no longer active and unable to do anything about it. Thus, what happens? Images are deleted and we will wait a year or so until someone else comes along to add new images. A ridiculous waste of editor time thus indicating that the bot really isn't all that efficient in the long-term. --Veritas (talk) 04:51, 14 February 2008 (UTC)

I agree we need a bot to tag these to conform to policy. But given that we have 40+ days, I'd urger that we work on this current backlog past the 7 day limit. Its worked before and can work again. Also, a schedule of when runs would occur, would be most helpful. MBisanz ^talk 04:54, 14 February 2008 (UTC)

This is an example of the intransigence. Just completely brushes aside any criticism. Not everyone has been here since mid 2006, especially many new uploaders, and many images pre-date 2006. This bot has got to be one of the most divisive things I've seen on WP, and yet, no direction is given to a precise summary as to consensus for this bot's usage. This is not about the policy, this is about the effectiveness of the bot to apply it. I have seen nowhere in the myriad of talk pages about this, any actual analysis of whether the bot statistically meets it's stated aim of not seeing an end result of deleted images, rather than being just a very fast and efficient deleter of content, valid or not. Denying that the bot plays any role in the admittedly final human decision to delete is just outrageous, especially given the timescale, numbers, and the reactions to it's tags from users of all experience levels. MickMacNee (talk) 04:57, 14 February 2008 (UTC)

Don't get me wrong, I'm not saying that BC is being intransigent, but that the use of this bot must be monitored and restricted due to its damaging side-effects that tags images faster than other, active, editors can step in to fix the backlog of destruction it leaves behind. BC seems to have a tendency to closely stick to policy. That's the safe route and there's nothing wrong with it save the fact that it doesn't always account for the human element of Wikipedia. --Veritas (talk) 05:01, 14 February 2008 (UTC)

Veritas, when ever BCBot makes a run that day's limit is normally extended. since we are getting closer I thought Id identify as many as possible so they could have time to fix them prior to the deadline. β^_command 05:03, 14 February 2008 (UTC)

I'm not completely sure of the technical aspects of how the bot works, but I know that I haven't had any complaints about it until the past couple days so I'm not sure if something has changed recently. It is troubling though since I do feel that the bot's actions are negating is usefulness. Perhaps we can extend these 7 day deadlines until the back-log is cleaned up? --Veritas (talk) 05:06, 14 February 2008 (UTC)

PS, why is this convo going on at AN as well as AN/I? --Veritas (talk) 05:09, 14 February 2008 (UTC)

That convo looks like a general gripe, I am raising issue with the incident of NFUR tagging 15,000 images in 1/2 days, and the multitude of issues that comes from that. MickMacNee (talk) 05:22, 14 February 2008 (UTC)

(ecx3)That is reasonable to ID them now so there is a greater opportunity of the uploader seeing and fixing them. Would it be possible to extend the delete date in the tag to say 14 or 21 days to reduce the incidence of uninvolved Admin X wandering across the image, not knowing about the extension on the Cat page, and deleting it? MBisanz ^talk 05:13, 14 February 2008 (UTC)

This issue has come up several times. Part of the problem is that instead of tagging a few hundred images each day, the bot runs on this task only once in a while but tags many thousands of images instead. There is no need to extend the deadline, really, this will create a huge backlog when it comes time to delete these things anyway, so the deadline will effectively slip. It would be nice, though, if this task was run more frequently but at a lower rate. Mangojuice^talk 05:11, 14 February 2008 (UTC)

What use is that if the tag says 7 days, and people just chuck it in because they haven't a clue what to do? There is no information anywhere about this deadline we are getting close to. Of the 7 images I have seen marked, 2 required a 10 second mod, most had specific tags that can direct to organised interest groups such as applying dvd cover templates, and none actually deserve deletion. Some even date from Nov/Dec 2005, so why are you rushing this through now? Why has there been no attention earlier by people who know what needs to be done to satisfy this bot? MickMacNee (talk) 05:12, 14 February 2008 (UTC)

The copyright tags on the images clearly state what is required of the uploader to fulfill policy. There are also a few hundred active admins and a help desk for anyone confused to contact about what to do. I really don't see how this is something the bot can be faulted for. Lara❤Love 05:20, 14 February 2008 (UTC)

I agree and disagree at the same time. Ideally, that is how it should work, but clearly not how it works in practice. I am tending to favor Mango's suggestion here that the bot run the task more often. This would create a far more agreeable tagging to editor-intervention ratio that is actually maintainable. --Veritas (talk) 05:22, 14 February 2008 (UTC)

I am planning more runs, I had a massive run about a month ago. this cleared what I missed last time and future runs should not be as bad, Im hoping to run this ~2 times a week. β^_command 05:28, 14 February 2008 (UTC)

That fits from the admin who said it wasnt her job to explain the procedures on the project page she patrolled, necessitating edits of the instructions by the user themselves. MickMacNee (talk) 05:27, 14 February 2008 (UTC)

What project? Lara❤Love 05:48, 14 February 2008 (UTC)

There's nothing at all wrong with this. I just cleaned up one myself, and i'm lousy at writing these things. Luckily, it's pretty much boilerplate, and if you can remember most of it, then you're fine, and if you can Cut n paste to fit the use of a given image, then it's easier. there's a bit of mix and match, running down your entire notification list with one clipboard text-set would be bad, but it's not that tough to do. And this doesn't affect every editor, just those who never read the full 'how to upload images' guides, or who disregarded the tough part about writing something. Those editors will get the notes, as will all editors who have any article with such an image in their watchlist. I just grabbed one, and will probably find others in the next few days. This isn't as big a deal as it is being made out to be. And, it does keep WP out of legal dangers and hassles. Would it be nice if the tagging bot ran more often than now? Maybe. It would probably irritate the serial violators more and a few might quit uploading, and a few might start writing, and then we'd have less for the rest of us to do. Or maybe not. But that we have 18K unfurred (sounds dirty, don't it?) images is even worse. ThuranX (talk) 05:28, 14 February 2008 (UTC)

I know how to write FUR's, it's easy, I did one right off the bat, but when 7 pop up in your watchlist in one day, in the middle of a big article project yourself, none of which you uploaded but want kept, all of which are not violaters, some pre-dating the policy at all, some that could blatantly be fixed in seconds, some you know full well will not get the attention (it's not all project based images), but the effect of lumping it all in one, and with attitude of the operator, and the complete lack of historical guidance, you have to wonder if the stated aim of not wanting to delete images is correct. Had I just seen one or two flagged, I probably wouldn't be here now, and be none the wiser as to the tip of the iceberg surrounding this bot. MickMacNee (talk) 05:42, 14 February 2008 (UTC)

Indeed. Three of us on the ice hockey project fixed about 40 images in a very short period of time. There is no reason why we have to hold to a hard 7 day deadline if the backlog cannot be adequately tackled, but if editors with an interest in affected articles and projects are dedicated, it can be handled fairly easily. It just takes some time. Resolute 05:32, 14 February 2008 (UTC)

Beta just said that he's planning on having the bots do more frequent runs. I am in favor of a large number of these being tagged now so long as the bot does regular and fairly frequent runs in the future so that users interested in maintenance have time to step in to fix the often minor adjustments that are required. --Veritas (talk) 05:35, 14 February 2008 (UTC)

Ah, for the day when there'll be a "11,000 images fixed in one day" thread. MickMacNee, if these images are so easy to fix, why are they not being fixed? Why is it so that years down the track, this bot finds thousands upon thousands of bad images each time it runs? --bainer (talk) 05:30, 14 February 2008 (UTC)

There are clear issues surrounding the lack of links and interest from the bot operator to inform aggrieved users of the past discussions about this bot (are there any agreed consensuses? Not about the policy, but the specific use of the bot). The issue is not clear cut when most advocates of the bot are expert admins, and most aggrieved by it are new editors. The issue is clearly affecting many many editors. Just needing a 17 point not my fault header on the bot talk page that has an archive for every month should tell people something is wrong. I say again, has any analysis been done on the effectiveness of this bot on gaining rationales to meet policy? As opposed to just hastening deletion of perfectly acceptable material? MickMacNee (talk) 05:35, 14 February 2008 (UTC)

The material is not acceptable if it's not fixed; that's the policy, and the bot is just alerting admins that there is a deficiency. As for Betacommand and his communication skills, yes, they leave something to be desired, and I've had my own 2 cents to say about that a time or two. Nonetheless, the bot shouldn't stop running just because it's delivering the bad news that there is a lot of stuff that needs fixing). Mangojuice^talk 05:42, 14 February 2008 (UTC)

You absolutely cannot ignore the scheduling issue, and the complete lack of any historical links or summaries. 'see the archives' is all you will get, if you're lucky. It has changed my willingness to fix things today, it has caused an admin with 60 odd tags to flip out, and caused others to just give up and say they want the images deleted rather than deal with bcb anymore, which were probably fixable. MickMacNee (talk) 05:47, 14 February 2008 (UTC)

In the time you've spent debating here, you could have fixed a few dozen images. Also, Wikipedia isn't really hurt by copyrighted images being deleted, appropriate for fair use or not, so it's not damaging the 'pedia. It's an inconvenience for some, obviously, but it's a bigger issue for Wikipedia to be improperly using copyrighted images. Lara❤Love 05:54, 14 February 2008 (UTC)

• I know this probably doesn't matter to the deletionists, but there are a lot of editors (myself included) who have simply given up on uploading images. BCBot takes a subsection of a subsection (10C is it?) and uses it to tag tons of perfectly acceptable images for deletion. Then we have admins who simply plow through the backlog, deleting without bothering to check if it's something simple to correct. I'm done with trying to add images to this project, at least until BCBot is reined in or shut down completely. Bellwether ^B_C 05:59, 14 February 2008 (UTC)

It's silly to say that deleting images doesn't hurt the articles. It's also a little strange to give up on uploading images. It's not hard to do it the right way, if the image is truly appropriate per policy and guidelines. I won't opine on the schedule though. Sooner or later we do need to get to a resting place where most of the images have the data they need. Wikidemo (talk) 06:01, 14 February 2008 (UTC)

The problem is that many of the images at hand were uploaded correctly, then the guidelines changed, and suddenly there's a bot coming by screaming about the coming image apocalypse. BCB is bad diplomacy; Wikipedia rules are in flux, and when they change, thousands of articles are affected. How do we deal with this? Do we make it easy to bring the old ones in to compliance with the new rules? Maybe get a bot to fix that? Or do we get lazy and just send a bot out to tell editors they're wrong and their work is being erased. Even if the form is easy now, why should I trust it? Why should I assume the rules aren't going to change again such that the band name or record number has to be included with every album cover, and suddenly BCB rolls by with another 11k nastygrams per day? Allowing this to proceed unabated costs Wikipedia the trust of its editors. —Torc. ^(Talk.) 11:00, 14 February 2008 (UTC)

Since when is the policy change BCB's 'fault'? Your issue is with Jimbo and the board.TheRedPenOfDoom (talk) 23:09, 21 February 2008 (UTC)

Isn't really hurt? A picture tells a thousand words. How about the editors chucking it in left right and centre every time this thing runs like this? In the time you've spent debating here, you could have fixed a few dozen images, As said above, I do fix images, but on days like today you think what's the point?, especially when you research and see the background to this issue, and see the massive effect one user is allowed to have without comeback. The copyright issue is all well and good, but again, this bot today has tagged in my watchlist sample, 30% of images that were loaded in 2005, and not a single actual copyright violator (after modification to meet a seemingly ever changing policy). Would you create content if you knew you might have to do it 3 times after each deletion retrieval? How does anyone expect any other jobs to get done in the face of that kind of lunacy. MickMacNee (talk) 06:04, 14 February 2008 (UTC)

• My main problem is that he's taking a subsection of a subsection, and applying it like it was the effin' tablets brought down from the mountain with Moses. And the admins that mindlessly plow through the backlog, without checking each image carefully aren't doing the project any favors either. There's just next to no common sense applied here. Bellwether ^B_C 06:07, 14 February 2008 (UTC)

In the time you've spent... is such a horrendously fallacious argument to make. In the time I've spent correcting correctly tagged images so the bot stops bitching I could have made substantive contributions to the project. We have X hours of editor manpower, yet a machine insists on determining exactly how we'll get to use those hours, and it has decided that forcing editors at gunpoint to perform dull, meaningless bureaucracy is the best use of our resources. And those hours are gone: whatever it decides, plus the lost hours from the editors who just bail from Wiki altogether when they get spammed by a machine, plus the lost hours from editors who will eventually have to reupload the same image because there was nothing wrong with it before aside from botardedness. —Torc. ^(Talk.) 10:48, 14 February 2008 (UTC)

What is silly to me is to stop uploading images because of BCBot. Just upload them under the policy. They don't get tagged if they meet the requirements. Lara❤Love 06:13, 14 February 2008 (UTC)

• No, what is silly is treating a subsection of a subsection like it was holy writ. And dismissiveness is usually the best option when dealing with those less experienced than you. I uploaded my first image maybe a month ago. I haven't uploaded a new image in a couple of weeks at least, after getting bludgeoned by BCBot, both on images I'd uploaded, and on images at articles I contributed regularly to. I'm tired of it, and I'm not going to be uploading (or working with images at all) until this bot is either reined in or shut down. Bellwether ^B_C 06:16, 14 February 2008 (UTC)
• The issue is also non-uploaders wanting to fix things, faced with ridiculous conditions to do so, the issue of minimal exposure images being lost due to an arbitrary 7 day deadline, and images that were uploaded under the now out of date correct policy, being tagged multiple times every time a phrase changes. The issue is also the us and them attitude, there is absolutly no link from the bot page for a collaberative effort for experienced editors to fix things in a coordinated manner, nor any links to major consensus regarding the bot, it's all hidden all over the shop, all that exists is a long list of excuses and get out clauses. Most loaders hit by the bot only know to go to the bot page if that, and realy struggle to even comprehend what is required of them. —Preceding unsigned comment added by MickMacNee (talk • contribs) 06:21, 14 February 2008 (UTC)

What would be a better time frame than seven days? "Arbitrary" doesn't seem appropriate considering these are images that infringe upon copyrights. Lara❤Love 06:33, 14 February 2008 (UTC)

Please don't confuse Wikipedia policy with U.S. law. The former is far more restrictive than it legally has to be. Most of these images do not infringe upon copyrights; they simply haven't had a specific rationale for fair use written yet. An appropriate time frame would be one that gives actual humans (not bots) time to give a thumbs-up or thumbs-down to each individual image, and to write a rationale if the image is to be kept. *** Crotalus *** 06:39, 14 February 2008 (UTC)

They are not infringing until examined and challenged in court. The one I corrected today was patently legally correct, it just did not meet whatever matching criteria the bot uses (these are I believe kept secret, why?). Many of these are not infringing at all, it is purely the bot design that tags them. Giving 7 days notice on an article loaded in 2005 along with 15,000 others at the same time is patently ridiculous. It is also my understanding that an image with a few keywords but filled with gibberish does not get tagged. MickMacNee (talk) 06:46, 14 February 2008 (UTC)

I know the quesiton was rhetorical but it would make more sense to tag all the images at once rather than haphazardly, and give people a very short time to correct new uploads but a long time (say, March 1 or March 15) to correct older ones. That way people could plan their work load, and rest assured that once they addressed all their notices they would be done with it. That's not going to happen, but just my $0.02. Indeed, the vast majority of these images don't violate copyright and are perfectly fine for Wikipedia, they just lack some data fields on the image file. Wikidemo (talk) 11:35, 14 February 2008 (UTC)

The problem here is that people who either don't understand the Wikimedia Foundation policies or don't understand how to write a proper fair use rationale are having their images deleted. It seems to me like they want someone to blame. Either Betacommand, his bot or the deleting admins are taking flak for enforcing the policies laid down by the Foundation. The policy is not decided by Betacommand, his bot or the deleting admin. Also it is not always simple to write a FUR if you are not the uploader. How can I know the source of an album cover that someone probably google searched? Also with regard to Bellwether's comment "No, what is silly is treating subsection of a subsection like it was holy writ." - If any part of this section is not met, the image may be deleted. It's written in the "enforcement" section right below the one I linked to. You may think that 10c is trivial but the Wikimedia Foundation (ie the owners of the servers and they set rules which we cannot override) do not. There's nothing silly about a bot pointing out violations of policy. James086^{Talk | Email} 07:22, 14 February 2008 (UTC)

There are issues on how it is being scheduled. There are people who want to fix things, this bot is not helping by going the extra yard and scaring new users and annoying old users that have complied with a now changed policy. At the very least this should have been an internal project bot to highlight first issues for an experienced group to review/quick fix, without going straight to tagging the uploader, placing a massive incomprehensible tag on the image, and deleting within 7 days (that isn't being enforced because it is not working, so why say it?) —Preceding unsigned comment added by MickMacNee (talk • contribs) 07:27, 14 February 2008 (UTC)

Plus nobody it seems is aware how the bot checks to decide when and when not to flag, it's whole design and operation appears to be in one persons hands. That to me is wrong when it affects so many other editors. MickMacNee (talk) 07:30, 14 February 2008 (UTC)

There is no-one more suited to writing a fair use rationale than the uploader. If they don't know how, then they shouldn't have uploaded the image. New/inexperienced users may not have known about fair-use and the policy, but the message given by BetacommandBot links to Wikipedia:Non-free use rationale guideline and Wikipedia:Media copyright questions. I think tags such as {{no rationale}} are pretty clear, but if you can think of a better way of phrasing the message that needs to be conveyed by the tag, go for it. Finally; even if the images aren't being deleted within 7 days it provides an incentive to provide a valid rationale now instead of procrastinating. We can't change this policy even if it is overwhelming for new users (I stayed away from fair-use for a long time). James086^{Talk | Email} 07:56, 14 February 2008 (UTC)

The images I'm often seeing tagged are ones where the uploader is usually long gone due to issues like this bot. And I don't get this idea that you can jump straight in and add text without any knowledge, and it will usually be fixed if wrong, yet woah betide you if you add an image with even the most trivial of non-compliances. Only a small subset of images really need the uploader's actual knowledge. Understanding the rules around nfcc is a nightmare. The bot is damaging efforts for new uploaders and experienced fixers at the same time. To suggest all is hunky dory at the moment is daft. The lack of group effort or consensus regarding this bot is also amazing, as fixing furs at the rate they are being tagged is way out of the league of even the most wiki-addicted. MickMacNee (talk) 08:08, 14 February 2008 (UTC)

Replying to MickMacNee: As far as I can tell, the bot has a very simple algorithm. It takes the name of the pages an image is being used on and looks for them in on the image page (note that the links at the bottom of an image page are automagically generated, and so don't count). If any of the images are precisely named on the page (doesn't even have to be linked), ie. including things like "(diambiguation bit)", then the bot passes it. There is even a toolserver link around somewhere to a tool where you can check your images to see if Betacommandbot will tag them or not. So MickMacNee is correct to say that you can write gibberish on a page and the bot won't be able to tell as long as you include the name of the article somewhere. Of course, if a human spots this, the image will be tagged or corrected, and the editor who did this would get warned and, if they didn't stop, blocked. There are images I know of, which lack rationales, but which have the article name on them for other reasons. The most common reason is when the description bit says "picture of random article name" (with or without the link). The bot won't be able to detect these. It does, however, detect ones that say "picture of subject of article" without naming the article. Carcharoth (talk) 08:01, 14 February 2008 (UTC)

If it's Just That Easy™, why does the bot fix the fair use rationale? What information is it missing? For that matter, if it only detects a linked article name if the article has been wikified, that's a bug. A fair use rationale has to name the article it's used in, but does not explicitly have to link to it.

Another bug is that the bot is too stupid to tell when a targeted page has been moved and replaced by a disambiguation page, which triggers a false NFC. If it was a human eidtor who falsely harassed a few dozen users and deleted images under false pretenses, they'd be banned for vandalism by the fifth instance. We should not be more lenient for bots than we are for people.

Another issue is how totally confrontational everything about this bot is. Look at User_talk:BetacommandBot: you're greeted with screen-tall stop sign and 17 rules that all essentially say "piss off I'm right about everything." Then read the discussions below where it gets even worse. (Seriously, read it.) —Torc. ^(Talk.) 10:39, 14 February 2008 (UTC)

Form issues

The issue is obviously that most uploaders are ignoring the page that appears prior to the upload formula, if the users uploading Fair Use images where presented a blank form and filling it was obligatory to complete the upload we wouldn't be dealing with a backlog of 11,000 pages due to something as simple as FCC#10c. - Caribbean~H.Q. 06:38, 14 February 2008 (UTC)

• No, the issue is that WP's policies regarding image useage are so arcane and stringent that they allow bots to tag perfectly acceptable images for deletion, based on a subsection of a subsection. When this project brings their image use policies more in line with actual copyright laws, perhaps the deletionists won't have such a field day, and new uploaders (and image workers) won't become discouraged and give up. Bellwether ^B_C 06:45, 14 February 2008 (UTC)

You've stated this four times now. What we're discussing now is a possible way to prevent images from being improperly uploaded to begin with. Lara❤Love 06:52, 14 February 2008 (UTC)

• And you've ignored the root cause I raise four times now. The problem isn't good faith uploaders who don't know the subsection of a subsection of the NFCC that BCBot uses to tag images for deletion. The problem is primarily with the policy that allows the "letter of the law" enforcement that BC demands from initial uploaders, and the tagging itself, which seems like killing a gerbil with a bazooka to me. Bellwether ^B_C 13:53, 14 February 2008 (UTC)

That's a good point. A change that I would think could be easily implemented. And what would be the negatives? Lara❤Love 06:42, 14 February 2008 (UTC)

There are many images that predate the bot specs and were uploaded perfectly correctly. MickMacNee (talk) 06:46, 14 February 2008 (UTC)

You really have to separate this issue into two parts, old images and newly uploaded ones. Or maybe three - brand new, a few months old, and really old. Older images are a real problem because people get blindsided by the tags and by deleted images. For people actively uploading new images they can get spanked around a little bit until they learn how. "Arcane" is an overstatement, it's just a weak user interface. People grumble but I think that's more in the delivery of the message than the task they have to learn. In less than 1/2 hour you could learn most everything there is to know about how to upload images properly. It's a lot faster than learning the markup language, or learning where all the policy pages are. Nevertheless, anything that could make the process easier to get right from the start is good. Wikidemo (talk) 11:42, 14 February 2008 (UTC)

• The negatives are that oftentimes an image is used for articles other than the one it was specifically uploaded for, which then triggers the bot to tag it per 10C. Bellwether ^B_C 06:47, 14 February 2008 (UTC)

I've pushed for this in bugzilla:12452. No word yet on its implementation. If someone could write the css/javascript code, we might be able to implement it locally. MBisanz ^talk 06:50, 14 February 2008 (UTC)

I am suggesting a option to prevent a simmilar situation in the future, this is supposed to kill the problem at its root. About sideffects, the only one that I can perceive is that the upload process would take longer, but then again when that is compared to the time that admins spend cleaning image backlogs that effect proves meaningless. Please note that this blank form (wich should be designed so even the most dumb of bots can understand the resulting rationale) would only appear if the user selects one of the Fair Use licences, wich means that users uploading free or public images wouldn't encounter this problem. - Caribbean~H.Q. 06:50, 14 February 2008 (UTC)

Most first time uploaders probably don't even know the difference between free and non free, but the editors that do are being given the runaround by the reverence being held for this bot, and ignoring the massive issues it creates and continues to create with it's tidal wave operation. The culture is also now I believe, we have a bot that fixes that, so let's not do it ourselves, or guide good faith violators. The tag box is huge and intimidating to a new user. MickMacNee (talk) 06:53, 14 February 2008 (UTC)

Again, if a new user is going to upload a image he/she will receive the instruction page before uploading, now if they choose to ignore that, chances are that they will most likely end uploading without a rationale wich will only add to the backlog, we can't have a competition between a bot that fixes images and several tagging them, and we can't let the 'fix bot' choose if he removes a notice placed by another bot before. - Caribbean~H.Q. 06:58, 14 February 2008 (UTC)

Actually, I would support a redoing of the image uploading procedure where incomplete uploads were prevented from happening. If a user does NOT supply a required piece of information, the page should send return an error message and stop the upload until all required info has been provided, INCLUDING the article where the pic is intended to go. This would reduce the problems on the back end that leads to all the grief this bot takes. I have no idea how this can be techincally done, but it seems a good idea. On the flipside, regardless of how inconvenient it is, the bot is doing necessary work, since images which are protected by copyright should NOT be used inappropriately on wikipedia. Remember, even "fair-use" is a violation of copyright, though being done in an "excused" manner. If you wish to excuse the violation of the rights of someone else, you should probably be very clear as to how and why and for what reason you are doing so. The bot only cleans up situations where people have not done this adequately. If it tags too many images, its probably because most people are doing it wrong, which is why we need to fix it at the "front end" before being uploaded. --Jayron32.talk.contribs 07:01, 14 February 2008 (UTC)

Let's not get hung up on first loaders, as I've said, there are wider issue with those about them not having a clue at all. But I am seeing a very large proportion of these flags being like this: [2]. Perfectly fixable in seconds by experienced editors, yet we are being swamped, and who is coordinating these efforts, and from where? And is there any information at the page causing all these tags, the bot? No. Leaving fixes to uploaders and pontificating as such by the bot is also wrong when like this, it was loaded 18 months ago. MickMacNee (talk) 07:16, 14 February 2008 (UTC)

WT:TODAY should give some insight into coordination. But its rather dormant now . MBisanz ^talk 07:33, 14 February 2008 (UTC)

For those images uploaded a year and a half ago we will have to deal it with patience and using the current methods. Please note that my proposal is focused in the images being uploaded in the very close future so we don't have a similar discussion eighteen months from now. - Caribbean~H.Q. 07:34, 14 February 2008 (UTC)

I was just about to make the same point - it's not fair to say uploaders should've enforced a policy that didn't exist at the time. At the same time, though, people should keep a check on their uploads. Will ^(talk) 13:03, 14 February 2008 (UTC)

Disputed fair-use image deletion deadlines

Partial repost of previous post at WP:AN

Please click "show" above and have a look at the backlogs arriving soon in the "disputed fair use images" dated categories. Normally, these images would be deleted after 7 days. The practice so far has been to extend the deadline some indeterminate amount. Given that these runs by Betacommandbot were done rather close together (looks like an extended run over three days), what would be the best way to determine a suitable extension here? An extra week? An extra two weeks? I asked Betacommand on his talk page recently (a few weeks ago), and he said he was near the end of doing these runs. Previously, I put dates of 10th and 17th February on the other backlogged categories. I'd suggest putting a date of 2nd March on these categories. It isn't terribly clear where this sort of thing should be discussed, or with whom, and it seems no-one else is attempting to manage the backlogs. To avoid future disputes, I'd appreciate it if people could object to or endorse this action, here. Thanks. Carcharoth (talk) 07:08, 14 February 2008 (UTC)

Well I only do logos (and some symbols and seals). Right now AWeenieMan's tools say I have ~3900 images to process. Given real life commitments and what, I would need about 3 weeks from today to get through them all. MBisanz ^talk 07:42, 14 February 2008 (UTC)

I'd recommend giving them a month at least. Give the original uploaders (who might not be very active) some fair time to see the notices. I know the policies are strict and all but 18,000 images obliterated in a week just seems brutal (though kind of impressive). • Anakin ^(talk) 18:23, 14 February 2008 (UTC)

The above discussion is preserved as an archive. Please do not modify it. Subsequent comments should be made in a new section. —Preceding unsigned comment added by Nakon (talk • contribs) 07:13, 15 February 2008 (UTC)

Solution (for the future)

Stop uploading so many non-free images, and use fewer non-free images. Less time spent writing rationales, less time spent fixing images, Wikipedia is more free — its a win-win situation. Mr.Z-man 18:31, 15 February 2008 (UTC)

Agreed. It is much more fun looking for free pictures, anyway! One thing I would say, is please, please, take care over old black-and-white photos. Some will be public domain though we are not aware of it, and some will be genuine historical photos and of great educational and encyclopedic value (and hence exactly what fair use is intended for). For now, to deal with the backlog, I've put a deadline notice of 2nd March (around two weeks) on those categories. I would also like to see Betacommand do a schedule for the runs he intends to do in March. Tagging thousands of images a few days before the Foundation's deadline (23 March 2007, or seven days later depending on your interpretation), would not be appreciated, I fear. Carcharoth (talk) 09:09, 16 February 2008 (UTC)

Given that we have about 500 non-free uploads per day, Im guessing that my future should be around 1000 images about twice a week. β^_command 16:14, 16 February 2008 (UTC)

Next BCBot Phase

What BCBot has been doing is tagging images without any valid rationales. The issue with this method is that if a image is used in multiple articles, but only one valid rationale, it violates our NFC policy, but the images should not be deleted. My next planned phase was to remove the image usage that does not have a rationale. Comments or suggestions? β^_command 16:58, 16 February 2008 (UTC)

What BCBot has been doing is tagging images without any valid rationales - This is fundementally an incorrect statement, your bot does not tag images that have complete garbage as a rationale but mentions each usage once. Your bot should not be the first and only assesment of the presence of a valid rationale before the placement of huge tags on images and talk pages with phrases such as invalid rationale (is it? on what basis did this get decided?), The rationale is (not) presented in clear, plain language, Unless concern is addressed by adding an appropriate non-free use rationale (which may already exist), or in some other way, the image will be deleted . MickMacNee (talk) 17:10, 16 February 2008 (UTC)

It seems a bit premature to talk about changes to BCBot when there is clearly no consensus on its continued operation AT ALL. My feeling is that until BCBot can accomplish its work without angering editors, it shouldn't be running AT ALL. If you don't choose to stop BCBot, then you should be forced to stop running it. The root problem with BCBot is that it is enforcing a policy that demands more work from volunteers. This must be done gently, if at all. And it's clear from reading the hostile messages here that BCBot is not gentle.

May I make a suggestion for a resolution of this issue? You and Hammersoft have claimed that you are operating with the support of many admins and editors. If you feel that this work is so important, then you should be willing to do it yourselves. If you cannot find volunteers to work on this project, then attempting to coerce editors (FIX YOUR IMAGE RATIONALE OR ELSE THE IMAGE GETS IT) does not improve Wikipedia. Better to let it die under the weight of copyright lawsuits than to create so much unhappiness. RussNelson (talk) 04:04, 17 February 2008 (UTC)

• I don't have much sympathy for people unwilling to comply with our policies, which have been in place for a very long time. Nor do I have much sympthy for people canvassing to get support to ban the bot [3]. And by the way, I'm not operating this bot. I support what it's doing. Betacommand operates it. And no, I'm not going to perform the work because it involves thousands upon thousands of images. No small number of people can handle it. I recently estimated that if it took 10 seconds per fix it would take 11 straight days of editing to fix just a portion of the images tagged by Betacommand. The problem is absolutely massive. The bot is a tool to get it under control, to change the culture of liberally uploading fair use under whateverthehellsomeonethinksistherightwaytodoit. It's GOT to change, or we might as well forget about ever getting into compliance with Foundation dictums on this. I personally don't care if the people liberally abusing this policy are unhappy about it. They should be *glad* to work towards compliance with policy, not fighting it like it's the second coming of Satan and the end of the project if their precious non-free image gets deleted. --Hammersoft (talk) 04:29, 17 February 2008 (UTC)

People aren't generally "liberally abusing" the policy. Many of them don't understand it, as it hinges on a highly technical definition of "rationale" that is not obviously different from copyright tags. Many of them uploaded images completely correctly, before the 10c-rationale policy existed. If you want all of Wikipedia to follow a new policy, you have to make the policy easy to follow and not demonize people who don't instantly change what they're doing to go along with it. rspeer / ɹəədsɹ 10:03, 17 February 2008 (UTC)

(ec)So BC, if I understand thise new phase, BCB will go through articles that use an image, but that the image has no FUR for. What sort of FUR will it look for? A valid backlink? the article title? a FUR template? etc. I'd say I'd support BCB generating lists of images that are used in articles where there is no mention of the article title on the image page. My fear is that if BCB edits the articles to remove the images, people might not notice and know to write a separate FUR. How many images are we talking about (FUs in more than 1 article)? My rough guess is 5%-10%, but I'd like a firmer number (xx,xxx) for example to figure out the impact of this phase? BTW, how many images are left in the current phase? MBisanz ^talk 04:35, 17 February 2008 (UTC)

I agree entirely with MBisanz. Betacommand needs to give the community an idea of numbers before starting new phases. He should also have the courtesy to state when a previous phase has finished, before rushing on to the next stage. I support a lot of the work that Betacommand does, I just wish he would be just a little bit more organised about it! I'm going to start a new section below, and maybe even a whole new page, like Wikipedia:Non-free content criteria compliance. We need to do this systematically as a community, not on the whim of Betacommand (though he has done much sterling work and should be thanked profusely). Carcharoth (talk) 13:54, 17 February 2008 (UTC)

BC, I don't remember the details of this phase, but could you please be sure to leave behind a redlink or some other notice, so that any editor who notices it can restore the image (if appropriate) by adding a rationale and a link? It would be best to leave something in the article itself, not just the talk page, because that way people who visit the article but don't necessarily have it on their watch list will be able to notice and pitch in. Thanks, Wikidemo (talk) 14:37, 18 February 2008 (UTC)

I think the original details of the four-phase BCBot programme is here. That may be out-of-date now, though. Carcharoth (talk) 16:02, 19 February 2008 (UTC)

Carcharoth i am basicly planning to start Wikipedia:Administrators' noticeboard/Archive91#Phase 4 in a month or so, (earily april). yes I know is after the deadline, but implimenting it sooner whould not be feasable. what I set out there is close to what Im going, Ive updated it for current issues and requests. β^_command 17:02, 19 February 2008 (UTC)

So much horror over a bot, and no horror over the policy violations

I find it amazing that certain vocal people here are absolutely desperate to do SOMEthing to stop this bot, to get it banned permanently. They stand in horror that 11,000 images would be tagged in a single day. You guys want to take a guess how many fair use images are improperly uploaded/licenses/tagged/rationaled every day? If we didn't have a bot to help manage this problem, we might as well give up on ever having the fair use situation brought under control on this project. All of you horrified at the work this bot is doing, which is completely in compliance with policy, should be absolutely MORTIFIED at the rampant abuse of policy performed by ungodly numbers of users who simply just don't care about our policy. Or maybe that's the point? You don't care about our policies? --Hammersoft (talk) 04:26, 17 February 2008 (UTC)

Sometimes the tag itself adequately sums up the rationale, and no additional words are needed (e.g. {{screenshot}}. Ron Duvall (talk) 06:30, 17 February 2008 (UTC)

But the tag doesn't say which article it should be used in (Example), or the source of the image (www.), or a description of what the image is (Logo of IBM from 1971). MBisanz ^talk 06:46, 17 February 2008 (UTC)

Sometimes it's obvious, as when the image itself clearly shows the screenshot, and "what links here" takes you to article on the software. And the source is irrelevant with screenshots; someone could have hit Printscreen and then pasted it into Paint. Ron Duvall (talk) 07:44, 17 February 2008 (UTC)

But, it could also be a screenshot from a news article, so we'd need the user to say either "I took it" or "I got it from NYT.com/132543". And the "What link here" section only shows articles containing the image, it doesn't define if there is a rationale under copyright law for the use of the image in that article. MBisanz ^talk 07:56, 17 February 2008 (UTC)

All of you ... should be absolutely MORTIFIED at the rampant abuse of policy performed by ungodly numbers of users who simply just don't care about our policy.

Why should anyone be mortified, when such behavior is completely and totally predictable. You set up a project in which every Tom, Dick and Mary who shows up at the door can jump right in and participate, and then you wonder why they don't care about your policies? Start with: Why should they care, since the system of enforcement is weak at best and random for the most part, and the chances of their being caught are minimal (unless, of course, they're just stupid or don't care if they're caught)? Then go to: there are so many damned policies that even if someone wanted to play it straight, it's practically impossible to do so -- and the interpretation of these myriad policies differs from administrator to administrator, from case to case, from circumstance to circumstance.

This is the system you've set up, which inevitably lead to the rampant disregard of your upload policies - so why be "mortified", just fix it.

Unfortunately, the only way you're going to be able to do that is to in some way change the essence of the project. It's going to have to be more tightly administered, but the only way to do that is to reduce the number of policies down to a manageable lot that both editors and administrators can understand without taking a lifetime course in Wikipediology.

The structure and design of the project set yourself up for this, and, ultimately, the only way to make it better is to change the structure and the design. Ed Fitzgerald (unfutz) ^{(talk / cont)} 08:34, 17 February 2008 (UTC)

You could apply that rant to pretty much any set of rules. Why bother with speed limits? Why bother with copy protection on music? There really aren't that many rules on Wikipedia, and the ones that are the most important are the ones based on common sense: keep a neutral point of view, don't be a dick, and don't put copyrighted material on a free encyclopedia, just to name a few. --clpo13_(talk) 08:53, 17 February 2008 (UTC)

Sorry to disagree, my friend, but that's not the case at all. Speed limits are just there, they're regulations created for a reason, but they can be there (as in most of the U.S.) or not (as on the Autobahn in Germany, and for a while in Monatana), they can be set high or low, they can change or they can stay the same for decades -- nothing about our system of justice or social regulation determines or requires that there must be speed limits. That's not the case here, where the problems are essentially systemically determined by the structure and philosophy of the project.

Oh, and not a lot of rules on Wikipedia? Are you kidding? We've got "guidelines" out the wazoo that most people treat like Holy Writ, and they specify, sometimes down to the comma or dash, what you can and can't do. Worse than that, the rules overlap, they conflict, and they are subjected to myriad interpretations, so if someone doesn't like what you're doing (regardless of its potential value) there's almost always a rule that can be cited to use to revert it. From this lack of clarity and too much complexity comes, inevitably, corruption. Ed Fitzgerald (unfutz) ^{(talk / cont)} 09:07, 17 February 2008 (UTC)

No need to apologize for disagreeing. You have your Wikipedia philosophy and I have mine. Of course, considering how you think things should be run, I'm beginning to wonder why you even bother sticking around Wikipedia. If Wikipedia is as deeply flawed as you think it is, and if you refuse to go through the usual channels to suggest a fix, then it's not going to change simply because you want it to. Perhaps you'd be better suited at Scholarpedia or Citizendium. I was actually visiting a fork of Wikipedia with stricter editing policies just recently, but I can't remember what it was called. Ah well. At any rate, if you oppose the open policy of Wikipedia, it's clearly not the place for you. --clpo13_(talk) 09:25, 17 February 2008 (UTC)

No, Wikipedia is the place to be, since it has become the current de facto standard, the first place people look for information on the Web. I think the project has fabulous potential, but I worry that its inherent contradictions will do it in before it has a chance to reach it. The infoworld moves fast and its past is full of de facto standards that fell by the wayside. My hope is that Wikipedia will resolve its systemic problems and become (in a phrase that's vaguely familiar) "all that it can be", so I'll stick around and do what I can to help out.

But also I think you may misunderstand me -- I'm not calling for stricter editing policies of things as they stand now, I'm calling for stricter and more coherent enforcement of a vastly simplified and more open set of policies. First the policies need to be whittled away to the essentials, and then they need to be properly enforced.

But this conversation is getting to be a little too esoteric for this venue. I simply wished to respond to the editor who seemed to think that we all should be beating our breasts about people uploading pictures they shouldn't, when, in fact, that behavior was perfectly predictable from the start! Ed Fitzgerald (unfutz) ^{(talk / cont)} 09:38, 17 February 2008 (UTC)

BTW, not just on Wikipedia, but everywhere all over the world, "usual channels" are the places where people who live to make sure nothin' ever changes no-how hang out. "Going through channels" is a perfectly acceptable thing to do in a bureaucracy or a strict hierarchy, if what you want is to order some new paperclips or get a soldier discharged, but channels aren't much good at dealing with change, since they exist to make sure the same kinds of things happen in the same kinds of way over and over again. Besides, (maybe you've noticed?) I'm not much of a fan of that kind of thing. Ed Fitzgerald (unfutz) ^{(talk / cont)} 09:44, 17 February 2008 (UTC)

Individuals come here and upload items and articles that for various, sometimes complex reasons, shouldn't be in an encyclopedia. In order to maintain openness and get as much good "stuff" as we can, we have admins, and bots, and experienced users, who use tools to weed out and sort that information. So should users be Mortified, probably not. Should we abandon policies simply to get more "stuff", probably not. Even though some things could be made better (and many of us are trying our darndest despite the lack of a paycheck), there is no deadline MBisanz ^talk 09:53, 17 February 2008 (UTC)

Paycheck doesn't enter into it. Your reward is that you get to delete stuff, and for your kind, that's priceless. Baseball Bugs ^{What's up, Doc?} 19:24, 17 February 2008 (UTC)

Hey! MBisanz is one of the "good guys". There probably are people who delete because they think it needs doing and don't think too much about anything else, but MBisanz isn't one of them. Carcharoth (talk) 21:22, 17 February 2008 (UTC)

Hammersoft, the problem is not WHAT BCB is doing. The problem is HOW it is doing it. Yes, I realize that some people don't want images tagged. We can disagree with them while at the same time admitting that BCB is harming Wikipedia. I think that you are so afraid of giving these people any quarter that you are unwilling to acknowledge any flaws in BCB. For example, BetacommandBot only knows about the case of English letters, so it wouldn't be able to see that Βcommand and βcommand could be different names for the same article! Why not be honest and say "Yes, BCB has problems, and harms Wikipedia, but the benefit it does is greater than the harm, so I can accept that." RussNelson (talk) 03:24, 18 February 2008 (UTC)

I'll say "Yes, BCB has problems", but I don't think those problems harm Wikipedia. I remember an instnace some months ago with an article that ended in 4 "!"s and therefore couldn't be recognized in templates. Now, that is a problem, but I don't think its a problem that harms the encyclopedia, because users recognize it and respond accordingly. Same thing with a certain song title that begins "Help:" thereby placing it in the helpspace instead of as an article. Another problem that doesn't harm the encyclopedia. Now if BCB was deleting images or not notifying creators or removing images at random from articles, that would be harming the encyclopedia. MBisanz ^talk 04:14, 18 February 2008 (UTC)

Betacommand has more power than an admin, because of closed source

Just by running a bot that's allowed to make so many edits so quickly, Betacommand has more power than admins to set policy. In particular, he uses this bot to enforce a very particular interpretation of WP:FURG 10c that is not required by the Foundation or by copyright law. I don't like the way 10c has been handled in general, especially the paranoid rush to beat some imaginary deadline, but even the kind of enforcement we do now -- by which I mean the kind of enforcement BetacommandBot does now -- could be made much more intelligent.

One very noticeable aspect of this: BetacommandBot approves of a rationale if and only if it repeats the name of the article exactly. Being a robot, BetacommandBot of course has no common sense, and there are many rationales out there that anyone with common sense can tell are correct but don't fit this narrow criterion. But when this discrepancy leads to disputes, Betacommand and Hammersoft (in particular) define the rationales that aren't understood by the bot to be against policy.

In effect, the bot is a de facto policy, but this policy cannot be discussed or edited by the rest of Wikipedia because Betacommand will not share the source code. When the bot does the wrong thing, sometimes the only way to get Betacommand to fix things is to block the bot. However, the bot holds other things such as the RfC process for ransom. If you block the bot, new RfCs can't be created. So BetacommandBot wins policy disputes with admins automatically -- it holds the power -- despite the fact that neither Betacommand nor his bot would realistically pass an RfA.

I believe we need to decentralize the tasks performed by BetacommandBot, and most importantly, open up the source code. Wikipedia runs on open source. I cannot see why Betacommand will not cooperate with Wikipedia and open up the source code to his bot -- "security through obscurity" is rather unhelpful here. Opening up the source code, in addition to being just a good idea for preventing bugs, also allows people to suggest patches and different ways to do things. Imagine if people who cared could fix these issues instead of just repeatedly complaining about them:

• BetacommandBot could tolerate near misses on article names instead of demanding an exact match. A rationale correctly identifying an album cover as being used to illustrate "the article on the album '!'" is a perfectly good rationale for using the image on ! (album).
• BetacommandBot has poor scheduling of the order in which it tags images and leaves notices, which causes it to disruptively spam users' talk pages.
• BetacommandBot leaves messages with misspelled words in them, making Wikipedia policy look amateurish.
• BetacommandBot edit wars, blithely retagging images when people contest the tag. (A possible alternative is to create a noticeboard for disputed tags, much like the PROD->AfD process. There are probably other ways to deal with this. They would require discussion, but right now discussion is hardly possible.)

My point here, remember, is not to simply point out these failings in the bot, but to point out that these failings could be fixed if Wikipedians could read the code and suggest patches or get their own version approved.

rspeer / ɹəədsɹ 10:26, 17 February 2008 (UTC)

Then people will just be warring over the code. Our best work is being done by bots who closed source their code. User:Zscout370 ^{(Return Fire)} 10:27, 17 February 2008 (UTC)

User:Zscout370 could you list some of these bots, and perhaps some that aren't closed source? Thanks. Mccready (talk) 10:48, 17 February 2008 (UTC)

There is such a noticeboard Wikipedia:Image copyright help desk for Beta related comments. Beta doesn't get paid to do what he does, so I really don't like the idea of trying to hold him to a schedule. 20,000 images is a lot at once, but its certainly not unmanageable with tools lik FURME. Yes, 10c is a rigid interpretation, but this process was approved by the Bot Approvals Group, which is sanctioned by the Bureaucrats, and the bot does enforce a policy that has consensus. For the RFC process, there is a backup bot, and I know Beta has been working for sometime on trying to split up the functions the bot performs. MBisanz ^talk 10:34, 17 February 2008 (UTC)

It takes approximately no time to release the source code to a program. It can take effort to make the code well-documented, or runnable on someone else's system, but I'm not actually asking him to do that. On the other hand, Betacommand is holding us to a schedule by running the bot so quickly that no one can keep up with fixing the rationales. rspeer / ɹəədsɹ 10:41, 17 February 2008 (UTC)

I fail to see how that's an argument for closed source, Zscout. You realize that there's a difference between the copy of the code that's running on a server and the copy of the code that's put up for public review? People can't just edit the displayed code and make the bot do something different. However, people could suggest that Betacommand incorporate changes in the bot he runs (this doesn't require a "war"), or split off their own version of the bot and get it approved through the BAG (and I hope that Betacommand would recuse himself from participating in the approval decision). rspeer / ɹəədsɹ 10:38, 17 February 2008 (UTC)

People are giving suggestions, here and on his talk page. Hell, there is a discussion about this bot on a weekly basis. So he is taking their suggestions to mind. But no matter what anyone says, the only person who can decide to put the code out there is BC. User:Zscout370 ^{(Return Fire)} 10:42, 17 February 2008 (UTC)

Yes, that is true. What I am saying is that the discussion could be more useful and less adversarial if the source code were open, because people could suggest actual patches to the bot instead of saying "BC, you need to figure out how to code this". How is your response an argument for closed source? rspeer / ɹəədsɹ 19:11, 17 February 2008 (UTC)

Comment. "10c" is part of the fair use policy (Wikipedia:Non-free content criteria). The resolution indicates that appropriate rationales are required for material used under an EDP. "10c" was part of the policy when the Foundation policy was passed, specifically mentioning our fair use rules as an example of an appropriate EDP.[4] The bot has no special control over the resolution and policy, but rather acts in response to policy in the fashion it was approved for. The Foundation resolution also indicates that non-free usage should be machine-readable, which could be construed to include the accurate linking/naming of the articles in which a non-free image is used. Just some thoughts. Vassyana (talk) 11:59, 19 February 2008 (UTC)

I know that our EDP requires we enforce 10c. It does not require that we enforce it like BCBot does. The "machine-readable" clause has also been construed as one we've satisfied for years, because all non-free use template names start with the text "non-free". If you're more paranoid than that, you want each rationale to be machine-readable as well. But even that is still different from "readable by BCBot", which is what is currently enforced and not required by the Foundation. rspeer / ɹəədsɹ 23:11, 19 February 2008 (UTC)

Numbers, numbers, numbers

Can we please have a central place to record typical numbers surrounding image issues? I suggest Wikipedia:Non-free content criteria compliance for now. Some typical questions:

• How many images does Wikipedia have (give an idea of numbers over time)?
• How many free and non-free images does Wikipedia have? (And under what circumstances are free images kept and not transferred to Commons?)
• How many free and non-free images date from (were uploaded in) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008?
• How does Wikipedia identifiy its free and non-free images? (License tags - are these systematically kept under control?)
• How many images are uploaded each day? (Split the number between free and non-free)
• How many images are uploaded each month and how many are deleted? (deletion needs to be averaged over a month - varies too much daily)
• How many non-free images have non-free use rationales?
• How many non-free images lack non-free use rationales?

After we have clear answers to these basic questions, we may be able to set up a system that works more smoothly and has more vision than the current one - or at least a more articulate vision that the current one. Carcharoth (talk) 14:08, 17 February 2008 (UTC)

Some stats at:

• Wikipedia talk:Non-free content/Archive 25#Updated stats
• Wikipedia talk:Non-free content/Archive 29#Towards keeping track of image-licensing template usage
  • Example of stats

But more are needed. Carcharoth (talk) 14:56, 17 February 2008 (UTC)

I'd say you'd need to contact a developer at MediaWiki to get most of that data. I just don't think its available in a format that even a script could extract from the encyclopedia. MBisanz ^talk 15:06, 17 February 2008 (UTC)

Complaints about BCB are often rebutted with replies claiming the size of the task. It would be nice to have a centralized source, as proposed in Wikipedia:Bots/BetaCommandBot and NFCC 10 c where I am trying to centralize a consensus for specific NFCC10c changes. MickMacNee (talk) 17:21, 17 February 2008 (UTC)

wikipedia currently has 773,658 media files Normaly it's about 300K unfree images. Tags for unfree images are pretty darn systematic. less so for free.Geni 19:17, 17 February 2008 (UTC)

Cool! Thanks. :-) Any tips on how to answer the other questions? By the way, where does the 773,658 number come from? Is that some magic word or something? Carcharoth (talk) 19:30, 17 February 2008 (UTC)

The number of media files is on Special:Statistics.

Yeah it's a magic word (or variable) {{NUMBEROFFILES}}, see Help:Magic words#Statistics. --Sherool (talk) 12:45, 19 February 2008 (UTC)

Most of these questions can be done easily with the database (toolserver) except that I can't differentiate between free and nonfree images that way. I do have a database query that will gather stats on taggings, deletions, uploads of images. Perhaps I can run it over the last year later today. --uǝʌǝsʎʇɹnoɟʇs(st47) 19:34, 17 February 2008 (UTC)

Generaly the best attack line for answering the questions is to take a smaller sample so for the simple free unfree spilt you pull up 100 random images and count.Geni 19:52, 17 February 2008 (UTC)

And by that method I estimate we currently have a little over 200K taged unfree images and a worrying number of copyvios.Geni 20:00, 17 February 2008 (UTC)

So my prediction that the obsession with the NFCC would lead to people failing to check for copyvios has come true? Well, that's good to know. Now, all sarcasm aside, what is the best way to deal with this? Should there not be as much focus on checking sources and weeding out invalid claims of free images, as there is on rationales for claims of non-free use? Geni, could you definre what you see as the difference between a non-free image and a copyvio? I think I know the difference you mean, but many people think that non-free image claims lacking a rationale are copyvios, when in fact they are not. I see a copyvio as something like an image with no fair use claim at all (not even a copyright tag), or a copyrighted image incorrectly tagged as free or public domain. Do I have that about right? Carcharoth (talk) 21:10, 17 February 2008 (UTC)

Nah the image namespace has been lousy with copyvios for years we could knock out a fair chuck of them by going orphan zapping. Fixing that is the next thing to do. In this case by copyvio I mean an image with say a GFDL tag that I don't belive. Non free image means something with a fair use tag of some sort.Geni 23:27, 17 February 2008 (UTC)

Here's my stuff. The 'date' is by week. This means that in the first full week of 2007, we had 16806 uploads, 6582 deletions, for a net gain of 10224. and none of the bots made any taggings. In the first full week of 08, we had a net loss of 841 images (free and non-free included here, remember) with bots making about 6000 taggings. --uǝʌǝsʎʇɹnoɟʇs(st47) 19:57, 17 February 2008 (UTC)

+--------+---------+---------+------------+--------+----------------+-----------------+
| date   | uploads | deletes | net_change | STBotI | BetacommandBot | ImageTaggingBot |
+--------+---------+---------+------------+--------+----------------+-----------------+
| 200653 |   15430 |    9668 |       5762 |   NULL |           NULL |            NULL | 
| 200701 |   16806 |    6582 |      10224 |   NULL |           NULL |            NULL | 
| 200702 |   19384 |    6691 |      12693 |   NULL |            442 |            NULL | 
| 200703 |   19027 |    9415 |       9612 |   NULL |            844 |            NULL | 
| 200704 |   20418 |   10131 |      10287 |   NULL |           6730 |            NULL | 
| 200705 |   18804 |   27043 |      -8239 |   NULL |           3299 |            NULL | 
| 200706 |   18204 |   10096 |       8108 |   NULL |           3217 |            NULL | 
| 200707 |   18672 |   10308 |       8364 |   NULL |           1997 |            NULL | 
| 200708 |   18940 |   10477 |       8463 |   NULL |           3122 |            NULL | 
| 200709 |   18868 |   11518 |       7350 |   NULL |           1644 |            NULL | 
| 200710 |   19119 |    7684 |      11435 |   NULL |           1774 |            NULL | 
| 200711 |   19470 |    7527 |      11943 |   NULL |           1163 |            NULL | 
| 200712 |   19011 |   11122 |       7889 |   NULL |            387 |            NULL | 
| 200713 |   18766 |    9469 |       9297 |   NULL |            653 |            NULL | 
| 200714 |   19975 |   11849 |       8126 |   NULL |           2972 |            NULL | 
| 200715 |   19276 |   10312 |       8964 |   NULL |           1407 |            NULL | 
| 200716 |   18532 |   11924 |       6608 |   NULL |           2956 |            NULL | 
| 200717 |   18456 |   13181 |       5275 |   NULL |            111 |            NULL | 
| 200718 |   17389 |   10570 |       6819 |   NULL |           5831 |            NULL | 
| 200719 |   17170 |   30934 |     -13764 |   NULL |           3805 |            NULL | 
| 200720 |   18253 |   13030 |       5223 |   NULL |            587 |            NULL | 
| 200721 |   18168 |   11972 |       6196 |   NULL |          22225 |            NULL | 
| 200722 |   19529 |   19790 |       -261 |   NULL |          62662 |            NULL | 
| 200723 |   17637 |    9196 |       8441 |   NULL |           2451 |            NULL | 
| 200724 |   16805 |   13605 |       3200 |   NULL |          13403 |            NULL | 
| 200725 |   17398 |   19702 |      -2304 |   NULL |           4392 |            NULL | 
| 200726 |   17731 |   16973 |        758 |   NULL |          12172 |            NULL | 
| 200727 |   18195 |   23334 |      -5139 |   NULL |           8473 |            NULL | 
| 200728 |   17544 |   16781 |        763 |   NULL |           9015 |            NULL | 
| 200729 |   17113 |   21325 |      -4212 |   NULL |           5464 |            NULL | 
| 200730 |   17632 |   20553 |      -2921 |     32 |           6331 |            NULL | 
| 200731 |   15914 |   12006 |       3908 |   2873 |           4632 |            NULL | 
| 200732 |   16492 |   10832 |       5660 |   3958 |           1364 |            NULL | 
| 200733 |   17258 |    9891 |       7367 |   2452 |           2783 |            NULL | 
| 200734 |   13122 |   10716 |       2406 |   NULL |          12550 |            NULL | 
| 200735 |   13843 |    7803 |       6040 |   NULL |           5308 |            NULL | 
| 200736 |   12876 |    9269 |       3607 |    994 |          14208 |            NULL | 
| 200737 |   12197 |   18171 |      -5974 |   2837 |           9523 |            NULL | 
| 200738 |   11993 |    9864 |       2129 |    491 |          10063 |            NULL | 
| 200739 |   12993 |   12442 |        551 |   3595 |           8437 |            NULL | 
| 200740 |   11960 |    9683 |       2277 |   4369 |           4921 |            NULL | 
| 200741 |   12612 |   15307 |      -2695 |   2278 |           4050 |            NULL | 
| 200742 |   12511 |   10498 |       2013 |   2683 |          31843 |            NULL | 
| 200743 |   12911 |   17762 |      -4851 |    864 |          10358 |            NULL | 
| 200744 |   18094 |   16616 |       1478 |    302 |          49991 |            NULL | 
| 200745 |   11421 |   37841 |     -26420 |   1003 |          15117 |            NULL | 
| 200746 |   11824 |    9134 |       2690 |    388 |           3840 |              20 | 
| 200747 |   12180 |    5378 |       6802 |   2069 |          19049 |            1145 | 
| 200748 |   10898 |   12379 |      -1481 |   2604 |           9599 |             716 | 
| 200749 |   11124 |   13256 |      -2132 |   2232 |           7292 |             854 | 
| 200750 |   17773 |   13009 |       4764 |   1166 |          12370 |            1478 | 
| 200751 |   12634 |    8681 |       3953 |   1766 |           4878 |             940 | 
| 200752 |   12367 |    8348 |       4019 |   4381 |          57858 |             613 | 
| 200801 |   12283 |   13124 |       -841 |   2865 |           1960 |            1119 | 
| 200802 |   13864 |   11159 |       2705 |   2506 |           5982 |             855 | 
| 200803 |   12623 |   11855 |        768 |   2854 |          31301 |             719 | 
| 200804 |   11495 |    7652 |       3843 |   3559 |            164 |             487 | 
| 200805 |   11005 |    7356 |       3649 |   5867 |            293 |             133 | 
| 200806 |   11323 |   11318 |          5 |   5437 |          76583 |            1056 | 
| 200807 |    1255 |     687 |        568 |    362 |            900 |             201 | 
+--------+---------+---------+------------+--------+----------------+-----------------+

Hey, thanks for that! :-) I've copied it over to Wikipedia:Non-free content criteria compliance. Could you put any more stats you find on the talk page there? Carcharoth (talk) 21:10, 17 February 2008 (UTC)

Over 600 THOUSAND images tagged by BetacommandBot. And people don't think a bot is necessary to do this work? People don't think the scale of this problem is enough to call it serious? --Hammersoft (talk) 17:02, 18 February 2008 (UTC)

As I said, more detailed numbers will help correct misunderstandings in future. Please help with the (lone) effort to actually write a useful page. Carcharoth (talk) 17:21, 18 February 2008 (UTC)

• I'm helping in my own way. In part, by showing how absurd some of the complaints are. --Hammersoft (talk) 17:24, 18 February 2008 (UTC)

Three more BetacommandBot threads

There are currently three more BetacommnadBot threads on ANI.

• Wikipedia:Administrators' noticeboard/Incidents#BetacommandBot is malfunctioning again
• Wikipedia:Administrators' noticeboard/Incidents#BetaCommandBot and NFCC10c - New discussion page
• Wikipedia:Administrators' noticeboard/Incidents#Betacommand's use of bot on MickMacNee's talk page

The last one is serious enough to keep in its own section for now, so that it doesn't get lost in the noise, and is more about Betacommand than BetacommandBot anyway. The other two could usefully be merged here though. Carcharoth (talk) 12:50, 18 February 2008 (UTC)

Avoiding drama with Betacommandbot during March 2008

Suggested action points:

• By 23:59 2nd March 2008, Betacommand will provide a clear schedule for his planned 'disputed non-free images' tagging runs using Betacommandbot during the rest of March 2008 and he will sticks to this schedule.
• Betacommand is asked to avoid using his bot to tag thousands of images in the week before the deadline (23 March 2008), and is encouraged to finish his tagging at least two full weeks before the deadline (ie. by 23:59 9th March 2008). This does not apply to images uploaded after 9th March 2008, or to images tagged before this date and untagged without being fixed.
• Betacommand is encouraged to tag older images and newly-uploaded images (ones uploaded this year) separately.
• When Betacommand announces that he is finished with his tagging runs, then existing tagged images will have their deadline extended closer to the 23 March deadline to allow editors time to work on them. As before, this does not apply to newly-uploaded images.

It is possible that Betacommand is already happy to do this, but I would like to get this on the record. Please add more action points above as needed, and discuss below. Carcharoth (talk) 09:42, 16 February 2008 (UTC)

Why do we need to do this, per my message above? Stifle (talk) 10:00, 16 February 2008 (UTC)

To avoid drama. Betacommand thinks there is a deadline. Others don't. This needs to be sorted out before the "deadline" arrives. Carcharoth (talk) 11:24, 16 February 2008 (UTC) Previous, unhelpful comment by me, replaced - sorry Stifle.

I agree. I don't particularly care much about the specifics, but Betacommand and his (her?) bot are the spearhead of this image-tagging drive and it'd be irresponsible not to have all our ducks in a row before proceeding with the tagging. Specifics like what images get tagged, what kind of response time uploaders are afforded, etc. etc. are important issues that should be sorted out so that later there isn't any confusion or worse, accusations of bad-faith editing. -- Y|yukichigai (_ramble argue ^check) 12:06, 16 February 2008 (UTC)

Surely it would make more sense to reprogram Betacommand into understanding the resolution correctly? Stifle (talk) 19:12, 16 February 2008 (UTC)

It seems there is deadline, so accordingly I am putting some of the unanswered bits above into subsections to try and get some clear answers. Carcharoth (talk) 12:57, 17 February 2008 (UTC)

Suggestion for reprogramming

I have transferred this from ANI where it was rolled up for starting forest fires, against the WP:don't start forest fires policy. MickMacNee (talk) 12:02, 16 February 2008 (UTC)
I am increasingly seeing how this bot works through its invasions of my watchlist and watching the results. What appears to be happening is that images with no rationale and images with an attempted rationale, but no article link, are tagged as being completely the same, i.e. deletabble in 7 days, i.e. here: [5]. I believe this is unnacceptable, as the cases of attempted rationales would clearly pass muster in a court, which is what the policy aims to meet. In order to decrease the stress and anger this bot is causing every time it goes on a 20,000 tag run, it should be reprogrammed to only run a few hundred images at a time, and to dump the resulting tagged images into an expert clearing house, before placing a huge warning and endless spamming of talk pages, only some of which are appropriate, so that images can be screened into the obvious 2 categories here, non-compliance full stop and in need of time and attention, and non-compliance with the bot's specification, but fixable in 10 seconds without stress and alarm by experienced editors, and the consequent uneccessary loss of images/editors. This would also increase the accuracy of the bot process as a whole itself, as the bot can be defeated easily by putting a gibberish rationale, as long as it contains the appropriate number of links. I know time/numbers is a factor, so I suggest images in the clearing house not given any attention in 7 days are then tagged and users warned as usual, irregardless. The clearing house can be accessed by anyone, in a collaborative effort that WP is supposed to be. MickMacNee (talk) 18:08, 15 February 2008 (UTC)

And still we continue with the "someone else should be doing rationales for my copyright image uploads" theme that has been running throughout this drama. What is stopping the uploaders doing these rationales for themselves? We can keep providing extra time, extra work spaces, extra help etc, but this doesn't solve the major problem: add the rationales in the first place (or when the nice bot tells you about a problem to give you a week to correct it) and the problem goes away. ➔ REDVEЯS knows how Joan of Arc felt 12:12, 16 February 2008 (UTC)

• The issue being, no one has shown this bot achieves this task, and even worse, it actively causes losses of images that are perfectly acceptable, due to it's harrassing and poorly written (in secret) code and operation. It is true that the specific issue of how to improve the process of getting all fur's to the correct format is being lost in the noise, this is purely down to no-one taking this issue seriously and letting betacommand do whatever he sees fit. The few images I have looked into shows me this bot is a blundering idiot at best, a complete incompetent at worst. MickMacNee (talk) 12:18, 16 February 2008 (UTC)
• And on "someone else should be doing rationales for my copyright image uploads", no, all my images are correct because I know what is required, but in trying to save some of the many that have popped up in my watchlist, I have seen what this bot is doing, and am making a suggestion as to how to stop the complaints, harassment, annoyance and general disruption it is causing. To pretend it is not is a pure head in the sand attitude. MickMacNee (talk) 12:21, 16 February 2008 (UTC)

Redvers, you think everyone knows the million rules about fair use and free use and the thousand steps about writing a fair use rationale? --Kaypoh (talk) 12:55, 16 February 2008 (UTC)

Perhaps if they don't understand how to license the images properly they should not be uploading them? Spartaz ^Humbug! 13:45, 16 February 2008 (UTC)

Again, that is not the whole issue with this bot, there are many many previously acceptable images being re-flagged, and the tidal wave operation of this poorly implemented bot, combined with the fact images from 2005! are being tagged and uploaders are long gone, this is causing anger amongst experienced editors, not just new uploaders. One admin had 65 tags placed on his talk page in one day. Other have just jacked it in and said 'delete them all, I give up', even though they were saveable images. Anyone minded to try and work with this bot in its supposed aim of not deleting images, is just overwhelmed. MickMacNee (talk) 13:56, 16 February 2008 (UTC)

Over in Wikipedia:WikiProject Trains we are having to deal with a rash of these, as BCB runs through and tags a huge percentage of the railroad herald images. The irony is that most of these appear in infoboxes which give mechanically interpretable evidence that image appears in a fair-use situation. I suspect that the album covers and DVD cases are similarly recognizable. At any rate, failing this, it would be useful to list the tagging where it can be gathered by the appropriate project. For the railroads we started going through all the articles manually, looking for problem images in the infoboxes. That's the kind of work we pay computers to do. Mangoe (talk) 14:23, 16 February 2008 (UTC)

Whether or not the images are being used appropriately, they need fair use rationales, and should have been given one when they were uploaded. J Milburn (talk) 16:41, 16 February 2008 (UTC)

See above about images uplaoded before the policy was changed. MickMacNee (talk) 16:57, 16 February 2008 (UTC)

Furthermore, BetaCommandBot frequently tags images that do have perfectly valid fair use rationales for deletion. Presumably because it is lonely and wants attention. Nandesuka (talk) 21:28, 16 February 2008 (UTC)

You are mistaken, if you think BCBot made a mistake, Ill gladly show you where your error is. BCBot does not tag valid rationales. either provide diffs or take back your comments. β^_command 21:36, 16 February 2008 (UTC)

I see at least three each day. If you can't be bothered to pay attention the edits your bot is making, then maybe it shouldn't be making them. And if you think "Find your bots mistakes by yourself" is a waste of your precious time, well, now you know how many of us who would rather be writing articles feel about your bot. Nandesuka (talk) 22:14, 16 February 2008 (UTC)

Nandesuka, Thats BS, I know for a fact BCBot does not tag images with proper rationales, either back up your claims, which you cannot do. or stop trying to make me look bad. β^_command 22:23, 16 February 2008 (UTC)

Fine, I'll give you one, since you're not inclined to check the work your bot does. The fair use rationale on this image was perfectly valid when Betacommandbot tagged it. Rest assured that I see a few of these every day. And rest assured I'm not "trying to make you look bad." Who needs to do that? Nandesuka (talk) 22:38, 16 February 2008 (UTC)

Guess what that image had an improper rationale. Rationales need to include the name of the article for which the rationale is for. that rationale did not have that. β^_command 22:54, 16 February 2008 (UTC)

I've never seen anyone miss the point so much as you. There is one article link on that image, as it is, that is a fully legal fully defendable rationale; however, if I removed the entire Fair Use Rationale section, and replaced it with Phantasie is the best game ever, making a demonstrably invalid rationale, your bot would not tag it in a million years, no matter how many 10,000 run days you do. I would estimate on the ones I see (as despite repeat requests, no analysis of this bot's effectiveness is forthcoming) that at least 30% of your tags are of this nature, a complete drain on the community and an annoyance to editors. MickMacNee (talk) 23:01, 16 February 2008 (UTC)

MickMacNee, correct the bot will not tag that image, but there will be a day that all images have been reviewed and the things that BCbot missed will get fixed. so stop complaining that the bot cannot ensure that every rationale is 100% correct. there are parts that cannot be examined by a bot. for that we have humans. but for what can be automated it is. so stop complaining and start helping. β^_command 23:44, 16 February 2008 (UTC)

87.2% of statistics being made up on the spot, I would estimate on the ones I see that 99.99999% of the bot's taggings are correct, including the one cited above. It's not hard to get an FUR right. Well, apparently it is hard for some people, who are also the ones who [don't want to tag their images at all] [want to wikilawyer and bitch in general and find this subject convenient] [want other people to do FURs or other stuff they don't want to do themselves] [want to claim they left because BCB drove them out but then keep editing whilst loudly declaring they've left because of BCB] please delete as applicable. Write correct FURs when you upload or when the nice bot tells you about a problem. Problem solved. Instantly, and with no drama. Ah, I see the flaw... ➔ REDVEЯS knows how Joan of Arc felt 23:38, 16 February 2008 (UTC)

Ther is so much wrong with this statement I don't even want to address, suffice to say you clearly don't get it. The only category I fall into there is possibly bitching, (which is nice and fluffily offensive), as explained to you by me ad nauseum, and if you think people would go to this much length to try and highlight what they see as an obvious shortcoming just for the sake of bitching, that's just sad, and speaks volumes for me about your attitude. Wind your neck in. You honestly think we're all idiots for not just ignoring the issues caused by this in my view (and not disproven at all in any sense of critical analysis, all we have to go on are the thousands of anecdotal examples, complaints and ye gods, drama!) inneffective piece of amateur-hour logic teamed up with an over-zealous operator, because you either a. don't care if acceptable images are deleted (non-comliance is non-compliance as some idiot on BCBtalk put it, how's that for a contrast to correct interpretation of IAR), or b. take no interest in images on WP at all. MickMacNee (talk) 01:06, 17 February 2008 (UTC)

Redvers, this is not about recent uploaders who have not bothered to add a rationale. This is about people who have tried to add a rationale (or did so a long time ago when things were more lax around here), got it wrong, and then have been deluged with noticed to fix their images. There is a problem here. This whole situation could have been managed better. I can guarantee 100% that if Betacommandbot stopped tagging today and the pre-23 March 2007 images were carefully sorted through by hand and fixed where needed and deleted where not needed, there would be zero drama. The drama comes from applying a blunt tool (Betacommandbot) to fix a problem with a single algorithm ("is the article named on the image page?"), instead of a more intelligent algorithm ("when was it uploaded?"; "has a rationale template been used?" "is there other evidence that someone tried to write a rationale and got it wrong?" and so on). You are usually good at understanding what people are saying, Redvers, so why are you seemingly wilfully misunderstanding things here? Carcharoth (talk) 00:50, 17 February 2008 (UTC)

MickMacNee is exactly right. Both Betacommand and Redvers have managed to ignore what has been said in this section, and worse, to misrepresent it. MickMacNee was talking about non-free images that clearly have something that any human can recognise as a rationale (sometimes even a section called "rationale"), but where the editor who wrote the rationale didn't name the article the image was used in. Sometimes such editors go to great lengths to carefully explain why they think the image should be used, and are clearly trying to do the right thing. It doesn't help that they then get a bot-generated message saying the rationale is wrong. Redvers responded by talking about people who upload images and don't bother adding a rationale. That is a completely different situation, and Redvers does those people who have tried to add rationales and "got it wrong" a great injustice by lumping them in with those who don't bother adding rationales at all. Furthermore, Betacommand's definition of a "valid rationale" is extremely narrow: the definition goes something like "if the article is named, it may be valid - if it isn't it isn't valid". Let's try, once again to lay this all out as clearly as possible. Consider three factors: date of upload (2005 compared to 2007), state of rationale (none; with article name but invalid; without article name but otherwise valid; completely valid), and who or what is doing the assessment (human or bot). That yields 8 different possibilities, but with varying results depending on whether BetacommandBot is doing the evaluation or whether a human is. Sure, the deleting admin should pick up the slack, but that doesn't always happen. And the bot, while being less discriminating than a human, still fails to pick up lots of invalid rationales, but no-one seems to care about these points. I've asked Betacommand several times to separate out those images with rationales (eg. by searching for a template being used, or for certain keywords) from those without no rationale at all. I think the response at the time was "it will take a few months to reprogram the bot", and my response, half-jokingly, was that that would be enough time to have this "deadline" arrive first. I'm no longer laughing. Why did Betacommand not reprogram his bot to separate out those images using a rationale template from those with no rationale? It would be nice to get an answer to that. Carcharoth (talk) 00:50, 17 February 2008 (UTC)

Carcharoth represents my position much more eloquently (and calmly) than I could. The example I gave -- which I assure you, is one of only many I could have picked -- was an example where any reasonable human being looking at the image page would conclude there was a valid fair use rationale. That Betacommand, being challenged, decides to dig in his heels and claim that this isn't so is emblematic of the problem. The bottom line here is that we have sent a robot to do a human's job, and, as a result, are thus discovering just how poorly a job can be done. Nandesuka (talk) 02:21, 17 February 2008 (UTC)

It appears that members of the WMF Board of Trustees that passed this resolution have a difficult time creating a valid Fair Use Rationale that will pass the Betacommand BOT. Erik Moller uploaded Image:Alexandria-sagan.jpg in December 2003. -- SWTPC6800 (talk) 02:21, 17 February 2008 (UTC)

One other consideration when admins go through these backlogs: some of these "fair use" images are, in fact, free. When the various image tags were consolidated (a year ago?), at least a couple tags used for some free images were lumped in with non-free. Gimmetrow 02:37, 17 February 2008 (UTC)

Gimme has a point, I've noticed that sometime ago a bot went through and changed all the logos from {{logo}} to {{Non-free logo}}. Not a major problem since 90% of logos are non-free, but that 10% that are simple non-copyrightable geometric shapes, also got caught up in the retagging. MBisanz ^talk 02:42, 17 February 2008 (UTC)

That's the main case I was thinking about. Logos. I've undeleted a few already that were "free", though the admins who deleted them didn't know it and apparently nobody who might have known was notified. Gimmetrow 02:48, 17 February 2008 (UTC)

The same applies to old black-and-white photos from the early 20th-century and late 19th century. Many of these are in fact public domain by age, but were tagged incorrectly or for lack of information. These shouldn't be deleted, but in the rush to clear these backlogs, many are being deleted. Looking at this list, I see Image:Furnesflooding.jpg, Image:Lawrence Brough Cranwell.jpg (there is a debate taking place on the image page that should be taking place on the talk page or at IfD), Image:Richard William Dowling.jpg (the uploader has misunderstood fair use - should have specified public domain), Image:Tsarbomb.jpg, Image:Aurora Ship.png (is that not public domain by age?), and so on. We really should be able to separate out images like this from the DVD covers and album covers and book covers and logos. Coverage of history is a clear educational and encyclopedic argument for fair use, and some Wikipedia image editors are woefully insensitive to historical matters. They just take a blinkered view that all non-free are equally bad and ignore whether a photo is of contemporary culture (last ten years) or whether a photo is of long-dead people and events from 70-80 years ago. I really hope that once the present image tidying up is over, we can concentrate on building up the historical images again, always remembering to search for free alternatives and remembering to use images where copyright has not been renewed. Carcharoth (talk) 13:14, 17 February 2008 (UTC)

I'm tempted to just say WP:SOFIXIT, if an image that is clearly public domain is labeled as non-free just be bold and fix the tagging (this would naturaly apply to admins who stumble across such images in the course of clearing out backlogs as well). If not entierly sure change the tagging and list the image on WP:PUI asking for a second opinion. If an image is just old, but still copyrighted, then it does need a rationale like any other non-free image, and tagging then for lack of such is entierly apropriate. If you are concerned about historicaly importnat images beeing lost in the purge I would reccomend starting a project to go though all images in Category:Non-free historic images for starters and make sure they all have apropriate rationales ahead of time. --Sherool (talk) 13:49, 17 February 2008 (UTC)

Sherool, I am thinking of starting such a project. I would also be interested in past members of Category:Non-free historic images that were deleted. Is it possible to get a list of images in that category that have been deleted? Carcharoth (talk) 15:27, 17 February 2008 (UTC)

Getting a list of images in a particular category that have already been deleted will probably be tricky. Best option I can think of is to have a bot generate a list of all the images currently there and then you can keep an eye on redlinks that crop up on that list from now on. To find images that have already been deleted would probably have to involve comparing older database dumps with the current list of images or something like that... --Sherool (talk) 16:33, 17 February 2008 (UTC)

I realise getting lists of what used to be in a category is tricky, but I'd assume much more good faith in people if they would actually spend time on such requests, and not just say "it could be done, but it is difficult", and then nothing gets done. Sure, it is undoubtedly difficult, but if you had to chose between clearing a backlog of images to be deleted and trying to do that request for me, which would you chose and why? I've already grabbed a list, and there are other lists and category trackers around as well, but it is the focus that is all wrong. An inability to distinguish between different sorts of non-free images and treat them all as "bad". The good ones get thrown out with the bathwater, and that causes huge amounts of resentment among good editors. I'm not talking about original uploaders who may be long gone, or current uploaders who fail to provide rationales, but the image-sensitive editors around now, who are trying to clear up backlogs and save the most needy images, but who are not always receiving all the help they could be getting from bot operators, who often are happier to focus on wholesale tagging and deletion, rather than fixing (though some do carry out fixing as well). Carcharoth (talk) 17:51, 17 February 2008 (UTC)

I herby retrack my biased comments We don't need no stinkin FUR!! 21:03, 19 February 2008 (UTC)

Second suggestion for reprogramming

It would be much easier to fix FURS as they are tagged if we could sort them into different types of media. A clearing house idea suggested in the preceding section is good but not as an expert clearing house; rather a place where we could inform the community of its existence and ask for assistance and spoonfeed how to do it with precreated templates (this would only work for media with essentially fungible rationales, such as those for dvd covers). So, using dvd covers as an example, we create a page: "Wikipedia:DVD cover mitigation" or something similar, and have the bot dump a log of all article it tagged of that media into it. I imagine the bot could be programmed to do this by logging which pages it tags contain the template {{Non-free DVD cover}}. Then it would just dump the links for all such articles it tagged into the associated mitigation page. On that page (and its brethren pages for film poster, book covers, album covers, etc.), we describe how to add {{Filmr}} to the article links dumped, and instruct users to remove the links as they fix them. Then we advertize the mitigation pages like crazy, community portal, the pump, maybe in one of those annoying pupups you get when you go to your watchlist, and so on. This would avoid placing any onus of fixing or adding rationales on just the admins seeing them ready for deletion, and might actually allow us not to lose thousands of images that actually improve the many articles they are in. The sheer size of the problem means that unless we get a realistic number of users involved in fixing the problem in some way, we will be overwhelmed, and will be forced by necessity to delete much that could be fixed. Of course this is dependant on betacommand doing the reprogramming. Is this feasible?--Fuhghettaboutit (talk) 04:02, 17 February 2008 (UTC)

Do [6] and [7] do this already? MBisanz ^talk 04:37, 17 February 2008 (UTC)

Wow, they sure look like the right type of thing! My question is are those at all comprehensive and how are they created. Are they hand sorted or compiled from logs, and how? If they are compiled in some compehensive fashion, and workable to keep up, then that is the data needed to implement my idea above. But of course they have to be moved or duplicated to central pages as I described, text added so that users know what to do with the data, and then those pages advertized. I'd be willing to start that soon if I had some answers to the questions I started this post with.--Fuhghettaboutit (talk) 04:48, 17 February 2008 (UTC)

User:AWeenieMan updates [8] and when he's not around, I do the update. Its a standard script he wrote that is comprehensive. User:East718 updates [9] on an as-requested basis, using a comprehensive cat-comparer. AWeenieMan's script takes about an hour to run and East's takes about 5 hours. They are both based on cross-tabing the Category:Disputed non-free images days in that cat when BCB runs his bot, with the all-inclusive cats like Category:All non-free Logos. When WP:TODAY was in its discussion phase, the idea of a site-notice on the watchlist was rejected I believe as being the wrong use of that feature. There are pages in the WP:TODAY space that do tell a user what to do, and of course, there is the FURME tool that tends to do it for them. MBisanz ^talk 06:04, 17 February 2008 (UTC)

Is it possible to cross-reference the "disputed non-free images" with "images using a rationale template"? Most of those will likely be in "non-free images lacking article backlink", and will have incomplete rationales, needing no more than an article link added to them. Carcharoth (talk) 12:57, 17 February 2008 (UTC)

That would be easy for someone with the tools to do. WP:AWB has that feature, but my internet connection limits me from doing it. East718 or AWeenieMan could probably do it rather quickly. MBisanz ^talk 14:59, 17 February 2008 (UTC)

As for my tool, the amount of time it takes to run is directly related to the number of images in Category:All disputed non-free images. Basically, it just pulls a category listing, then pulls the categories for all the images on that list, the separates based on the list of categories I program in (currently Category:Album covers, Category:All non-free Logos, Category:Book covers, and Category:Non-free historic images, because that is what I have built into FURME, but it is trivial to look for other categories). That part actually goes pretty quickly, it is the image usage part that eats up a lot of time. Hopefully I will get this amount reduced when my toolserver account is created. I could most likely do what Carcharoth is asking with little trouble. I would just need a list of what rationale templates to include (as far as I know there isn't an actual category "images using a rationale template," or am I missing something?). That and do you just want one big list? If you leave me a message with exactly what you want over on my talk page, it shouldn't be too hard to write a script that will create (and update) the list for you. - AWeenieMan (talk) 19:06, 18 February 2008 (UTC)

Here's a list of such images, if anyone is curious. - AWeenieMan (talk) 21:54, 18 February 2008 (UTC)

Third suggestion for reprogramming

• Betacommand (or others) are encouraged to tag older images (pre-2008) and newly-uploaded images (ones uploaded in 2008) separately and place the images in lists or categories to enable the workload to be separated between newer and older images.

If you support this, please say so here. Carcharoth (talk) 12:57, 17 February 2008 (UTC)

This is a rather odd suggestion. Why does the beginning of 2008 become a magical cut-off date? If there is to be a date which marks the end of the "legacy" period for images lacking rationales, it ought to be 4 May 2006, the date when lacking a rationale explicitly became a criteria for speedy deletion (even though the requirement for a rationale has always been part of non-free content policy going back to the earliest revisions). --bainer (talk) 14:51, 17 February 2008 (UTC)

Yes, you are right, 4 May 2006 is a much better cut-off date, but really, it should be possible to sort all the tagged images by upload date. See this user analysis of how image notification sometimes fails, for an idea of why legacy images should be treated differently. For what it is worth, there was also Wikipedia:Non-free content criteria/legacy image proposal, and there was Wikipedia:Non-free content criteria/Proposal, which was (I think) the main reason why Betacommand held off on legacy images until 2008. That proposal, like many others, unfortunately never gained traction, and so once 2008 had arrived, Betacommand went ahead with running his bot over the large lists of legacy images he had, which is why the complaints reached new levels recently. I've been trying to gather these links together, and make some sense of all this, at Wikipedia:Non-free content criteria compliance. I'm going to add the 4 May 2006 date there, and if you or others could provide more, that would be great. Carcharoth (talk) 15:23, 17 February 2008 (UTC)

March 2008 schedule for BetacommandBot

• By 23:59 2nd March 2008, Betacommand will provide a clear schedule for his planned 'disputed non-free images' tagging runs using Betacommandbot during the rest of March 2008 and he will stick to this schedule.
• Betacommand is asked to avoid using his bot to tag thousands of images in the week before the deadline (23 March 2008), and is encouraged to finish his tagging at least two full weeks before the deadline (ie. by 23:59 9th March 2008). This does not apply to images uploaded after 9th March 2008, or to images tagged before this date and untagged without being fixed.

Please can people give some feedback on this. I think it really will help if we plan how things are done in March. Hopefully Betacommand will agree to this voluntarily (and I would be the first to thank him profusely for doing this), but ultimately if there is support in the community for this, the community could go to WP:BAG and ask them to talk to Betacommand, or indeed take other measures. If you want to see things done carefully and calmly in the final month before the WMF non-free image deadline, please support this proposal. Carcharoth (talk) 12:57, 17 February 2008 (UTC)

Carcharoth, about 5 months ago I gathered some stats, we had ~500 non-free image uploads per day. (that number is net non-free images). at that point about half of them did not pass NFC. Im guessing that BCBot will tag about 1000 images per run from now on. I dont know for a fact, but all older images should be tagged, (BCBot might have missed a few) but for the most part I think the mass tag runs should be over. Due to real world issues I cannot give exact dates for each run, but I plan on running the bot twice a week. I plan on generating a new list of NFC and doing ORFU tagging shortly. when that is done I am going to re-generate the NFC list and do a NFCC10c run. β^_command 18:30, 17 February 2008 (UTC)

Cool. Really good to know that the main tagging is effectively finished. Would you object to the fixing deadlines being rolled back another week? I wanted to look through Category:Disputed non-free images as of 24 January 2008, but won't have time tonight. The net image uploads seems to be around 10,000-20,000 per week (around 2000 per day), so non-free images seem to be about a quarter of the total uploads. I share Geni's concerns that a large number of copyvios are sneaking through and are being labelled as "free" images. Do we have any way of checking that? Carcharoth (talk) 23:45, 17 February 2008 (UTC)

Nothing automated, though a fair number of those also don't include any source info so they get picked off by bots that way. There are a number of users who spend quite a bit of time browsing though various free license image categories looking for at last the most obvious copyvios, but yeah there is a lot of dodgy stuff out there, and the more eyeballs we can get double checking our images the better. --Sherool (talk) 12:28, 18 February 2008 (UTC)

• Sounds fair. Will ^(talk) 14:28, 17 February 2008 (UTC)

  Yep, seems like that could work. MBisanz ^talk 15:00, 17 February 2008 (UTC)

New discussion pages

• See Wikipedia:Bots/BetaCommandBot and NFCC 10 c. MickMacNee (talk) 14:53, 17 February 2008 (UTC)
• And Wikipedia:Non-free content criteria compliance for an idea of how to handle things in more general terms, involving other bots and humans to work together and reduce the reliance on BetacommandBot for doing this important and necessary work. Carcharoth (talk) 15:25, 17 February 2008 (UTC)

I have created the folowing page Wikipedia:Bots/BetaCommandBot and NFCC 10 c to attempt to centralise discussion on BCB and specifically its NFCC10c tagging operation.
MickMacNee (talk) 14:26, 17 February 2008 (UTC)

• This page is nothing more than an attempt at forming a lynch mob for the bot and its operator. From the get go, it is inherently biased. I strongly urge it be shut down as the sham it is. --Hammersoft (talk) 16:13, 17 February 2008 (UTC)

The page is a legitimate effort to fulfill requests to me and others to stop starting 'forest fires' on ANI, and to clear up the confusion caused by general replies to what are specific concerns with the bot, this comment being a good example. You are a complete and utter wind up merchant. MickMacNee (talk) 16:28, 17 February 2008 (UTC)

• The page is a witch hunt, nothing more, nothing less. If you continue the personal attacks against me, I will file a formal complaint against you. Please read WP:NPA. There is no need for you to insult me. --Hammersoft (talk) 23:03, 17 February 2008 (UTC)

• Hammersoft, do you think that this behaviour is acceptable? Betacommand is clearly responding to MickMacNee's creation of that page by deluging his talk page with notifications of images that MickMacNee has not uploaded. The block of MickMacNee looks unrelated, but I will ask SwatJester (who is active on that talk page) to investigate. Carcharoth (talk) 11:51, 18 February 2008 (UTC)

• I guess bot operators are excused from WP:POINT. Good to know. rudra (talk) 11:55, 18 February 2008 (UTC)

Bot should be blocked pending explanation. Either it's malfunctioned, and a malfunctioned bot with a high edit rate is very dangerous (ask me and thehelpfulone, who had to rollback 800 of BCB's contribs when it started tagging all images as orphaned, and all those edits took part within 6 minutes), or it's editing to make a point, which is much worse, and people have been indef blocked for less. Yes, I do normally support the bot (I supported on the discussion page as I think it'd help to know when it'll run, and be a bit friendlier with 10c tagging), but the MickMacNee talk spamming is very serious. Will ^(talk) 12:01, 18 February 2008 (UTC)

This is serious enough to warrant a new section. See Wikipedia:Administrators' noticeboard/Incidents#Betacommand's use of bot on MickMacNee's talk page. Thanks. Carcharoth (talk) 12:38, 18 February 2008 (UTC)

BetacommandBot is malfunctioning again

BetacommandBot (talk · contribs · deleted contribs · logs · filter log · block user · block log) just removed an image from an article "beacuse "";" (a rationale which I quote in its entirety) , and tagged an image for speedy deletion on the basis of a decidedly vague complaint (perhaps the image needs to be speedily deleted "beacuse "";", I suppose). The bot also substitutes instances of templates occurring in template documentation, as a result of which the description of the template output won't be automatically updated when the template is changed -- see [10] [11] [12] [13], for example. If that weren't bad enough, the bot makes bizarre null edits to userpages -- see [14] [15] [16] [17]. John254 05:41, 17 February 2008 (UTC)

I don't think BcBot uses WP:TWINKLE, Betacommand may have just been logged in as the wrong user. SQL^{Query me!} 06:01, 17 February 2008 (UTC)

Yes, it would be possible that Betacommand accidentally logged into his bot account, then removed an image "beacuse "";" -- except that User:BetacommandBot/monobook.js was blank at the time of the offending edit, so Betacommand wouldn't have access to WP:TWINKLE when logged into the bot account. Which, of course, leaves us with one of three possibilities:

(1) Betacommand logged into BetacommandBot, then removed an image "beacuse "";", and entered an edit summary indicating that he used WP:TWINKLE even though he didn't.

(2) BetacommandBot is a compromised account

(3) BetacommandBot is malfunctioning

John254 06:30, 17 February 2008 (UTC)

Didn't see that. Have you notified him of the problems at all? SQL^{Query me!} 06:33, 17 February 2008 (UTC)

I have informed Betacommand of the discussion here. However, given the amount of damage which a malfunctioning bot can cause, I would suggest that BetacommandBot be blocked until Betacommand can fix the problem. John254 06:47, 17 February 2008 (UTC)

Given one messed up edit summary, and the MP templates, I think Beta is testing some code. Note the regex-like edit reverted as betacommandbot.[18] I'm not inclined to block for this, although I'm curious why some edits are recorded but appear to do nothing.[19] Gimmetrow 06:35, 17 February 2008 (UTC)

May I respectfully suggest that further test edits be made in a userspace sandbox, rather than damaging articles, images, and template documentation? John254 06:47, 17 February 2008 (UTC)

I notified Betacommand of this. Tiptoety ^talk 06:43, 17 February 2008 (UTC)

I believe he is testing out what he proposed here Wikipedia:Administrators'_noticeboard/Incidents#Next_BCBot_Phase MBisanz ^talk 06:53, 17 February 2008 (UTC)

The discussion linked there contains no mention of whatsoever of disruptively substituting templates on their documentation pages, or of making null edits to userpages. In any event, the next time Betacommand or BetacommandBot removes an image from an article or tags an image for speedy deletion, I surely hope that a better explanation than "beacuse "";" is provided. John254 07:04, 17 February 2008 (UTC)

Yea, the null edits in the userspace is bizzare, didn't think the system would let that actually be recorded. I'm looking at the templates, and they contain images with instructions that the templates should be subst'd in actual use. That leads me to think this is either some weird transclusion error with the image that is messing up the bot or that the bot is searching for instances of a template that should be transcluded and then actually doing it. MBisanz ^talk 07:09, 17 February 2008 (UTC)

The templates should be substituted in normal user talk page usage, to record the warnings that were actually given, and to minimize the use of system resources (though there's a disagreement as to whether bot template substitution actually consumes more resources than it saves.) The problem here is that the bot has apparently been programmed to substitute all usages of the templates, even where they are used in template documentation, in which they should be transcluded, so as to update the text of the templates contained in the documentation when the templates themselves are updated. John254 07:20, 17 February 2008 (UTC)

The GTA/Brooklyn Bridge was a one-off; that will no doubt be fixed. The bot has not edited for almost two hours. Gimmetrow 07:16, 17 February 2008 (UTC)

Why is BetacommandBot so controversial? It is performing a necessary task, but must it anger so many people in the process? I suggest instead that Betacommand enlist willing volunteers to work on rationales, rather than trying to dragoon editors into having to volunteer even more of their time. Part of the problem is that the editors are untrained, and BetacommandBot's notifications 1) come across as a hostile challenge, 2) do not adequately tell people exactly what is wrong with their rationale, and 3) will spam an editor who has uploaded multiple images, rather than inserting a single text block pointing to multiple images. These problems are technical problems and could be addressed, but Betacommand is seemingly unwilling to anger less people. RussNelson (talk) 17:17, 17 February 2008 (UTC)

Again I ask people who insist on bringing Betacommandbot issues here to stop doing so. Betacommand has a user talk page, and any admin can block a malfunctioning bot. We don't need to clutter this page with every single instance where this very, very busy bot does something unexpected. --Tony Sidaway 18:10, 17 February 2008 (UTC)

There is a cryptic message on the user's talk page that advises editors with complaints to go to another source page. Inexperienced users still see the bot's messages on their image and will tend to revert or delete the image rather than revising the image information. The bot does not seem to read anything outside the new templates very well even when the image rationale is written out clearly. FWIW Bzuk (talk) 19:03, 17 February 2008 (UTC).

See Wikipedia:Bots/BetaCommandBot and NFCC 10 c MickMacNee (talk) 19:11, 17 February 2008 (UTC)

Betacommand archives the bot's talk page without responding. I can understand why he doesn't want to deal with all the complaints, but Tony, don't suggest this as a way to accomplish anything. rspeer / ɹəədsɹ 19:25, 17 February 2008 (UTC)

So, who is that authorized this Betacommand character to become lord and master of wikipedia? He can't be getting away with this stuff on his own. Someone higher-up must have told him it was OK to do what he does, attitude and all. Baseball Bugs ^{What's up, Doc?} 19:33, 17 February 2008 (UTC)

• Nobody made Betacommand lord high almighty of Wikipedia. That said, all the bot's routine functions have been approved by WP:BAG. --Hammersoft (talk) 01:29, 18 February 2008 (UTC)

• the TW edits where made by me, (forgot to switch back to my account) I was doing a little cleanup with BCBots watchlist, Im not sure wny mediawiki choose to save null edits, it normally does not do that. β^_command 03:21, 18 February 2008 (UTC)

Edits by Betacommand to MickMacNee talkpage

I would like the administrative community to review the following incidents concerning User:MickMacNee and User:Betacommand:

• Following heated debate at ANI and AN and User talk:Betacommand, User:MickMacNee created Wikipedia:Bots/BetaCommandBot and NFCC 10 c at 14:21, 17 February 2008.
• Betacommand's main response seems to be here (02:06, 18 February 2008): "that page is full of lies and bullshit and is a complete farce and it will not affect how I operate BCBot".
• Earlier in the day, but after the creation of Wikipedia:Bots/BetaCommandBot and NFCC 10 c, Betacommand had left this message for MickMacNee: "Since you so gladly asked for more communication and want to assist in helping fix images that BCBot identifies as failing WP:NFCC, I am going to have the bot notify you when it tags an image." (19:50, 17 February 2008)
• Betacommand then, after the "bullshit" note at 02:06, presumably got the bot to make 46 edits to MickMacNee's talk page between 04:33 and 04:38, 18 February 2008. This resulted in 23 image deletion tag notifications on MickMacNee's talk page. None of the images are ones that he uploaded or edited, as far as I can tell. Some edits, like this one are rather strange, seemingly replacing one image notification with another, but that is a secondary concern to the spamming of MickMacNee's talk page that seems to be an angry and pointy response to MickMacNee's criticism of Betacommand and his bot, and an abuse of BetacommandBot's bot flag as well. Is this acceptable behaviour?
• Additional notes: (a) It should be noted that User:MickMacNee has recently been blocked for 3RR at Bear Grylls (along with an IP address that was also edit warring), but that block (at 19:29, 17 February 2008) looks to be unrelated. (b) Another note is that the page MickMacNee created was nominated for deletion by User:Kylu at 05:44, 18 February 2008. See Wikipedia:Miscellany for deletion/Wikipedia:Bots/BetaCommandBot and NFCC 10 c. (c) Finally, a warning was left at User talk:Betacommand by User:Stifle, backed up by me here (11:12 and 11:38 18 February 2008).

What needs to be done here? Have previous lessons been learnt? Carcharoth (talk) 12:23, 18 February 2008 (UTC)

Block BetacommandBot until this is fixed. --Kaypoh (talk) 12:32, 18 February 2008 (UTC)

Block BetacommandBot until Betacommand agrees that it is not his private toy, to be used to attack anyone at will. RussNelson (talk) 14:10, 18 February 2008 (UTC)

I agree that the spammed warnings were completely inappropriate, but the bot hasn't edited in 7 hours, and it made over 2000 edits after it stopped leaving notices on MickMacNee's talk page. I'm not sure I see an immediate need for a block to prevent further disruption. --Onorem♠Dil 12:38, 18 February 2008 (UTC)

How about a very stern warning that the next time something like this happens, both the bot and the bot operator will be blocked for a week? And please (talking to everyone in general), can those who don't like what the bot does, not start baying for blood? There are clear problems with Betacommand's temperament and whether he really can learn from his mistakes. These issues go to the core of whether he should be running a powerful tool like a bot. Let's keep the focus on that. I would have no problems with someone else running the bot - someone who wouldn't abuse it in the way Betacommand just did. Carcharoth (talk) 12:44, 18 February 2008 (UTC)

Agreed. Beta shouldn't have spammed MacNee's talk page, but to be honest, given the amount of unwarranted abuse he gets, I can understand it occasionally. I am a patient person but in Beta's place I'd have given up trying to assist the encyclopedia long ago. Black Kite 12:50, 18 February 2008 (UTC)

As I've said above: bot should be blocked pending an explanation. Either it's malfunctioned, and a malfunctioned bot with a high edit rate is very dangerous (ask me and thehelpfulone, who had to rollback 800 of BCB's contribs when it started tagging all images as orphaned, and all those edits took part within 6 minutes), or it's editing to make a point, which is much worse, and people have been indef blocked for less. Yes, I do normally support the bot (I supported on the discussion page as I think it'd help to know when it'll run, and be a bit friendlier with 10c tagging), but the MickMacNee talk spamming is very serious. Will ^(talk) 13:00, 18 February 2008 (UTC)

Agreed. I'm prepared to block it now, pending an explanation. I'm not going to take the chance that it will continue to spam MMN's talk page when it resumes again. We need some assurances. El_C 13:07, 18 February 2008 (UTC)

What explanation do you need? As per above, it was obviously a pointed response to what Beta saw as an attack page on him created by MickMacNee. Censure Beta if you wish, but blocking the bot, especially at this point, would be extremely unhelpful to the project. Also, as pointed out, the bot has made 2,000 useful edits since the spam incident, so I think we can say it's not an ongoing problem. Blocks are preventative not punitive. Black Kite 13:09, 18 February 2008 (UTC)

You'd rather we do nothing and let it continue with the pointy disruption upon resumption? A plan of inaction? El_C 13:12, 18 February 2008 (UTC)

(unindent) My view would be that it hasn't spammed for 2,000 edits. It was clearly a short one-off "f*** you" to MacNee, and it was a Bad Idea, but as I say, censuring Beta is the best way here. If it recurs, then fine. Black Kite 13:15, 18 February 2008 (UTC)

So, you're willing to take that chance, then. Do I got this right? And no consequences, either? I didn't realize image work gets one full immunity from conduct policy. (I'm being somewhat flippant, of course I knew that, we all know that) El_C 13:19, 18 February 2008 (UTC)

(EC)If this were a vandal, I wouldn't block because the report is stale - the activity is not ongoing, and the vandal is not active now. The same applies here; neither the bot or the op are active right now, and so a block would not prevent damage to the encyclopedia. If, having been warned that this sort of shenanigan is unacceptable, BC continues notifying unrelated users who did not explicity request such notices, then a block would be in order. BC does not get a pass, but I don't think a block in this case would be helpful. UltraExactZZ ^Claims ~ Evidence 13:20, 18 February 2008 (UTC)

(e/c) No, I am not saying that, merely pointing out that censuring the operator would be better than blocking the bot, which would be counter-productive at the moment. I am probably kicking a deceased equine given what passes as consensus on the bot's work at the moment, but there we go. That's my view, I'll recuse from this now. Black Kite 13:25, 18 February 2008 (UTC)

(after a gazillion edit conflicts)Unless the pointy disruption is on-going, or there is evidence that it will resume shortly, a block would merely be punitive at this point. I don't agree at all with BC's reaction, but MickMacNee does appear to have done everything possible to get a rise out of BC (and anyone who agrees with BCB's operations) and has achieved that objective. Throwing further petrol on this with a block wouldn't help. ➔ REDVEЯS knows how Joan of Arc felt 13:21, 18 February 2008 (UTC)

This incident is clearly an abuse, clearly vandalism, and worst of all was clearly retaliation for discussion. And it needs to be taken seriously. And that's all I'll say about it. Nandesuka (talk) 13:22, 18 February 2008 (UTC)

We are told that our preventative measures are limited to hoping that the pointy disruption isn't still programmed in the bot's next run. El_C 13:24, 18 February 2008 (UTC)

This is getting to be intolerable, frankly, I'd block the bot, remove it's flag and ask someone else with experience of coding bots to takeover this task. This heavy handed and frankly unacceptable pointy approach to dealing with problems is causing more problems than it solves, and it's alienating users who would normally be willing to help. Nick (talk) 13:23, 18 February 2008 (UTC)

(Responding to BlackKite up there somewhere) It was not the bot that carried out this action. It was clearly Betacommand abusing the bot and/or instructing it to leave notices in the wrong place. If MickMacNee had done a "f*** you" series of pointy edits like that, he would be sitting on a long block right now. I agree that the bot does not need to be blocked, but until Betacommand apologises and says he won't do this again, there may be a need to indefinitely block Betacommand as a temporary measure until he says he won't do this again. Carcharoth (talk) 13:26, 18 February 2008 (UTC)

I think that's pretty much what I was saying. Black Kite 13:30, 18 February 2008 (UTC)

Yeah. Or at the least leave a very strongly worded warning that this is the last chance. Next time this happens, there will be a block to record this ongoing disruption in the block log. This is exactly the sort of behaviour that people should be able to see when loooking through a block log to see if there is a pattern of unacceptable behaviour. Carcharoth (talk) 13:33, 18 February 2008 (UTC)

By the way, what if MickMacNee logs on while his (unrelated 3RR) block is still active? Should he be unblocked to say something here, or not? Oh, and should the tags be removed from MickMacNee's talk page? Carcharoth (talk) 13:26, 18 February 2008 (UTC)

If MickMacNee has a substantive comment to add to the discussion, then I will be glad to copy it over here, as would several others. If this were an ArbCom discussion (Not that I'm suggesting that!), I'd offer a parole limited to editing their talk page and this discussion only - but we're not there. UltraExactZZ ^Claims ~ Evidence 13:33, 18 February 2008 (UTC)

And the tags? The tags being there are embarassing Betacommand more than anything, and ideally Betacommand himself would remove them with an apology, but I don't want the tags being there to cause MickMacNee to react in a way he might regret. It would be easy to link to an old talk page version to show what the result of the talk page spamming was. Carcharoth (talk) 13:38, 18 February 2008 (UTC)

This also raises another problem with this bot (and others) that I mentioned many times: their gross inefficiency when it comes to multiple entries. Rather than have multiple items noted in a single thread, in a list format, it repeats an individual notice as many times. It's totally absurd, and to a great many users (many of whom uploaded images, at the time, by-the-book) overwhelming. El_C 13:51, 18 February 2008 (UTC)

I'm still concerned with preventatives. How about we do this: either I block BC now, indefinitely, until he provides assurances that the disruption isn't still programmed in, then unblock. Or, we wait and if it is still going, I block the bot for two weeks for abuse -cum- incompetence to curtail-it-in-the-second-round (oh, and I take to arbitration anyone who reduces the block, it having been announced). BlackKite, you choose. El_C 13:37, 18 February 2008 (UTC)

If I may, I think you should wait. I doubt BetacommandBot will still be doing this when it is set to editing again. And anyway, the two-week block wouldn't stick as the one-week one last week by Tim Starling didn't stick for more than a day or two. Carcharoth (talk) 13:41, 18 February 2008 (UTC)

Sure, I'll wait. The arbitration case against whomever would lift it would stick, in that event. El_C 13:45, 18 February 2008 (UTC)

As for your to be honest, given the amount of unwarranted abuse he gets, I can understand it occasionally. I am a patient person but in Beta's place I'd have given up trying to assist the encyclopedia long ago, I don't buy that for one second. I take as much abuse, certainly in quality if not in quantity, and I don't habitually curse at people. Somehow, we treat bot programmers as if they are more special than the rest of us. They're not that special. El_C 13:43, 18 February 2008 (UTC)

Blocking, by either of you, at this point would be punitive and would pour petrol on the fire. Please don't. ➔ REDVEЯS knows how Joan of Arc felt 13:45, 18 February 2008 (UTC)

Note: Redvers comment "Blocking, by either of you" refers to El C and East718. See here and here. Carcharoth (talk) 09:58, 19 February 2008 (UTC)

Again, no problem. Two-weeks it is, then, in that event. El_C 13:54, 18 February 2008 (UTC)

Is a concensus emerging here? By the way, a warning was left by Stifle at 11:12 this morning, which was after Betacommand stopped editing, but I've noted it above for the record. Carcharoth (talk) 13:52, 18 February 2008 (UTC)

BetaCommand Break

• People conducting attacks against other people rarely achieve the results they want. More often than not, it hurts themselves far more than it hurts their target. Betacommand has committed a serious judgment error. Betacommand's actions in this are out of line and completely unacceptable. I do not feel a block is the correct behavior, even pending an explanation/assurance from Betacommand. You do not block someone and them demand they acquiesce to your demands under pain of not being unblocked. That's a punitive action. I do feel that a very stern warning should be given to Betacommand, and perhaps have the Bot Approvals Group be made aware of this abuse of the bot flag. This is clearly unacceptable behavior and must not be tolerated.
• With the above said, I agree with others above; the amount of abuse that has been heaped upon Betacommand is unreal. The large majority of editors would have already snapped under the pressure. An increasingly vogue method of getting your way on Wikipedia these days is to unceasingly harass your target, skirting the edges of policy under thinly veiled attacks (such as Wikipedia:Bots/BetaCommandBot and NFCC 10 c). Eventually, you wear your target down and they go apeshit crazy and do something stupid. In my opinion, MickMacNee has done precisely that. He's been unrelenting in his abuse of Betacommand and has routinely been willing to use various methods of attack to get his way. No surprise that Betacommand snapped, even if it is completely unacceptable. --Hammersoft (talk) 13:57, 18 February 2008 (UTC)

• (countless EC) *Yes it was a minor violation of POINT, Given the repeated abuse and misleading comments and an attack page that was set up, its a very minor response. yes I lost my cool, My response was a very limited and controlled outburst. El C you have no clue the amount of abuse I take, yes you might get the run of the mill abuse toward admins, but its nothing compared to what is done to me. users who make these attack seem to be immune to NPA and CIVIL when they talk to me, Ive been called everything in the book and a few more. what I did was a small demonstration to make a point. β^_command 14:00, 18 February 2008 (UTC)

• Conversely, I don't think you have a clue about the abuse I take. Volume isn't everything. I often deal with the scum of the earth. "People" that'd haunt your darkest dreams, and they tend to go far beyond generic attacks. El_C 14:04, 18 February 2008 (UTC)

• Anyway, please make sure to take out MCN's page from BCB, if you haven't already. Thanks. El_C 14:09, 18 February 2008 (UTC)

• it was removed after about 50 warnings. β^_command 14:34, 18 February 2008 (UTC)

• Betacommand, the MfD doesn't support your accusation that MickMacNee wrote an attack page. There is legitimate criticism there. And I am seriously concerned that you think it is OK to use a bot to make a "minor violation of POINT". You are a member of WP:BAG and need to be more responsible than that. Carcharoth (talk) 14:10, 18 February 2008 (UTC)

Users who make these attack seem to be immune to NPA and CIVIL when they talk to me, Ive been called everything in the book and a few more. what I did was a small demonstration to make a point.

I am highly concerned, along with Carcharoth, that Betacommand believes that it is acceptable to seriously abuse a bot's function and vandalise a user's talk page to prove a point instead of initiating civil discussion with the user. BCB placed approximately 50 tags, none of which had anything to do with MickMacNee, in a single second (although bizzarely some of the edits replaced the other tags), this behaviour is seriously unnacceptable and should be recorded and dealt with appropriately instead of dismissing the matter entirely. This discussion is also quite confusing so I apologise if I'm just reiterating what others have said, I just felt like having my say on this very serious matter. .:Alex:. 14:26, 18 February 2008 (UTC)

I don't think he does see it as acceptable: "yes I lost my cool", whilst falling short of a much-needed ringing apology, would seem to indicate that. Please can people start reapplying that ol' looking on the best side of people thing again, rather than finding further sticks to beat BC with? Wikipedia extends miles and miles and miles of AGF to every stinking sociopath, troll, sockpuppet and POV-pusher on the entire interweb, but suddenly the cupboard is bare for BC. This place can be very odd sometimes. ➔ REDVEЯS knows how Joan of Arc felt 14:37, 18 February 2008 (UTC)

What I find actually to be odd about this place is how many times we've seen BC lose his cool, fail to apologize, and then heard these defenses before, always phrased as something new, however. Like memory, selective AGF is a harsh and/or the sweetest mistress. El_C 14:46, 18 February 2008 (UTC)

I think we may have finally got to the point though where everyone agrees - one more stupid stunt and blocks will have to be imposed, as the gamut of asking, requesting, cajoling, pleading, and ordering have all now been tried to get BC and his bot to behave. Any more flagrant WP:POINT violations and the only remaining solution is a block (although no doubt Betacommand would soon get it overturned via IRC). Everything else has now been tried. I would probably have supported a block at this point (if a 24 hour block prevents further asshattery at a later date, it is preventative). Neıl ☎ 14:59, 18 February 2008 (UTC)

What's the saying about admins are given more leniency for their snippy behavior because of all the shit they take from people and all the work they do? Well, it's basically that. It's my opinion that Betacommand is one of the last people you'd really want to piss off, because if he leaves our image workload skyrockets. Read his talk page archives. It's completely ridiculous. If I had to take all the shit from people that he takes, I would have walked in, guns blazing, and killed everyone already... okay, just gone rogue, but you get the point. Oh noes, a point. Speaking of those, this was a fairly minor one and has not damaged the project. What basically boiled down to an attack page against Betacommand and his bot was created. BC spammed his page with a couple dozen notices and then went on his merry way. One click of the [rollback] and it's over. It's just more of the same. Everyone smacks BC around over his behavior, but he's constantly got a lynch mob after him. Look up. How many threads have been ignored or have practically no response? And look at this. Betacommand's name comes up on AN/I and people flock. Calling for his indefinite block is just... stupid. Let's be frank. It's stupid. That's somehow supposed to help the situation? Since when do we indef block editors seven hours and 2,000 edits later to wait for their apology for spamming a talk page in retaliation of relentless harassment. And anyone who wants to claim that it wasn't harassment, needs to go read that ridiculous page. Be glad that Betacommand has thick skin and all he needs to deal is to make the occasional harmless point. Considering the tone in which people approach him, there should not be an expectation of rainbows and butterflies from Betacommand. Lara❤Love 15:21, 18 February 2008 (UTC)

I think the argument is to block the bot, not the editor, as it could be seen as malfunctioning, and it is standard protocol to do so when bots make mistakes - my bot was blocked a few months ago because pywikipedia had not caught up to a fairly recent change to redirects. On the other hand, the bot flag, like any, is a position of trust (and in the case of BCB, a large one indeed), and people should be wary not to let it be lost in situations such as these. Will ^(talk) 15:31, 18 February 2008 (UTC)

I have an expectation that editors will not vandalize Wikipedia, and any suggestion that it's somehow acceptable or "understandable" for editors (particularly admins) to vandalize is completely insane. Nandesuka (talk) 15:30, 18 February 2008 (UTC)

To be clear, I'm calling for a 24 hour block of BC if he pulls any more stupid stunts (they can then escalate), not an indefinite block. The reason he gets so much crap is not because of the task he does, it's because of the way and the manner in which he carries it out. Neıl ☎ 15:38, 18 February 2008 (UTC)

I was the only one to mention an indefinite block, and that was only a temporary measure until an explanation emerged. It is common for admins to indefinite block and then stonewall and say "no shortening or unblocking until they show they know what they did wrong and say they won't do it again". I don't agree with that in general, but in some cases it is useful to be able to do that - to shock some sense into people if nothing else. Once Betacommand turned up and it was clear that the spamming had stopped, a warning was the logical next step, especially given the lack of an apology. Carcharoth (talk) 15:47, 18 February 2008 (UTC)

In summary

• Betacommand should not have got his bot to spam MickMacNee's talk page.
• Betacommand is aware that the community strongly disapproves of said spamming.
• The bot is needed to enforce violations of WP:NFCC
• Blocking the bot would be similarly WP:POINTy.

Therefore, my suggestion is that Betacommand be, and hereby is, given a really strong warning that while he's doing a good job, even the patience of those of us who happen to possess it in greater degree is not infinite, and it would be highly advisable to not engage in similar endeavours again. Stifle (talk) 15:00, 18 February 2008 (UTC)

Fully Support - With the addendum "Blocking BetaCommand at the point would fail the 'preventative, not punitive' mantra of blocks. ThuranX (talk) 15:13, 18 February 2008 (UTC)

Are we all agreed that the next time he vandalizes a page, he should be blocked? Nandesuka (talk) 15:19, 18 February 2008 (UTC)

I would say so. I think someone should place a notice on Betacommand's talk page. Template and stern warning hand or not, I don't care. Just get a clear warning in place so that he realises that a block will happen next time he "loses his cool". I would also strongly encourage efforts to develop a way to enforce NFCC without the need to use BetacommandBot. I've supported this bot and its operator in the past, but enough is enough. Carcharoth (talk) 15:24, 18 February 2008 (UTC)

• Support I am in full agreement of a "one strike" approach towards Betacommand. A stern warning should definitely be given. One more serious incident of this nature and blocks should be imposed, any further incidents involving Betacommand and his bot should prompt a review into the continued operation of BetacommandBot. .:Alex:. 15:32, 18 February 2008 (UTC)

• Betacommand warned as per the above - [20]. Neıl ☎ 15:56, 18 February 2008 (UTC)

Proposal: Betacommand stops running his bot

• Betacommand gets a ridiculous amount of criticism for the operation of his bot. 99% of the time it runs appropriately and follows the letter of policy, but that doesn't stop the abuse.

• The most recent and concerted campaign against Betacommand has been run by MickMacNee. I haven't counted, but there are more than a couple of new threads against BCBot and Betacommand started by MickMacNee in the last week - and there were already a half dozen active threads. Starting the discussion page was an attempt to bring his disruptive attacks and expressions of outrage to another forum, it was not to solicit constructive discussion.

• Betacommand made yet another failure of judgment in sending the bot notices to Mick's talkpage.

• Provocation can be a mitigating factor, but it does not dismiss the problem.

• No one person should have to shoulder the responsibility and abuse that running this bot has given Betacommand. The community should determine a better method for accomplishing this task, and select someone or a group of people to take it over.

• For his own good, Betacommand should stop running the bot until the above has been completed. He isn't singlehandedly responsible for compliance with policy - if we fail to move into compliance, then we fail and arrangements will be made to deal with the problem.

• In the future, the folks who take such joy in attacking BetaCommand for pointing out their own failures (particularly those who say they have stopped uploading images because of the 'attacking' response they get from BCBot, when you only get the tag notice if you have screwed up the uploading process through your own laziness...) should suggest solutions rather than attacking an imperfect one.

Avruch ^T 15:10, 18 February 2008 (UTC)

• NO Support. No one else has come forward to help out, and split the heaping, steaming, piles of abuse, so the only other option would be shutting down BCB, and giving MickMacNee a victory over policy and community support. This is a proposal to laud the villian and lynch the victim. hell no. ThuranX (talk) 15:15, 18 February 2008 (UTC)

  I'm sorry, but since when is someone who uses an extremely volatile bot to spam a good-faith user's talk page a victim? Will ^(talk) 15:21, 18 February 2008 (UTC)

• Support Betacommand not running the bot. Beta has been, on several times, rather uncivil, the seventeen-item box being part of that. Calling the RFC a "bullshit attack page" was bad, for instance. But the user talk spamming has caused me to lose all my trust in him. Allow some other established bot runner to take over the bot as the task does need to be done - a member of the BAG or pywiki team perhaps? I'd trust them enough to handle the delicate source code and fix errors in it. Will ^(talk) 15:21, 18 February 2008 (UTC)

  I had a nice discussion on #wikipedia-en with Beta about the bot, and did answer some questions quite openly and competently, and assured me that it wouldn't happen again. If he tries to keep his cool as best as he can (I don't think Mick was intending to attack Beta, just get some centralised discussion), then I'm happy for him running the bot. Still, releasing the source code to the BAG wouldn't hurt, as it can be helpful if they know exactly how the bot ticks to recommend on how to improve it so it acts less inflammatory. Will ^(talk) 23:52, 18 February 2008 (UTC)

• Hell no. We should sooner consider blocking MickMacNee, who is instigating this whole thing, than shutting down the bot that's doing most of the work in this area. Since when is it acceptable to go forum shopping to now at least 3 new areas against a single user, who is trying to help? (Those areas being here, ANI, and the 'betacommandbot and nfcc 10 c' subpage of WP:BOT.) You say that noone should have to 'shoulder this responsibility', and you suggest that the community take over. Has anyone come forward? The community is, in general, not sufficiently competent to do what betacommand does, but they're so quick to whine and so slow to volunteer assistance. If this bot was put in the hands of the community, it would be shut down for months until they decide what they want to do with it and then a watered down version would be running far too late and nothing would get done. The people who constantly disrupt the operation of this bot are the ones who should be being discussed. Instead administrators are permitted to run amok and block this bot for the smallest misstep, a misstep that other bots would receive a message on their owner's talk page about, and all because of MickMacNee and others causing such commotion. As a result of this disruption, the bot is under constant fire and BC is more concerned about protecting his reputation than working on new tasks. Why don't we leave the bot running to those who know what they're doing? --uǝʌǝsʎʇɹnoɟʇs(st47) 15:33, 18 February 2008 (UTC)

  As a matter of interest, how would you feel about the bot/sourcecode being transferred to the BAG/pywiki developers? As I've said, they can be trusted enough. If it's done quickly, most people won't see a change in its operation. Will ^(talk) 15:38, 18 February 2008 (UTC)

  Source code so recommendations can be made? Sure! Bot operation transferred? No! The best person to maintain a program as complex as I would guess betacommandbot is is always the one who wrote it, because they're usually the only one who knows what they were thinking when they wrote it. --uǝʌǝsʎʇɹnoɟʇs(st47) 16:01, 18 February 2008 (UTC)

  ST47, I would be happy to leave bot owners alone if they didn't, well, spam talk pages of editors they are in dispute with, maybe? Stop using the other issues to obscure the fact that what Betacommand did was completely unacceptable. Carcharoth (talk) 15:40, 18 February 2008 (UTC)

  So, betacommand demonstrating a point, perhaps somewhat inappropriately, but harmlessly, to a user who didn't understand the sheer magnitude of images that the bot works on (50 edits?) should result in all edits by this bot stopping? (sometimes well over 150000 edits a week, if you count both images tagged and warnings sent separately) I'd rather tolerate 50 edits than have 75000 more images to review. Of course, if you'd like to volunteer to work on the backlogs, I'm sure betacommand can get you a list of images that need to be fixed, but I don't think any human, or even group of humans, can do in a year what this bot can do in a week. --uǝʌǝsʎʇɹnoɟʇs(st47) 16:01, 18 February 2008 (UTC)

  Also, the harassment was a problem long before that particular incident. --uǝʌǝsʎʇɹnoɟʇs(st47) 16:05, 18 February 2008 (UTC)

  Carcharoth, you were on BC's case before he spammed MMN's page, no? Lara❤Love 16:14, 18 February 2008 (UTC)

  If you mean Wikipedia:Administrators' noticeboard/Archive126#Adding useless revisions to pages to make them undeletable, then yes. That incident was resolved as far as I was concerned at User talk:Betacommand/20080201#Bot unblocked. As far as I'm concerned, this is a new incident. Does that answer your question? Carcharoth (talk) 16:31, 18 February 2008 (UTC)

Support - This causes far too many issues. Just because we don't have a good alternative in someone else running a bot or multiple people to shoulder the workload doesn't mean we should continue with the current situation blindly. Its very obvious there is an issue here and we shouldn't continue to plow forward.--Crossmr (talk) 15:35, 18 February 2008 (UTC)

• Comment I would support this proposal if and when we find anyone willing and capable to write and run a new bot. Neıl ☎ 15:40, 18 February 2008 (UTC)
  • I would write a new bot, but I want to see BetacommandBot's source code in order to do it instead of reinventing every wheel. I have asked for this in several places, but Betacommand has ignored me. rspeer / ɹəədsɹ 01:01, 19 February 2008 (UTC)
• Comment I think that a group of users would be more effective for this sort of task, but that might be too complicated. I do not believe that simply stopping operation of BetacommandBot is a good idea, as this will cause massive backlogs and problems across Wikipedia. I cannot support this proposal if there is no alternative given. .:Alex:. 15:45, 18 February 2008 (UTC)
• Support Betacommand is consistently unhelpful and borderline incivil whenever any issues regarding his bot show up. It should be put in someone else's hands. Jtrainor (talk) 15:53, 18 February 2008 (UTC)

Do you seriously think that other people would deal with this constant harassment better? --uǝʌǝsʎʇɹnoɟʇs(st47) 16:13, 18 February 2008 (UTC)

• reluctant (believe it or not) support. The bot serves a necessary function, but both the bot operator and the bot would have been banned long ago if it weren't such a necessary function. I think we would be better off deleting all images (with a tagged list) and restore those for which rationales can be provided, than the bots arbitrary (by necessity) and incorrect (not by necessity) interpretation of NFCC. — Arthur Rubin | (talk) 16:19, 18 February 2008 (UTC)

I don't know too much about bots, so this might be a stupid question, but why isn't the code of the bot open source? Or rather, why can't we give the code to someone who, well, will not spam someone's talk page after being provoked? Maybe I've missed something and the code is open source, tho. If so, could someone else maybe do the job that the bot is doing now? There might be some obvious reason(s) against this, but (as I don't know a lot about bots) I don't see any right now. --Conti|✉ 16:01, 18 February 2008 (UTC)

Betacommand has not released the code, although he is (if I remember correctly) happy to release it on request to people he trusts. Neıl ☎ 16:03, 18 February 2008 (UTC)

That's probably something I should ask him directly, but.. why not? Is there, generally speaking, a reason not to publish the code of a bot (Apart from bots that deal with vandalism, I suppose)? --Conti|✉ 16:07, 18 February 2008 (UTC)

Well, I don't know his real reasons, but one could be the same as vandalism bots: If everyone knew the bot's detection rules, then anyone could evade them by putting in only what the bot looks for without following the complete policy. --uǝʌǝsʎʇɹnoɟʇs(st47) 16:15, 18 February 2008 (UTC)

There is an element of WP:BEANS, I believe, yes. Neıl ☎ 16:17, 18 February 2008 (UTC)

Aren't the bot's detection rules already known, tho? As far as I know, the bot checks if there's a link to the article the image is used in. It should be easy to fool the bot even without knowing its source code, anyways, since a bot can't decide whether a fair use rationale is vaild or not. We still need humans for that, so that doesn't seem like a very strong reason to me. Of course, there could be other reasons to keep the source code closed. I just think we should use open source code wherever possible. We are the free encyclopedia, after all. --Conti|✉ 16:23, 18 February 2008 (UTC)

• Oppose - I doubt there are many other people capable of running this bot and taking all this crap that would be able or willing to deal with it without snapping now and then. Everyone has their limits and the fact that people don't see how far BC gets pushed on a regular basis tells me they aren't looking hard enough. Lara❤Love 16:14, 18 February 2008 (UTC)

• Oh good grief hell no: If the Bot is shut down, there will be a gaping hole. If you get someone else to run the bot, the abuse will be just as bad. The problem isn't the bot. The problem is the people unwilling to abide by policy and mad enough to spit nails when they get called on it. --Hammersoft (talk) 16:40, 18 February 2008 (UTC)
  • Comment. I can't let Hammersoft's comment go unanswered. The problem I and many other users have is when we uploaded our images AT THE TIME WE UPLOADED THEM, the various fair use tags, rationale text, etc etc was perfectly fine and acceptable. At some point the powers that be decided that some sort of template needed to be included with every image, or that certain T's had to be crossed and I's dotted. And so those of us who spent hours uploading perfectly allowable images are being asked to reinvent the wheel on hundreds, sometimes thousands of images? That's the issue for a lot of us. Yes, there are those who didn't do it properly but there are plenty of good faith users who are being penalized by a bot that needs to be fixed (not eliminated, fixed - I have been very vocal about the bot but I don't think I've ever said it should be deleted). If you want an analogy, consider this: imagine that the speed limit in a city is 60 MPH. One day the city council decides to drop the speed limit to 30 MPH. And they hire somebody to send retroactive speeding tickets to everyone who was driving 60 MPH before the law changed. That's exactly what I'm seeing happen here. 23skidoo (talk) 16:58, 18 February 2008 (UTC)

I'd like to note that there appear to be quite a number of non-compliant images uploaded by you after September 2005, when the policy on reviewing each use (earlier form of 10c) came into play. For example, two images from March of 2007 [21][22]. I also note that you later edit warred BetacommandBot on those images [23][24] against policy. I'm not terribly sympathetic to people complaining about this bot when they won't adhere to policy themselves. --Hammersoft (talk) 17:59, 18 February 2008 (UTC)

• • • No, the city council is asking you to go to your local garage to install the new safety device they are telling you about for almost a year. You are not forced to do it, but you won't be able to continue to use your car if you don't. -- lucasbfr ^talk 17:35, 18 February 2008 (UTC)
      • When we say that there needs to be a more gradual and lenient transition for images that were perfectly copyright-tagged when they were uploaded, Betacommand and Hammersoft paint it as if we're supporting evil copyright violators. The fact that the images were right when they were uploaded is relevant, even though it's also true that we need to add rationales now. rspeer / ɹəədsɹ 06:00, 19 February 2008 (UTC)
        • The policy requiring 10c has been in place for 2.5 years. The Foundation's resolution is cognizant of the fact that things don't change quickly on this scale, and allowed for a deadline a year in the future. Yet, you and others want MORE time to fix the problems???? How much is enough? Ten years? Twenty? --Hammersoft (talk) 13:58, 19 February 2008 (UTC)
        • Actually, it's even older than 2.5 years. It goes back at least 3.5 years. Quoting [25], "each "fair use" must be explained and a rationale must be established for that specific use". --Hammersoft (talk) 16:53, 19 February 2008 (UTC)
• Give me a break. People can be so childish when it comes to nonfree images. Here we have a helpful bot that does a lot of good work for wiki, and since people don't want to put in the ten seconds necessary to write a stupid rationale we're considering shutting it down?? Think how much more work this whole process would take without the bot. There are other alternatives that could be considered (e.g. allowing templated rationales for certain classes of uses, which has been considered before and found a lot of support), but shutting off BCB is utter silliness. Stop being so lazy! No one thought this process was going to be easy, and the minor disruption this is causing only illustrates what sad shape policy compliance is in right now. BCB does a good job, and I applaud BC for continuing his work in the face of all this unwarranted criticism. Calliopejen1 (talk) 23:56, 18 February 2008 (UTC)
• Support shutting down the bot. Betacommand has shown that he can't be trusted with this kind of power, and that he isn't interested in making his bot less disruptive. Threats of impending copyright doom don't change this. rspeer / ɹəədsɹ 01:05, 19 February 2008 (UTC)

• Support bot shutdown. Betacommand is not a bad person, and I'm quite certain he means to improve the project as a whole. That said, this has become a major point of contention, and quite likely the head of unfree content hysteria that's gripping Wikipedia right now. It does not reflect well on the project to see this type of miserable infighting occurring - a mountain's been made of a molehill, to quote an old cliche, and this particular item isn't helping make it any smaller. I have a few ideas which might help reduce the level of nonsense occurring in both sides of the content dispute, but this isn't the proper forum to express them in. MalikCarr (talk) 04:12, 19 February 2008 (UTC)

• Support proposal for Betacommand to stop running his bot, at least temporarily. Enigma ^msg! 04:24, 19 February 2008 (UTC)

• Support the bot is inflammatory, bitey, and just plain rude. The User doesn't respond civilly to criticism and doesn't respond at all to anything else. I understand that this is an important bot but if humans can't check all of these images in the time before this deadline, Humans can't be expected to fix the problem before the deadline. I don't care how much abuse BC takes for attemting to enforce a rule which is, in general, not followed. He signed up to take the abuse by running the bot, several others have offered to take over, and yet he still hides behind the excuse that people's complaints aren't legitimate. Beta and his bot should not be above the rules that govern the rest of us. I ask, if any other bot caused this much trouble would anyone hesitate to block it? If any other user made people this angry would you hesitate to reprimand or block them? I don't think so. Adam McCormick (talk) 05:03, 19 February 2008 (UTC)

• Support - How has this guy not been blocked? You don't let cops get away with beating a few random people up to relieve stress just because most of the time they're doing something you want them to do. —Torc. ^(Talk.) 01:42, 20 February 2008 (UTC)

• Given the comments and tone in the paragraph with big red font below, I ask again: How has this guy not been blocked? Start treating somebody like they're untouchable and they're going to act like it. —Torc. ^(Talk.) 04:51, 20 February 2008 (UTC)

• Strong Oppose - This is completely the wrong timing, with a month and a few days before we need to be compliant. If someone had a bot right now ready to go and tested as a replacement, I would likely support that, but there's a vacuum without it, and I doubt within a week one could have a fully tested bot functioning to do the same tasks. The bot is trying to save images by bring them to attention of users, and doing it now before the Foundation requires us to take a stricter approach; without a bot, more images would be deleted when we have to implement the stricter requirements of the Foundation after March 23. After we pass March 23, we can talk about getting a new bot in place, but right now I see no viable replacement that is needed for a time critical task. (Yes, we still do need a bot after March 23 to keep compliance). (Personally, I would rather have BCB get the rest of his runs done in as fast as he can, so that there's a huge outpouring of anger over a short amount of time (and longer grace periods to fix) than some outpouring over a longer period of time, as the former "gets it out of the system" faster.) --MASEM 01:53, 20 February 2008 (UTC)

Comment re this whole thread

This really, really should not have been moved to a subpage after only a few hours. Particularly as whoever moved it here has placed it in the wrong order, above two older sections. Neıl ☎ 15:29, 18 February 2008 (UTC)

Yeah I didn't get the order right. I'd play with it to fix it, but it looks like someone has. There were discussions in multiple threads on both WP:AN and AN/I. That is not helpful towards reaching a conclusion or having a worthwhile discussion, centralizing the discussion is. Avruch ^T 15:32, 18 February 2008 (UTC)

Yes, I fixed it. The issues are seperate and centralizing them isn't relevant - one is a technical discussion of how the bot should work, the other (very recent thread) is examining Betacommand's behaviour, is less than a few hours old, and should have remained on the main AN page. I won't edit war over it, but I'd suggest we need to have some kind of loose guidelines for when and when not threads should be excised to subpages. Neıl ☎ 15:35, 18 February 2008 (UTC)

If folks want to move it back, I don't mind. It seemed like they were all essentially related discussions that would helpfully move forward in the same place. If folks disagree, revert away (although, the sooner the better here). Avruch ^T 15:40, 18 February 2008 (UTC)

It could be moved back now, but I agree, sooner the better. At the moment, the ANI subsection has a link directly to the start of the thread, so that is OK for now. Carcharoth (talk) 15:42, 18 February 2008 (UTC)

Ironically, one of the reasons for moving was the size of AN and ANI. This page is 166kB and larger than AN. It is also over half the size of ANI. Carcharoth (talk) 15:51, 18 February 2008 (UTC)

Sub-subpages? Neıl ☎ 15:58, 18 February 2008 (UTC)

Thats part of the reason why I moved it - its a huge mess, and partly because everything is getting repeated in different threads. Unless someone is going to switch it back, I'll move the remaining thread Carcharoth pointed out. Avruch ^T 16:04, 18 February 2008 (UTC)

Non-notification

Regarding the Feb 13-14 spree of around 20,000 images tagged for deletion I have attempted to initiate conversation with betacommand over possible glitches in his bot. It seems that it often fails to notify either the image uploader or the article talkpage. All those images will be up for deletion tomorrow and I'm certain that some "helpful" admins will burn through the lot in no time. I think that it's out of order that images can be deleted in this manner and I attempted to discuss the issue with Betacommand here but with this edit he moved the conversation to an obscure location on his talkpage where his responses were "that [MacNee] page is full of lies and bullshit and is a complete farce and it will not affect how I operate BCBot" and "as for the other images uploaded by English peasant, that was caused by a user re-name while the bot was running" and that he excempted Blofeld of SPECTRE from receiving bot warnings (long story) I don't feel that these responses adaquately address my concerns. If there are glitches preventing users from being notified of iminent deletion the issue should be seriously considered rather than quickly dismissed.

I'm also concerned about the state he leaves inactive users talkpages [26] 532,500k and growing, and the vast majority of the messages concern images that are only lacking a backlink to the article. My (admittedly sh*t) PC can barely open the page to go through and fix the missing backlinks. English peasant 13:48, 18 February 2008 (UTC)

Without commenting on any of your other points, there's been an tacit agreement in place among the admins who regularly handle image deletions of this nature to just postpone this specific backlog indefinitely; it's around three weeks behind now. Also, the exemptions from notices are done only upon request, one situation that jumps to mind is the exemption of Melesse, who does lots of work in reducing non-free images and shouldn't be bothered for images she didn't upload. east.718 at 13:52, February 18, 2008

(to English peasant; ec with above) A very pejorative way of stating the first issue. The "obscure location" was to his personal talk page rather than the bot's talk page - more useful as it causes him the orange bar. Not at all unreasonable. His reply to you also seems adequate: "I found the issue, it was a database lock that caused the error". These things happen to every bot (to humans too - I hit one a few minutes ago when User:Nick deleted Man). I'm not seeing an issue (other than people trying to find more sticks to beat BC with). Also, coupling the first complaint - not enough warnings - with the second - too many warnings - highlights the hell that editors put BC through. No way he could win that one, is there? ➔ REDVEЯS knows how Joan of Arc felt 13:55, 18 February 2008 (UTC)

User:Nick, not User:Neil(!) - I've fixed it :) Neıl ☎ 14:52, 18 February 2008 (UTC)

He's lucky to have you, I'll say that much. El_C 13:56, 18 February 2008 (UTC)

(4EC!!!)This part of this thread's is somewhat misleading. Betacommand did NOT hide the comment, he brought it over to his talk page, his edit summary at BCB's talk makes that clear. it wasn't hard for me to find it on his talk page, I checked BC's edit history, and BC's VERY next edit was to paste it onto his talk page and note that EP should read an above section. He did NOT respond with the 'bullshit and lies' comment there. As for his explanations, that a user changed names and BCB couldn't find the user seems reasonable to me, that Blofeld and he have an accord to not notify Blofeld is between them (This only effect on others is that Blofeld's images stand a slightly higher risk of deletion because the uploader doesn't know; articles will still be tagged that the image is endangered). This section can only add fuel to the fire, and is thoroughly unnecessary. —Preceding unsigned comment added by ThuranX (talk • contribs)

(ThuranX) Just to clarify, Blofeld was unaware of his "accord" with Betacommand. I believe that I am raising a legitimate issue, I'm not at all bothered about the spat between BC and his detractors, I'm concerned about the retention of images on Wikipedia, and the effect of non-notification, and yes I get the message, sod off and quit complaining. English peasant 14:38, 18 February 2008 (UTC)

No, more "please don't use wikidrama as an excuse for creating more wikidrama". Remember when we come to press 'Save page', we must always say to ourselves, "am I pressing this to improve Wikipedia, or just to provoke a response?" The 'Show preview' button is also useful for giving us the opportunity to look at what we've written and ask, "Am I presenting this fairly and accurately, or am I just hoping to get what I want however I get it?". ➔ REDVEЯS knows how Joan of Arc felt 14:44, 18 February 2008 (UTC)

I saw nothing anywhere that indicated Blofeld didn't know. It may be that it was handled in Email, or by community consensus. You'd have to ask for the 'long story'. I took it at face value, so shame on me for AGF. I guess you'll ask for me to be blocked next? Redvers has it right. Stop the fucking drama. go write a page. ThuranX (talk) 14:51, 18 February 2008 (UTC)

Excuse me, I have never said anyone should be blocked. I have written over 1,000 pages but my time is being used protecting images at the moment and I am not raising the issue for dramatic effect. I'm just concerned that there may be a glitch in the bots programming and I get shouted down. English peasant 15:32, 18 February 2008 (UTC)

no sprees or deadlines, please

I don't have time to follow the whole debate, but I'd like to add my opinion that the "problem" of non-free images is not so serious that it requires automation at this level, and not so important that it requires deadlines (let alone ones measured in days). Let's not destroy the encyclopedia in order to save it. —Steve Summit (talk) 16:41, 18 February 2008 (UTC)

• The Foundation disagrees with you, as they have set a deadline. --Hammersoft (talk) 16:46, 18 February 2008 (UTC)
  • And this whole issue has distracted from my question about what happens after that deadline. If you really, really want to avoid drama when that deadline arrives, please go to the thread at Wikipedia:Administrators' noticeboard#Deadline does exist (spirit of resolution) and say what you think should happen after the deadline passes. No-one has bothered to try and answer my bolded question there: Is it possible to determine which images should and should not be deleted after 23 March 2008? Well? Is it possible? Carcharoth (talk) 16:54, 18 February 2008 (UTC)
  • Apparently the deadline for en: actually passed 11 months ago, on 23 March 2007. And yet the Wiki hasn't been turned into a smoking crater by copyright lawyers. Imagine that. This, to me, is a sign that we can take our time, set realistic goals, and not cause unnecessary strife in an attempt to complete an uncompletable task by a certain date. rspeer / ɹəədsɹ 06:03, 19 February 2008 (UTC)
    • If you look at the link I provided above ("Deadline does exist"), you will see that the 23 March 2008 deadline was probably intended in spirit, if not wording, to apply to projects with an EDP as well (like en-Wikipedia). I would support the bot running up until the deadline, and then being retired with grateful thanks from those portions of the community that recognise the vital job it has done. I'm hoping that the community will step forward and help get alternative processes in place to deal with the 500 or so non-free images uploaded every day, or even better, to thoroughly patrol the 3000 images uploaded every day. I'm guessing we patrol more new pages than that every day? Carcharoth (talk) 10:05, 19 February 2008 (UTC)
      • All images still need to be patrolled. An old image could be added to a second page without a rationale added to the image, which creates a problem; ideally, of course, the number of images tagged will be much much lower post March 23, but it still must be done, and it can't be done by humans. This is not to say it has to be BCB but it does need to be a bot, and at that point, you might as well have that bot also patrol new image uploads for the same purpose. --MASEM 14:02, 19 February 2008 (UTC)
        • New uses of old images could easily be detected by bot. The trouble is that if bots are used to notify, people will just copy the existing rationale and avoid human review that will pick up excessive fair use. Ideally, there would be a way to require an actual discussion and approval for such new uses, and a box can be ticked in a template to link to the discussion and to exempt such cases from the next bot run. All rather complicated, but might be needed. Carcharoth (talk) 16:21, 19 February 2008 (UTC)
          • I'm going to continue some ideas on this into Wikipedia talk:Non-free content criteria compliance, since this is getting away from BCB directly, but we do need to think about how we handle post-March 23 enforcement. --MASEM 16:31, 19 February 2008 (UTC)

Drama

I wonder whether anyone has considered that just directing a bot to go through every image and add links to the associated articles would effectively cancel out all of these issues? Adam McCormick (talk) 00:35, 19 February 2008 (UTC)

That misses the point. If the image use was wrong, it needs to be removed, not just linked. Humans need to review each non-free image use and make sure it is OK. Bots can't do that. The drama here comes from a variety of sources, but mainly people (from both sides) jumping in and making only a few points without really looking at the whole picture. To really get a handle on what is going on here, it is best to read lots of archived discussions, hang out at various noticeboards for a few weeks, try fixing a few images and uploading a few images yourself, and then slowly you will get an idea what the common types of image problems are, and the volumes of various groups of images. Carcharoth (talk) 00:58, 19 February 2008 (UTC)

I've been reading these discussions (and similar) for nearly a year now, but it all seems to come down to "Images that don't follow the letter of the policy, as we see it, they must be deleted" versus "Not everything this bot does is accurate, and can it change" and "lots of these rationales are withing two word of perfect" with some random other comments in between. Both sides see the other position as utterly untenable and around and around we go. This just reeks of one stubborn editor who refuses to see reason and a groundswell of those who feel downtrodden about this policy. It all seems very bitey and pointed. I mean the code to this bot sound extremely simple (could probably be replicated in a few hours testing) and the idea makes sense it just seems that the execution is flawed. Why won't Beta allow community consensus to control his bot? Why is this policy still in place if there is a groundswell of opposition to it? I wonder. Is this really that important of a bot, and if so why is it being run outside of the Admin Community? It just seems like everyone, including Beta and Mac, needs to chill out and just discuss things instead of inciting such a riot. Adam McCormick (talk) 01:10, 19 February 2008 (UTC)

• The bot could add the link if the picture is only used on one page and has correctly indicated a license. Assuming that a non-documented use of the image means the image is being misused is assuming bad faith - we wouldn't let human beings get away with doing that, so we're going to let a machine? —Torc. ^(Talk.) 20:23, 19 February 2008 (UTC)

• the bot assumes nothing. bots cannot write rationales. and bots cannot assume good faith, neither can they assume bad faith. images that are non-compliant need reviewed and fixed. just adding a link goes against policy. β^_command 20:28, 19 February 2008 (UTC)

• The bot inherits its assumptions from its programming. This bot was created with a faulty, bad faith assumption: anything that doesn't meet its requirements has not met Wikipedia's requirements. Plenty of examples of this assumption being false have already been provided. The bot is faulty and should be taken offline. —Torc. ^(Talk.) 20:42, 19 February 2008 (UTC)

• Torc2 you are mistaken. Policy clearly states otherwise. bots cannot add proper rationales. if you think they can you need to review our non-free policy. β^_command 20:50, 19 February 2008 (UTC)

• The faulty assumption is that the rules that apply WP:NFCC must be written in a way that BCB understands. That is false; the resolution only calls for it to be "machine readable" - not "readable by a poorly-programmed bot". When it reads an image fails to see what it is expecting in a format it can comprehend, it assumes (in bad faith) that the editor is to blame. Also, given this proposal, BCB activity should cease until that is settled. The {{Non-free album cover}} or {{Non-free DVD cover}} tags (for example) for images used on a single page should be sufficient indication of fair use. —Torc. ^(Talk.) 21:04, 19 February 2008 (UTC)

• question Torc2, since you obviously did not bother to read WP:NFCC let me quote it. The name of each article (a link to the articles is recommended as well) in which fair use is claimed for the item, and a separate fair-use rationale for each use of the item,

whats so hard to understand about that? that is POLICY instead of making bullshit comments based on things you pulled out your ass and are not based on fact, why not stop making unfounded attacks and BACK IT UP WITH FACTS AND DIFFS which you cannot do. all BCBot does is follow policy if you write a rationale that meets policy BCBot will not tag it. β^_command 02:55, 20 February 2008 (UTC)

I read it. I saw nothing that said the rationale had to be readable by BCB, or what the bot would be looking for. And since when does "recommended" mean "required"? That's a faulty assumption. Even WP:FURG doesn't require the use of a template. The bot needs to be intelligent enough to work around reasonable interpretation of the requirements; editors shouldn't be expected to guess how the bot will read the image page. —Torc. ^(Talk.) 05:05, 20 February 2008 (UTC)

Torc2, let's draw this out. The policy requires rationales for each use (the link to the title, as opposed to just typing out the title, is recommended). BCB looks for those typed out article titles (link optional) that are required by policy. If your rational doesn't meet policy, then BCB tags it. If your rational does meet policy, BCB doesn't tag it. What part of that don't you get? Lara❤Love 06:50, 20 February 2008 (UTC)

I don't get the part where the bot is too poorly programmed to recognize simple misspellings. I also don't buy that the algorithms and coding are robust enough to spot valid rationales that are not formatted in a way that the bot is expecting, or that the bot can't be easily fooled. Of course, the bot is a big black hole, and by keeping the code secret, BC has ensured that the flaws won't be found, and that the bot won't undergo any serious QA testing. Does the bot recognize FURs without links? I don't know, because nothing is published about how the bot works.

It's easy for Beta to hang himself on the cross and yell "bullshit!" and "lies!" at everybody, because it's easier to hurl insults than it is to be helpful and discuss potential flaws in his code and the tone of the messages he programmed. I also don't get how one editor is given cart blanche to run roughshod over whatever civility guidelines we pretend applies equally.

Just think how much easier this whole process would have been if the instructions were remotely informative, if the warning message the bot gave encouraged users to fix their images and pointed them to clear, helpful instructions rather than simply berating them for not having it right, if User_talk:BetacommandBot had actually helped editors solve their problem instead of defensively belittling them for being annoyed by an obnoxious message, and if BC and his legion of apologists actually acknowledged that positively contributing editors might have a justifiable reason to be annoyed by the way the bot was instituted and run. —Torc. ^(Talk.) 10:09, 20 February 2008 (UTC)

Torc2, Betacommand and Lara are right. We can't assume that because an image is in use that it is in use in the right article or that the use is valid. That needs to be reviewed by a human. Think of it as image patrol, the same way as new pages are patrolled, with the exception that some of the images are not new and that deletion is proposed. The equivalent would be if a bot put PROD notices on all new pages that failed some agreed upon text. That is what BetacommmandBot is doing to non-free images. Carcharoth (talk) 08:56, 20 February 2008 (UTC)

Yet we can assume that if an image doesn't meet whatever specific, yet insufficiently-documented conditions BCB is looking for, that it is a copyright violation? That doesn't seem like an assumption of good faith to me. Rather than do the analogous of PRODding, why don't we do the analogous of {{references}} tagging or {{notability}} tagging? Why did BCB go directly to a deletion warning? Why didn't it drop an warning with a polite tone stating that the image might lack a sufficient rationale and needs to be fixed soon, maybe with links on how to fix the rationale and why this is important? Then why weren't these images reviewed by humans to confirm BCB's suspicions before the deletion warning was dropped on them? Where is the discussion where BC asks how he can make this whole thing friendlier to editors and less confrontational and actually listens to the responses? The warning BC received for attacking another editor with a bot was more gentle than the warning a couple thousand users received for posting images that don't fit the obtuse, mechanical requirements of an obscure guideline.

I don't buy the argument that a bot cannot write a fair use rationale for, say, an album cover that already has a license. If the person who coded BCB cannot do that, there's no reason to have any faith in BCB. If there's a legal issue of saying a human has to confirm the information, the bot could do the formatting work and prefill the information, add the image to a list, and a human editor could verify it later. There are better solutions than what we have. Another better solution would just be to append the rationale to the appropriate non-free use templates, so users fill out both in one go. The licenses and their relation to fair use are another source of confusion: why do we need these two separate templates that deal with essentially the same thing?

So, are all these tagged articles going to be reviewed by human beings before deletion? Are they going to fix the simple mistakes? Are images that are used multiple times and have one sufficient rationale but not another going to be deleted, or just removed from the unauthorized article page? —Torc. ^(Talk.) 10:13, 20 February 2008 (UTC)

overwrought comment removed (please, Betacommand, calm down)

Torc, I've removed Betacommand's comment, but if you look past his tone and language, he is actually right. For multiple uses we do require a separate rationale for each use - there are good reasons for this. "Are images that are used multiple times and have one sufficient rationale but not another going to be deleted, or just removed from the unauthorized article page?" - Betacommand has answered this elsewhere - if there is one valid rationale, the image will be left alone no matter how many times it is used without a rationale in other articles. There are plans to remove "images with rationales" from articles where they don't have a rationale for that article, but that will be later and there should be notification about this. "are all these tagged articles going to be reviewed by human beings before deletion?" - ideally, yes, though if the backlog sits there unfixed for a long time, eventually they will be deleted. Work has been ongoing on this issue throughout the year, from March 2007. There is a deadline in March 2008 that the whole year's work has been aimed at meeting. Have a look at Category:Disputed non-free images as of 24 January 2008 (currently 1899 images) - the deadline there has been extended to 24 February, though it has not been possible to indicate this on the image page, user talk page and talk page notice tags, which, in hindsight, is unfortunate. Similarly, Category:Disputed non-free images as of 12 February 2008 and Category:Disputed non-free images as of 13 February 2008 and Category:Disputed non-free images as of 14 February 2008 (a total of around 13,000 images) have a deadline of 2nd March. There used to be a lot more images in those categories. The numbers are going down in part because people are adding rationales, probably using tools like WP:FURME. Compared to the total number of images on Wikipedia (nearly 400,000, of which possibly a quarter are non-free, though that is a wild estimate), that is not a lot. The proportion of free to non-free has varied over the last few years, and is complicated by many free images being removed to Commons, but this gives some idea of the numbers involved. Wikipedia is a big place, and this whole project has been about more than just small groups of images here and there, it has been about bringing the entire set of images into better compliance with Wikipedia's non-free content policy and the image policy of the Wikimedia Foundation. The process has not been smooth at all times, and could have been prepared and implemented better, but it could also have been worse. Please read and re-read what I've written before replying, and see whether your questions have been answered. Thanks. Carcharoth (talk) 12:55, 20 February 2008 (UTC)

The fact that you had to act as handler for BC kind of goes to show what the problem is. I saw his reply, so I'll ask again: How has he not been disciplined for this behavior? Why is it continually excused? You see nothing wrong with that? Is there anybody else on this whole site who gets this kind of protection from their own actions?

I'm not really asking the questions for myself. I have no images at risk and was largely unaffected by this whole process. (I had two images that were tagged, instantly fixed once I deciphered that the the big template that said "It is believed that the use of low-resolution images of such covers solely to illustrate the audio recording in question, on the English-language Wikipedia, hosted on servers in the United States by the non-profit Wikimedia Foundation qualifies as fair use under United States copyright law" wasn't actually the fair use template the machine wanted, deleted anyway, and restored by request without any further changes required). I'm asking about where this process is documented for editors who don't read this page, or get turned off by the whole talk to the hand approach? Thank you, by the way, for being the one person who seems interested in discussing this sanely. —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

Addressing Torc's concerns individually

Torc, let me address your concerns one at a time.

OK, let me respond where I have time to do so. I've interspersed the replies since they wouldn't make much sense lumped at the bottom... —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• I don't get the part where the bot is too poorly programmed to recognize simple misspellings.

  The bot shouldn't have to be programmed to pick up spelling errors. It's the responsibility of the uploader to get it right. This is just more ridiculous demands on Betacommand.

Even my cell phone will correct a misspelling. I find it hard to believe that a bot wouldn't be able to scan an image description, and if the image is only used in one article, detect a name that is similar and put that on a list of potential misspellings, then send out a more appropriate message that alerts editors of the potential error, or that simply corrects it. There is also the issue described above of moved articles whose previous names are used for disambiguation —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• I also don't buy that the algorithms and coding are robust enough to spot valid rationales that are not formatted in a way that the bot is expecting.

  The bot expects the title of the article the image is used in to be present. This has been repeated to you countless times above. The article title, that's it.

• That's not what I said there. I said I have my doubts about it because the code isn't made public. Telling people what the intent of how the bot works isn't the same as letting people see what it's doing to accomplish that. —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• Does the bot recognize FURs without links?

  This question has been answered in big black bold letters for you already, more than once. How can you possibly not get it? This is blatant disruption.

I'm not asking for me. I'm asking for editors who are unaware of of this discussion. Where is this documented? Where does the average user go to find that information out, and whatever other questions they have about the functionality of the bot itself? And thanks again for your assumption of good faith. —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• It's easy for Beta to hang himself on the cross and yell "bullshit!" and "lies!" at everybody, because it's easier to hurl insults than it is to be helpful and discuss potential flaws in his code and the tone of the messages he programmed.

  He's hurling insults because he feels attacked, which I happen to agree with considering the sheer stupidity of some of the claims being made. These messages, for example, that you and others are bitching about are not messages written by Betacommand. Those are the standard image templates that were written by others. I went to go soften them a bit, but there's absolutely nothing wrong with them. They are clear, to the point, and polite.

• Read this. That's the introduction I, and most editors get to the work BCB is doing: defensiveness, insistence of absolute infallability, a totally dismissive tone, and 21 pages of archives editors are expected read before speaking. Everything on that page was said to other editors before they said a word to BC. Compare that to even the edit summary for ClueBot: "Reverting possible vandalism by [...] to version by [...]. False positive? report it. Thanks, User:ClueBot." Perfect. "Hey, I think this is vandalism, but just in case I'm wrong, let me know." Is it just a coincidence that this bot induces anxiety and combativeness instantly while ClueBot doesn't? —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• I also don't get how one editor is given cart blanche to run roughshod over whatever civility guidelines we pretend applies equally.

  I wouldn't consider it cart blanche. He takes a lot of shit from people, most of it unwarranted. Sure, he's an ass sometimes, but most people probably would be after two years of this bullshit.

  • I've been involved in this for a week or so, and actually, it didn't have any huge effect on me. It's not personal for me. That's why I'm telling you, from a fresh perspective, the editor relations side of this whole operation is horrendous. If he (or whoever is responsible for this) is going to set up a system that is this aggressive and this defensive, he should be willing to accept the criticism or he should not be doing this work. Think about what he did: he used a machine in which the community trusted him to operate to coerce and abuse another editor who was critical of its use. His punishment for this was nothing. People were so afraid of his reaction, many were hesitant to even require him to apologize. So what happens after that? He comes right back and abuses another editor, and somebody else steps in and protects him from himself. If he cannot deal with the criticism and defensiveness this generates, he should not be doing this. —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• Just think how much easier this whole process would have been if the instructions were remotely informative, if the warning message the bot gave encouraged users to fix their images and pointed them to clear, helpful instructions rather than simply berating them for not having it right.

  {{Di-disputed fair use rationale-notice}} reads:

---

Thanks for uploading IMAGE NAME. However, there is a concern that the rationale you have provided for using this image under "fair use" may be invalid. Please read the instructions at Wikipedia:Non-free content carefully, then go to the image description page and clarify why you think the image qualifies for fair use. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to ensure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page. If it is determined that the image does not qualify under fair use, it may be deleted within a couple of days according to our criteria for speedy deletion. If you have any questions please ask them at the media copyright questions page.

---

That clearly explains the problem, where to read about why it's a problem, where to go to fix it and an explanation of what will happen if it's not fixed. I tried to fix it, but I came up with nothing. Well, I changed that "may" from "will" so you nor anyone else would have room to call BCB a liar if the admins couldn't get to it in two days.

{{Missing rationale2}} reads:

---

[[:{{{1}}}]] is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to [[:{{{1}}}|the image description page]] and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to ensure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images lacking such an explanation can be deleted one week after being tagged, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

---

Look at all those links. Links to what everything means, various areas to go for help, a tempalte to use. I mean, it's like Missing Rationales for Dummies. I should check to see if it was written by Hungry Minds.

First off, you forgot Template:Di-disputed fair use rationale, which is what actually shows up on the image page and includes none of those links; it's the first thing most editors are going to see. Here's the problem with that: most images are used on only one page. It's entirely reasonable to expect that editors will believe something like Template:Non-free_album_cover is sufficient fair use for a single article, and there's really no logical reason to believe it's not. From an outsider perspective, for somebody who hasn't been dealing with this for months and months, for an image only used in one article, Template:Non-free_album_cover is the reason it's used in this article. Yes, the warning links to both the boilerplate and the fair use templates, so users can eventually figure out that there's two distinct components that say largely the same thing required. It's workable, but it does cause some frustration - the frustration is normal and expected, but people involved in the tagging and deletion side of this process sure seem to act like it's not and have no sympathy for editors who get frustrated with it.

The formatting of these messages is a minor part of the issue, although it doesn't seem like the psychological effect of seeing three or four of them on a talk page, or ten or twelve show up in a single day was consdered. The tone issues are things that seem trivial, but will be magnified when issued 10,000 times to 3,000 people. For example, "remember that you must complete the template. Do not simply insert a blank template on an image page." is rather condescending. Substituting "Adding and completing" for "Using" in that would make the last two lines unnecessary. Use of the word "you" is confrontational: "Please read the instructions at Wikipedia:Non-free content carefully, then go to the image description page and clarify why you think the image qualifies for fair use" can be less confrontationally stated "Due to copyright restrictions, the image description page must include a reason why the image qualifies for fair use; instructions for how to do this are available at Wikipedia:Non-free content." That at least makes the instructions sound less abrassive.

And yes, while BC doesn't control these templates, from the perspective of the editors receiving them, he is the one responsible for them being issued. And BCB is responsible for the "Image is about to be deleted WP:NONFREE" edit summary inserted into talk pages, which gives the impression that a decision has already been made to delete the image rather than indicating something just needs to be fixed. —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• Yet we can assume that if an image doesn't meet whatever specific, yet insufficiently-documented conditions BCB is looking for...

  Insufficiently-documented? Oh, I guess you forgot to read the policy that's been beaten into this conversation and like a dozen others this week. THAT is where it is CLEARLY written out what the bot is looking for. Also, again, in big black letters, repeatedly, on this page.

• Again, what I'm asking is where does the average user, who hasn't been involved in this discussion, easily find out exactly the conditions the bot is looking for? I want somebody who is going to complain that their image was correctly tagged, who isn't sure who to inform after fixing the images, or who just has a general question about the procedure to be able to go to a page that says essentially "here are some common mistakes that are made: the license tag is different from the fair use rationale; the article has to be spelled precisely in the rationale; after the rationale has been added, the warning messages can be deleted; if you have any questions, feel free to ask them on this talk page specially dedicated to this specific process (not one for the general guideline on which it's based)" Yes, this information can be divined from reading this page, or plowing through a few dozen archived talk pages, but if you're interested in making the whole process less irritating for the user, you're going to keep them better informed any way you know how.—Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• ...why don't we do the analogous of {{references}} tagging or {{notability}} tagging? Why did BCB go directly to a deletion warning? Why didn't it drop an warning with a polite tone stating that the image might lack a sufficient rationale and needs to be fixed soon, maybe with links on how to fix the rationale and why this is important? Then why weren't these images reviewed by humans to confirm BCB's suspicions before the deletion warning was dropped on them

  We don't tag images for improvement when they may be copy-vios. We tag that they may be copy-vios and give a few days to be fixed. If you don't like that process, you're in the wrong place. Betacommand didn't write the policy, he simply enforces it. And the messages the bot drops are discussed above. They do exactly what you just requested they do. Perhaps it's assumptions of bad faith on your part and that of others that these completely non-confrontational messages are taken as hostile.

• Given the number of people complaining, perhaps it's a reasonable belief that the messages are not non-confrontational? And perhaps the experience after receiving the messages when one seeks discussion and explaination simply reinforces that impression? —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• Then why weren't these images reviewed by humans to confirm BCB's suspicions before the deletion warning was dropped on them?

  They are tagged so that they are reviewed by humans. Admins review the images before deleting them. If they were incorrectly tagged, which practically never happens, if ever, the tag is removed. If there's an easy fix, they're fixed.

• That begs the question: why is the warning so stern and why doesn't it state something more like "Our automated system has flagged this article as possibly lacking a Fair Use rationale. It will be reviewed by an admin no sooner than ten days from now, and may be removed due to copyright restrictions if it has not been updated to meet the current fair use requirements." —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• The warning BC received for attacking another editor with a bot was more gentle than the warning a couple thousand users received for posting images that don't fit the obtuse, mechanical requirements of an obscure guideline.

  Tell it to the Foundation. You have serious comprehension issues.

• That must be why I'm the only editor to have ever complained about this, and why this page is so short and uncontentious. Thanks for making this personal, by the way.

• I don't buy the argument that a bot cannot write a fair use rationale for, say, an album cover that already has a license... There are better solutions than what we have.

  I don't buy that it's necessary as album covers are probably the easiest FURs to write. How lazy are you people becoming that you demand a bot be coded to do this for you? If it's such an easy code to write (certainly much easier than just writing your own FUR), stop complaining and get to coding.

  • So...BCB is absolutely critical because checking these images is too hard to do manually, but asking that a bot who checks them also complete the simpler ones is "laziness"? Are you interested in trying to solve any problems or are you just trying to make people angrier? —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)
• Another better solution would just be to append the rationale to the appropriate non-free use templates, so users fill out both in one go. The licenses and their relation to fair use are another source of confusion: why do we need these two separate templates that deal with essentially the same thing?

  Feel free to raise this issue in the appropriate forum.

• Which would be...? —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• So, are all these tagged articles going to be reviewed by human beings before deletion? Are they going to fix the simple mistakes? Are images that are used multiple times and have one sufficient rationale but not another going to be deleted, or just removed from the unauthorized article page?

  I think this has been stated above multiple times as well. Bots don't delete images. Admins, all of whom, with maybe one exception that isn't of concern here, are human, do the deletions. And they review the images and fix the ones that can be.

• Great. How would I know that if I didn't come here and ask? How will the next editor who receives the message know that? —Torc. ^(Talk.) 11:11, 22 February 2008 (UTC)

• Glad to get all that cleared up for the umpteenth time. Lara❤Love 19:58, 20 February 2008 (UTC)

  Yes, and there have been cases of admins deleting thousands of images in a short space of time, and then apologising later for mistakes made in that deletion run. Do you not think that admins should take time to carefully review each case, rather than semi-automating the deletions? There are cases where a cursory inspection doesn't find the simple fix. Sometimes a closer look is needed to spot what the mistake was, and to then tweak the image of article to fix the problem. Carcharoth (talk) 15:14, 21 February 2008 (UTC)

  I do check before I delete. If others don't, that's something you need to address with them. Lara❤Love 18:30, 21 February 2008 (UTC)

Questions

• I would be interested to know the following.

1. About what part of the abuse BC gets is from people who have received ONE notice about ONE of their images? What part is from people who have received MULTIPLE notices at once? (Obviously these would be guesstimates, at best, but I suspect the answer would be enlightening.)
2. Is there a way--note, not "is there an EASY way"--to modify the workings of this bot so that, instead of twenty or thirty separate notices splashed all over a user page each time BCB runs, the run would be completed, a list for each user would be generated, and ONE SINGLE message, with a list of affected images, could be placed on the user's page?
As you might have guessed, I have a suspicion here: I think much of the rage accorded to this bot (and thus, its operator) is not due so much to what it does, but to the manner in which it's accomplished. Knowing little about GDFL and free use, I'm willing to assume that the bot performs a necessary task; at any rate, it's a task mandated by the Foundation. But I think that people who see what appears to be a Sisyphean task's-worth of "disputed image" messages on their talk page (just for an example, take a look at User talk:Azumanga1--note: this is a completely uninvolved user whose page is only an example) are more likely to get angry than a user who logs on to find a message saying "okay, here's the list" and a link to a clearly-written page of instructions. (I'm a reasonably-intelligent human, but all the "how to write an acceptable rationale" instructions I see have confused me.) Also--and I really hope BC doesn't take this as yet-another attack, because he's been nothing but helpful to me--I think the constant attacks have gotten under BC's skin, and he's developed some less-than-ideal ways of handling perceived criticism. It becomes a vicious cycle: the bot makes edits that honk off users, the users spit fire at BC, BC spits back, matters escalate, and all of a sudden we have twenty-eleven AN threads calling for a ban of both BC and his bot. I'm wondering if reducing the aggregate mass of the warnings would reduce the drama as well. Gladys J Cortez 15:26, 19 February 2008 (UTC)

as to your point about BCBot and multi post, Ive been thinking about how to do it for months. there are two main issues that I am having.

1. go back to a single threaded checking method.

the main drawback is what now takes about 8-10 hours to check will take 800-1000 hours to check per run.

2. hack together some method to get 100 seperate threads to talk and play nice together, and avoid random thread crashes due to random server errors.

the main issue here is that it increases the risk of users not getting notified, something that I do not want to happen.

β^_command 16:03, 19 February 2008 (UTC)

Can't you just make a list as the bot does the tagging over 10 hours, and then run a program to consolidate the listings, and then do a separate run to notify each user once? Delaying the notification of the user by a day could be workable if you make the date and category eight days after notification instead of seven. BTW, what is a "thread crash" - do you mean an "edit conflict"? Carcharoth (talk) 16:16, 19 February 2008 (UTC)

please see Thread (computer science) for what a thread is. but one of many things can cause it to crash, (Ive tried to catch as many as I can) it could be an edit conflict, spamblack list error, page protection issues, image deletion, or some other random problem that the bot cant handle. it then stops the bot. (given that there are in essance 100 copies of the bot running at a time loosing a few is nothing major.) β^_command 16:44, 19 February 2008 (UTC)

While I agree that users not getting notified would be a bad thing--and again, please recognize that I'm coming from a tech-support-but-no-programming-whatsoever background--but if a user ISN'T notified due to a thread crash or other event during a given bot run, wouldn't the "missed" users from one run get picked up during the next? Not that it's ideal, of course--but I can't help but think there's got to be SOME way to improve the current situation. And again, please don't see my ideas and questions as criticism of you personally; I think you'll agree with me, though, that the current mode of operation is apparently untenable for both the users AND for you. Gladys J Cortez 18:47, 19 February 2008 (UTC)

users would not be notified. once an image is tagged BCbot does not re-tag an image if its still tagged. that would mean any images tagged during a run would not get notified if there was a thread crash. Ive been thinking of methods of improvement but they all get shot down for one reason or another. β^_command 19:00, 19 February 2008 (UTC)

Sounds like you are trying to do too much stuff in parallel (sounds like multi-tasking) and coming up against technical limits. What I'm asking is can it not make a list somewhere off-wiki of the users that need notification - the list gets consolidated - and then the bot runs again and makes notifications. Actually, this really is a year too late now, but for the future could you really, really try and sort this problem out? There may be other bot runs in future where users will be notified of multiple problems. In fact, I'm sure there could be consensus generated to require bot notices to be given in consolidated format, rather than separately. Call it the Betacommandbot clause if you like. Carcharoth (talk) 18:53, 19 February 2008 (UTC)

multi-tasking is what I am doing, I give 500 images to each thread and have that thread work on them. Ive got one idea, that Im working on but Im not sure how it will pan out —Preceding unsigned comment added by Betacommand (talk • contribs)

I have a couple of possible solutions:

1. If I am correct in assuming that your 100 threads don't edit conflict notifying the same user, then have the bot look for a previous note by itself (using comments or whatever) and find the end of the list and add the name to that instead of simply appending it to the user's talk page.
2. If your 100 threads do edit conflict notifying the same user, you could have another thread/process running which waited for incoming UDP packets notifying it of new notifications to post. It could then limit it to one thread per user so it wouldn't edit conflict.
3. Another idea is that you could have the bot log to a MySQL database and have a process come behind and tally everything into a list and post it, maybe once every hour or so and update lists that already exist on users' talk pages.
4. Or you could continue as you currently do and then run over BCBot's notifications for that run and consolidate them after the run is finished.

Those are my ideas. If you want, I could even help you with implementing any of these, though Python (which is what BCB is written in, if I am not mistaken) isn't my forte. -- Cobi^(t|c|b) 07:11, 22 February 2008 (UTC)

Write the list to a subpage of the user's talk. At first it won't exist, so when the first item is found, create the page and leave the user a notice on their talk. As the bot finds new items, add them to the list. The notification only happens once per run (because the page already exists when the second item is found, leaving a notice can be skipped). That would be my suggestion. It can be made fancier by taking dates into account and leaving notices when the first time THIS run an item is found, or make the subpage different for each run. ++Lar: t/c 04:11, 20 February 2008 (UTC)

That seems brilliant, but I don't code, so I don't know. Here's a simple thought I had, if that isn't possible or something. Instead of the whole long message that the bot leaves now, what if it were just one sentence that linked to the whole template somewhere else. I get that there'd still be multiple section headers, but big deal. Either you're going to fix the images then delete or archive the notices, or you can just rollback the notices. All this over talk page messages. Lara❤Love 08:00, 20 February 2008 (UTC)

"All this over talk page messages" - exactly - so why not fix that complaint? Lar's suggestion sounds workable. Carcharoth (talk) 08:57, 20 February 2008 (UTC)

The problem of the user subpage (as with a local file on BCBot's system) is that the threads can edit conflict with each other, and then you're screwed. The user would miss a few images (and complain, a lot :)) -- lucasbfr ^talk 13:36, 20 February 2008 (UTC)

How does the current system avoid such edit conflicts? Carcharoth (talk) 13:58, 20 February 2008 (UTC)

I never looked at BCBot's code so this is just speculation, but my guess is that every message left on a talkpage uses a new section ("section=new"), which prevents conflicts (Mediawiki does not throw an edit conflict if the conflicting edits aren't in the same section). -- lucasbfr ^talk 14:03, 20 February 2008 (UTC)

So then you just use new sections on the subpage. The point of the subpage is to avoid an overwhelming list on the main talk page. All that is needed is a link to the list. Carcharoth (talk) 14:19, 20 February 2008 (UTC)

Yes, because the undo/rollback button or the archive option are such a hassle. Lara❤Love 01:55, 21 February 2008 (UTC)

Rollback was not designed as a tool for people to tidy their talk pages with. And anyway, I think you are suggesting that if a bot creates a long messy list, a human should tidy it up. Why not get a bot to do the tidying up or avoid the mess in the first place by figuring out a way to add one notice and a list of images with the notice? If you were leaving the notices by hand, would you leave 30 identical notices or one notice and a list of 30 images? Carcharoth (talk) 15:10, 21 February 2008 (UTC)

If a user subpage were to be used, I would think it should be an opt-in kind of thing. Perhaps something like a blank template on the talk page or a category users can add themselves into that basically says (to the bot), 'I know what your warnings are all about, just make me a list on my subpage.' That way people who aren't familiar with the bot's work are still getting talk page notifications (with a paragraph of information, whatever it says) and those that are can just watch their subpage (where it would leave perhaps a note with the title being the image name, and the text being bullet pointed reason and date scheduled for deletion, just one idea). I have concern about creating date based subpages, because then you end up with a ton of pages that will never get looked at and potentially never deleted. It would seem smarter to me to put the list on one subpage (which would be standardized) and let the user clean it up as he/she cleans up the images/deadlines pass/etc.

Of course, if we are throwing out ideas that could involve a lot more work, I would expand upon the template on the talk page idea and add another bot into the picture. This bot would scan the BCB list subpage, remove any redlinked sections (i.e. images that had already been deleted), then update the template on the user talk page with the number of sections on the page, which would then create a note saying 'You have X many images on your BCB page that need attention' (this message can say whatever, this is just an example). This update in turn would then create the watchlist notification (once per day if the bot was run once per day). I guess depending upon how many people opted for this (and unless we have a bot writer like Betacommand who can write a multithreaded bot to attack this problem, I would imagine the number of pages to scan could eventually get a bit large to do quickly. Perhaps the bot could even nominate for deletion any blank BCB subpages. Of course you would end with people who don't clean out their list (which goes back the opt-in bit) and end up with a notification to check images that other people have fixed (to fix that you would need to go back and check if the image fails again, which would be even more work). And of course what I am talking about adds an entire layer on top of the seemingly simple idea of a subpage list (which I would bet most people would ignore or forget about, which is one of my concerns) and perhaps is a bit overkill to a problem that could be solved other ways. Just food for thought (from someone who doesn't really mind the BCB notifications on his talk page). - AWeenieMan (talk) 18:27, 20 February 2008 (UTC)

The entire concept of using a bot to raise issues with image tags is not user-friendly. It's about on the same level as automatic voice-mail machines; those delightful systems we all enjoy using so much. Beyond that, I would suggest that some of the comments from the bot's operator are not particularly civil, which surely does not help matters.

Just because there are issues with image tagging, and possibly even most of the users alerted by the bot are not conforming to image upload policy, the end of sorting that out nevertheless does not justify the means. It seems to me that many of those speaking in favour of the bot are merely supportive of policy (which really, there's not much of a leg to stand on for anyone who isn't). However, they might want to consider how the issue should be addressed, even if using a bot is part of the solution. Also, surely it should not be up to one individual to act on image tagging? It seems to me that Betacommand has far too much latitude on the issue, merely through being a bot operator. Indeed surely there should be far stricter conditions on people running bots? zoney ♣ talk 20:52, 20 February 2008 (UTC)

I don't only support Betacommand because I support the policy. Betacommand and the work he does with his bot are invaluable to the project. It would be unmanageable without him. As far as the messagse go, they're the same message templates left by editors who do image tagging. It's like the opposition to the welcome bot. People said it was too impersonal. Who cares? The point is that the information gets across. A bot can get so much more done so much faster, which is necessary for a project of this size. Lara❤Love 01:55, 21 February 2008 (UTC)

As I said, the ends do not justify the means. Undertaking the work of tackling image licensing does not place Betacommand or his bot outside criticism. That's the kind of logic that allows all manner of unpleasantry to occur on Wikipedia simply because the contributors are "valued". The unpleasantry meanwhile ensures that a lot of other contributors end up packing it in altogether and making better use of their time.

It is not the case that there is absolutely nothing can be done if the bot is done away with. For example, it might be a better idea to look at avoiding the workload in the first place. Fair use images are a bad idea for one thing. zoney ♣ talk 11:45, 21 February 2008 (UTC)

Comments from AjaxSmack

Although I've tangled with User:Betacommand and his lap dog User:Hammersoft over other issues before, their direct rudeness, sarcasm, and incivility is a refreshing change from the stultifying mandated consensus and community that is frankly geo- and culturally biased toward the sensibilities of the dominant Euro-American editor base. Yes, there are problems but the Betacommandbot seves a useful purpose and it's opponents seem to be reacting in anger toward perceived rudeness in auto-generated deletion warnings but have few viable alternative proposals of their own. Having said that, I will admit sympathy with some of the arguments presented above:

1. The uploading process for non-free images should clearly include all elements needed to satisfy BCB in a simple format.
2. The deletion notification should cast a wider net and allow for more time to add FUR

Imagine a newer editor with around 500 edits who edits infrequently wants to upload a non-free image. He is likely not that well versed in policy and follows the upload form and includes a FUR but the article name is misspelled or not wikilinked. BCB catches the mistake and posts a deletion notification on the user's talkpage. However, the user is an infrequent editor and doesn't see the notice until after the article is deleted. He wants to re-upload the image with the proper information but the relevant information is on several policy pages and is not readily distillable from overly verbose prose so he gives up. No doubt the anti-fair use crowd is cheering but death of an image though bureaucracide should not be the ideal. The upload process has gotten better but there is room for more clarity as well as tweaking of the bot. — AjaxSmack 00:37, 21 February 2008 (UTC)

The bot doesn't look for a link, which is not required. If it's just a mispelling, the reviewing admin or someone else doing image patrol will correct it. So, for your hypothetical situation, the image most likely would not be deleted. Lara❤Love 01:55, 21 February 2008 (UTC)

Lara, you are aware that when there are large numbers of images involved, that some admins sweep through the category with less care than is ideal? The "admins will catch all mistakes" attitude is understandable, but doesn't always work. Carcharoth (talk) 15:07, 21 February 2008 (UTC)

The case I gave above was not a hypothetical; it was a distillation of my experiences. I have had images I have uploaded that are totally within fair use guidelines deleted because of a non-wikilinked article title and, since I don't always edit Wikipedia weekly, I missed the notice. Another image was deleted because the target article was moved (renamed) and I never got a bot notice of the impending deletion. Only after 20-30 minutes of carefully reading several pages of policy did I accrue the knowledge necessary to ward off similar deletions of future uploads. Although these are anecdotes, the list at User:Blathnaid/Sandbox seems to back up my experience. I can imagine that quite a few editors more casually attached to Wikipedia than myself would be unwilling or unable to wade through the Slough of Despond that is Wikipedia non-free-image-use policy. I'm not arguing against usage of the bot, just for improvement of upload policy and procedure. — AjaxSmack 17:12, 21 February 2008 (UTC)

Then that's something you need to address with the careless admins. It's been stated several times that people think it would be better of users were given more time to make corrections. Betacommand simply runs his bot to find images that fail to meet policy and tags them with a generic community-written template. Bring up on AN/I that admins need to avoid deleting images that they can correct. That you believe there is no need to rush. Backlogs will be backed up for some time, but it will give editors more time to make corrections, and fewer mistakes will be made. Win/win. Lara❤Love 18:36, 21 February 2008 (UTC)

Wikiproject Religion tagging

Sorry, I wasn't aware of this page, so started Wikipedia:Administrators'_noticeboard#Mass-spamming_by_User:John_Carter_and_User:Betacommandbot which has generated some discussion. Johnbod (talk) 01:01, 21 February 2008 (UTC)

This isn't really a Betacommand issue. Lara❤Love 01:44, 21 February 2008 (UTC)

Several people there think it is, and are not convinced by by his "I voss only following orders" attitude. Clearly the requester is the primary problem. Johnbod (talk) 02:30, 21 February 2008 (UTC)

Why must people always link such stupid bullshit to Nazis? Seriously. Can't you take such things just a little more seriously? Avruch ^T 00:27, 22 February 2008 (UTC)

OK, what's another handy example everybody will understand of somebody deflecting blame by claiming they were doing as told? —Torc. ^(Talk.) 11:19, 22 February 2008 (UTC)