Wikipedia:Bot requests

; Archives
	August 2004 – September 2005; June 2005 – November 2005; August 2004 – January 2006; February 2006 – April 2006; November 2005 – February 2006; February 2006 – April 2006; May 2006 – July 2006; August 2006 – December 2006; January 2007 (Part A); January 2007 (Part B); February 2007-current;

Shortcut

[[:|WP:BOTREQ]]

This is a page for requesting work to be done by a bot. This is an appropriate place to simply put ideas for bots. If you need a piece of software written for a specific article you may get a faster response time at the computer help desk. You might also check Wikipedia:Bots to see if the bot you are looking for already exists. There are also quite a few "frequently denied requests", for various reasons, such as a welcoming bot, as it would de-humanize the process, and an anti-vandalism bot, as several already exist. If you want to request a bot to populate a category for a wikiproject, please create a subpage with a full list of categories to be used, as most bot operators who can complete this task will not go into all subcategories, as some members may be irrelevant to your project. Also note that if you are requesting that an operator change or add a function to an existing bot, you should ask on his talkpage, if you have questions about certain bots, they should be directed to the bot owner's talk page to the Bot Owners' Noticeboard, and that if a bot is acting improperly, it should be posted to the owner's talk page, the Administrators' Noticeboard, or AIV, listed in increasing levels of severity, and a link to the discussion may be posted at the Bot Owners' Noticeboard if appropriate. Please add your bot requests to the bottom of this page.

If you are a bot operator and you complete a request, note what you did, and archive it. Requests that are no longer relevant should also be archived in a timely fashion.

See also: Current policy on bots and Wikipedia:Bots/Frequently denied bots, to make sure your idea is not listed.

This talk page is automatically archived by Werdnabot. Any sections older than 5 days are automatically archived to Wikipedia:Bot requests/Archive 12. Sections without timestamps are not archived.

Add a new request

External Links => Incline Citation Bot

In many wikipedia articles, there are external links after a sentence which is used in a number formating (so the external link has no extra info attached to it); example [1].

Would it be possible for a bot to remove "[" replacing with "<ref>" and remove "]" replacing with "</ref>"?

After that the bot would search if there is <references/> in the article.

If it cannot find it, the bot would make a new sub-section "==References==" and place "<references/>" below that.

The bot would have to make a list from the last dump of all the mainspace articles, and perform the operations (hopefully it will get over within one week).

Example Nonvolatile_BIOS_memory

--Paracit 23:24, 19 March 2007 (UTC)[reply]

In my view, end of section references are preferable when there is a textual description of the reference. For a pure html link, the reference section just obscures matters, requiring an extra click-through. However, putting raw links into a reference section might encourage people to change them to proper citations. That's a testable proposition, and if it's true this would be a good idea. Derex 00:12, 20 March 2007 (UTC)[reply]

Some editors might consider it controversial to change an inline link to a cite.php reference. Even if it encourages adding full citation info, some will view this as a short term detriment, by making the link one step removed. Gimmetrow 01:34, 20 March 2007 (UTC)[reply]

It is possible to create such a bot and not very complicated actually. But I share the concerns mentioned above. Maybe you should see if you can reach a consensus in a discussion on this question at WP:CITE and/or WP:EL. Perhaps this has already been decided on and you can provide a link to it? I'd be interested in helping with the bot / programming it, if there's such a common agreement. I suggest continuing to talk about a bot when we are sure your suggested changes are supported by the community. — Ocolon 08:33, 20 March 2007 (UTC)[reply]

I've run across other articles where an editor has manually (I assume) converted embedded links to references/footnotes, without adding anything else. I suppose that encourages editors to work the references to improve them; I'm not sure (because I didn't systematically follow up over the months) that anyone actually did.

As far as starting a discussion, I also support that before a bot is written and approval for it requested. I note that WP:EL is the wrong place, however; that policy has to do with the "External links" section, which isn't an issue here (except that the bot should be programmed to stay out of that section). In addition to) WP:CITE, it's worth noting the proposed change at Wikipedia:Embedded citations (which isn't a policy or guideline) and Wikipedia:Manual of Style (links). -- John Broughton (♫♫) 01:33, 29 March 2007 (UTC)[reply]

I would almost suggest to just be bold, and manually do a few articles and see the reactions. Do the links get improved? Do you end up just annoying people? etc. —— Eagle101 ^{Need help?} 04:03, 2 April 2007 (UTC)[reply]

Check this out so many external links converted to inline citations:

http://en.wikipedia.org/w/index.php?title=Clinical_depression&diff=118654983&oldid=118576074

--Parker007 21:14, 4 April 2007 (UTC)[reply]

I agree that the Clinical depression article *did* benefit from converting the external links to inline citations. A problem that this conversion did not address is that the reference sections contain a lot of raw link text that ought to be replaced by useful 'metadata' in the form of authors, titles and complete names of publications. (Each raw link could be replaced by a citation template, and the link itself could be filled into the 'url' field, so the citation would be clickable). Someone could go through manually and fix that. Another more general problem is that this article seems to be overwhelmed by its excessive references. Wikipedia is not a directory or a bibliography. Not sure what your tool could do about that, but it might suggest to us that manual fixup can do things that a bot cannot. EdJohnston 16:43, 8 April 2007 (UTC)[reply]

See m:standardize_notes.py. Needs modification for <ref>. (SEWilco 04:11, 2 April 2007 (UTC))[reply]

WikiProject Cycling

Hi. Could someone please tag all articles, categories and templates that fall under Category:Cycling and it's child categories with {{cycling project}}, and automatically assess all stub articles with class=stub and all non-articles with class=NA? Thanks, Severo^TC 23:41, 7 April 2007 (UTC)[reply]

Wasn't this done back in February? -- SatyrTN (talk | contribs) 00:57, 8 April 2007 (UTC)[reply]

Yes, but there are more pages now, and assessment has started and I'd appreciate the bots ability to automatically assess loads of articles based on stub templates. Regards, Severo^TC 07:50, 8 April 2007 (UTC)[reply]

Is it just the "Cycling" category and one level deep? -- SatyrTN (talk | contribs) 15:39, 8 April 2007 (UTC)[reply]

All articles and cats should fall under the Project's span :). Severo^TC 18:42, 14 April 2007 (UTC)[reply]

You would be surprised... If I start with the category Cycling->Utility cycling->Green Vehicles->Fuel cell vehicles... You can see it would get sticky if the bot tagged Type 212 submarine with the cycling banner :) I've created a list of categories from the category tree that starts with Category:Cycling - please review User:SatyrBot/Cycling and remove the categories that don't belong. Then I can run the bot to tag articles if they don't already have the tag. Thanks! -- SatyrTN (talk | contribs) 19:23, 14 April 2007 (UTC)[reply]

Bot

Can someone please develop a bot for me. I would love to have a bot as this would allow me to help contribute to Wikipedia and as well as that the more bots we have, the more pages we can edit. The more the merrier! --pizza1512 09:05, 10 April 2007 (UTC)[reply]

Could you specify what the bot would do? Funpika 22:02, 10 April 2007 (UTC)[reply]

Could I have a bot which removes repetition of internal links, i.e. when a word comes up more than once, it would not be linked more than once. Um.. I'm not sure if this is clear. Let me give an example... take the first paragraph of...

Ludwig van Beethoven was a German composer. He is generally regarded as one of the greatest composers in the history of music, and was the predominant figure in the transitional period between the Classical and Romantic eras in Western classical music. His reputation and genius have inspired — and in many cases intimidated — ensuing generations of composers, musicians, and audiences. While primarily known today as a composer, he was also a celebrated pianist and conductor, and an accomplished violinist.

The word composer is linked more than once and would be removed by the bot to only link the first 'composer':

Ludwig van Beethoven was a German composer. He is generally regarded as one of the greatest composers in the history of music, and was the predominant figure in the transitional period between the Classical and Romantic eras in Western classical music. His reputation and genius have inspired — and in many cases intimidated — ensuing generations of composers, musicians, and audiences. While primarily known today as a composer, he was also a celebrated pianist and conductor, and an accomplished violinist.

I hope you understand what I mean. --pizza1512 10:24, 11 April 2007 (UTC)[reply]

The burden of proof is on the bot-maker to demonstrate that the bot:

is harmless
is useful
is not a server hog
has been approved
abides by all guidelines, policies and common practices
— Wikipedia:Bot policy

I don't see how this is useful, and it may actually be counter-productive. ST47Talk 10:51, 11 April 2007 (UTC)[reply]

Well, the reason why I asked for a bot with that requirement is that it only processes articles that requires cleanups. I just tried to clean up a whole article full of links, and it took a very long time. Having a bot that does that, would allow clean ups to be made more easily and so the articles can be rewritten more easily. I'm only trying to help. --pizza1512 19:18, 11 April 2007 (UTC)[reply]

I think you need to specify exactly what types of clean ups you would like to do. WatchAndObserve 20:17, 11 April 2007 (UTC)[reply]

It would be the removal of repetitive links in a document then. I would like the bot to look through articles that require clean ups and it would remove those unnecessary addition links.

As well as that, could I have a bot that automatically create information boxes for biographies of people or would that be a bit ambitious? --pizza1512 21:38, 11 April 2007 (UTC)[reply]

Is the problem that your first idea is hoping to correct really that big of a problem though? I agree that there are articles with entirely too many repeated links, but are there enough of them to justify the bot searching through the thousands of clean-up articles looking for them? As for the second bot: Could you elaborate a bit? What information would the bot use to generate these infoboxes?--Dycedarg ж 05:18, 12 April 2007 (UTC)[reply]

Well, I think it is a problem that needs to be solve. Many editors use up too much time, editing these 'clean up' articles. However, by creating a bot that removes these, it would allow articles to be rewritten and back up to the true quality of Wikipedia. I really do think that there are enough links to justify the bot searching through them. The other day I had to delete over 20 links in an clean up article - and boy it took me a lot of time...

As for the info boxes, the bot could I suppose detect a few key words that are related to them. As well as that, because most professions are marked as categories on Wikipedia, it could create a userbox using that knowledge also. --pizza1512 08:41, 12 April 2007 (UTC)[reply]

Hello is there anyone there who can help? --pizza1512 ^{Talk Autograph} 18:17, 15 April 2007 (UTC)[reply]

Signature bot

I'm considering a slight username change and I was wondering if anyone had a bot that could exchange my old signature, to one with my name change in? SinceI started here, I've only had 2 different signatures so it's not like there's loads of different signatures togo through. My only concern is there's probably around 3,000 edits that the bot would needto make. I've considereda re-direct from myold userpage to my new one, but I'd prefer if it was set in stone. Cheers Ryanpostlethwaite contribs/talk 18:03, 12 April 2007 (UTC)[reply]

For what it's worth, you could do this easily in WP:AWB. That said, I think that the established precedent is to redirect (protected) your userpage and avoid clogging up recent changes with switching your signature. I've seen a few people doing that, but it's generally frowned upon. alphachimp 18:10, 12 April 2007 (UTC)[reply]

Yeah I was considering AWB, would have been nice to have something automated to do it with though, oh well, I guess I'll settle for a protected redirect Ryanpostlethwaite contribs/talk 18:13, 12 April 2007 (UTC)[reply]

If you get the task approved on WP:BRFA (I'm sure they won't), I can have Alphachimpbot do it. It'd be extremely easy. alphachimp 18:13, 12 April 2007 (UTC)[reply]

Wikiproject Georgia

Could a bot add {{WikiProject GeorgiaUS}} to the talk pages of all the links in these articles: List of cities in Georgia, USA and List of towns and villages in Georgia (U.S. state) Akubhai 12:10, 13 April 2007 (UTC)[reply]

SatyrBot can do that for you. Please see User:SatyrBot/Current project for a couple questions. -- SatyrTN (talk | contribs) 13:41, 13 April 2007 (UTC)[reply]

Welcoming bot

Could someone create a bot for automatically welcoming new users when they create a new account? ♠TomasBat (@)(Contribs)(Sign!) 02:01, 14 April 2007 (UTC)[reply]

That might have been suggested before...you'll have to check the archives. However, I think that such a bot would make the system even more impersonal.--Ed ^{¿Cómo estás?} 02:07, 14 April 2007 (UTC)[reply]

That's correct. A welcoming bot is be very easy to implement, but unnecessary. The whole point of welcoming somebody is to build connections between users. Oleg Alexandrov (talk) 02:09, 14 April 2007 (UTC)[reply]

Well, couldn´t it be like the official welcome or something of the sort? It would really help all new users get started... ♠TomasBat (@)(Contribs)(Sign!) 02:28, 14 April 2007 (UTC)[reply]

Users are getting started by being welcome by other users. :) Sometimes that takes a while, but I bet that any user which has made say 20 edits will be noticed and welcomed by somebody. Also, I would doubt that people are going to be helped so much by the welcome template right away. Anyway, on the whole, I think a user would be better off waiting a bit and being welcomed by a real person, than being welcomed immediately by a bot. Oleg Alexandrov (talk) 02:54, 14 April 2007 (UTC)[reply]

Ok, but couldn´t we instead use the Smiling bot mechanism (see section right below) but , intead of handing a smiley, make it hand a welcome to a random user which has got no posts on the talk page? ♠TomasBat (@)(Contribs)(Sign!) 19:30, 14 April 2007 (UTC)[reply]

No - A random user is easy to get, a random user who happens to have no messages is not. Use one of the fancy JS scripts that works on the newusers log? ST47Talk 19:34, 14 April 2007 (UTC)[reply]

OK. ♠TomasBat (@)(Contribs)(Sign!) 20:47, 14 April 2007 (UTC)[reply]

Smiling Bot

Could somebody make me a bot that helps in handing out random smiley awards? Now, it would work in the following way:

You go to the bot´s userpage.
You click on an image of a button (say a green button).
The bot detects your username by detecting which user clicked the button and automatically enters a random user page and inserts a random smiley, with your detected signature.

Idea: Maby, once clicking the buttom and detecting your username, it could automatically enter special:random, go to the page history and send a smiley tothe first user on the list. Or it could, once the button is clicked and the username detected, travel to special:listusers and select a random user from there.

With this bot, handing out random smiley awards would be a lot less tediouse and users who wish to do so would lose a lot less time.

♠TomasBat (@)(Contribs)(Sign!) 19:07, 14 April 2007 (UTC)[reply]

The burden of proof is on the bot-maker to demonstrate that the bot:

is harmless
is useful
is not a server hog
has been approved
abides by all guidelines, policies and common practices

What's the point? We don't need a bot that picks random users, use Recent Changes and do it yourself. ST47Talk 19:15, 14 April 2007 (UTC)[reply]

It would be even faster with this bot; extremely fast... Which is somethig good... ♠TomasBat (@)(Contribs)(Sign!) 19:22, 14 April 2007 (UTC)[reply]

But we have the same problem as above - a bot smiling at people - yes, it can sign your name, but the history would still say it would be from the bot...also, it'd have to be on toolserv and involve a <form>, as you can't put CGI scripts on your userpage...by the way, your signature is disgustingly long, can you shorten it a bit? ST47Talk 19:32, 14 April 2007 (UTC)[reply]

How about if we add automatic/bot-styled code to the users who wish to hand in smileys in this fast way, instead of using a seperate bot? I think this would certainly fix the problem... ♠TomasBat (@)(Contribs)(Sign!) 20:50, 14 April 2007 (UTC)[reply]

I don't see how this will help us write an encyclopedia. ^High_InBC^{(Need help? Ask me)} 19:34, 14 April 2007 (UTC)[reply]

Users hand in smiley awards through bots, the users who recieve those get cheered up and start contributing more. ♠TomasBat (@)(Contribs)(Sign!) 20:46, 14 April 2007 (UTC)[reply]

But it's a BOT! I'd say, look, a little metal box has been programmed to like me, big deal. ST47Talk 20:48, 14 April 2007 (UTC)[reply]

You don´t seem to understand. The bot is a semi-automatic one which aids users who wish to send random smileys by enabling them to hand them faster. The user orders the bot to make a task faster, but the bot does that because the users orders him to do so, not because he is programmed to like that user and be friendly. ♠TomasBat (@)(Contribs)(Sign!) 21:16, 14 April 2007 (UTC)[reply]

However, the bot is making the actual edit? 24.0.52.44 21:43, 14 April 2007 (UTC)[reply]

If not, then I think you just want some javascript that will select a random user, load their talk, and make a new section. 24.0.52.44 21:45, 14 April 2007 (UTC)[reply]

Hmmmmmm... I suppose so... But, could that java-script also automatically define the section name as Random Smiley Award and the text as {{subst:User:Pedia-I/SmileyAward1}} ~~~~? ♠TomasBat (@)(Contribs)(Sign!) 21:58, 14 April 2007 (UTC)[reply]

You do realize that randomly selecting a user to give this award to is as likely to turn up vandals, spammers, banned users, and users that made three edits and left as it is to turn up good contributers? How is this helpful?--Dycedarg ж 22:44, 14 April 2007 (UTC)[reply]

Since the Smileys are given out at random, it may sometimes happen that one is mistakenly given to an editor of bad reputation. This should not be seen as an endorsement of such a person, but merely a mistake. Besides, perhaps the sight of a Yellow Smiley on their page may inspire efforts at self-reform. Alternately, for a genuinely Evil Editor the sight of a Yellow Smiley spreading warmth and good cheer might serve as a just punishment, like well-wishers wishing a Merry Christmas to grumpy Ebenezer Scrooge. ♠TomasBat (@)(Contribs)(Sign!) 23:45, 14 April 2007 (UTC)[reply]

I don't have any good knowledge of JS files, but your proposal might be able to be written on your monobook. A bot wouldn't be needed.--Ed ^{¿Cómo estás?} 01:08, 15 April 2007 (UTC)[reply]

(outdent) If this gets approved (which I don't see happening), I would want a way to opt out...I don't want my name to be in the drawing for random talkpage spam. ^demon^{[omg plz]} 03:17, 15 April 2007 (UTC)[reply]

Ok, so this isn´t the right place to ask; so, where can I ask for such js files for my monobook? ♠TomasBat (@)(Contribs)(Sign!) 20:28, 16 April 2007 (UTC)[reply]

Change ordinal º to degree ° and remove space between ° and C or F

I read an article where the masculine ordinal sign º (º) was used some places instead of the degree sign ° (°), and there was also some inconsistent usage of space between the degree sign and the scale denominator (C or F). I edited that article, but looking at a few related articles I realized that my contribution was just a drop in the ocean. Could a robot please change incorrect use of ordinal º to degree ° and also remove inappropriate space between ° and C or F?

Change	To	Change	To
ºC	°C or °C	ºF	°F or °F
º C		º F
º C		º F
ºC		ºF
º C		º F
º C		º F
° C		° F
° C		° F
° C		° F
° C		° F

Unless the search can be constrained to solitary capital C's and F's, the change from ordinal º to degree ° should be monitored while the robot is running. The removal of space between ° and C or F, however, is straightforward. On the other hand, space between numerals and degree sign is optional and, also considering geometric angles, should not be robot'ed. --Eddi (Talk) 01:39, 15 April 2007 (UTC)[reply]

Send this over to WT:AWB they can take care of this. Betacommand ^{(talk • contribs • Bot)} 01:41, 15 April 2007 (UTC)[reply]

Will do. --Eddi (Talk) 01:48, 15 April 2007 (UTC)[reply]

BirthdayBot

Hello, As I am a constant active member of the Birthday Committee, I would like to run and manage a bot that is run once each day to read the special calendar entries for the day and post a random template on the user's talk page. It has been discussed in the past with support from other members and I would like to take this further. If a kind Wikipedian out there would be willing to construct such a bot for me, I would be very grateful. Many thanks, Extranet ^{(Talk | Contribs)} 11:46, 15 April 2007 (UTC)[reply]

I'll have a look. ST47Talk 11:55, 15 April 2007 (UTC)[reply]

Looks like it works - if you run a linux machine, I can send you the source? ST47Talk 13:57, 15 April 2007 (UTC)[reply]

Conflict of Interest Finder Bot

Someone should make a bot that searches articles for IP addresses that made many major edits to closely related articles(in the same category, or linked) in a short period of time with out making many other major edits and add it to a watch list for someone else to check, because I noticed that people(2 out of 2 so far) who probably work for a company related to the subject tend to never make any other edits other than a couple huge edits to the article. The bot could also have a list of categories that have a conflict of interest often(e.g. List of United States companies, or Category:United States presidential candidates), and therefore check those more often. I could maybe make the bot because I have some programming skill but not with internet stuff. 80.109.79.136 13:51, 15 April 2007 (UTC)[reply]

Talk to User:Beetstra, he has a similar bot at the moment. ST47Talk 13:57, 15 April 2007 (UTC)[reply]

What bot? 80.109.79.136 14:01, 15 April 2007 (UTC)[reply]

User:COIBot. ST47Talk 14:02, 15 April 2007 (UTC)[reply]

(edit conflict) That sounds like a nice idea, but I am not sure on how to implement this.

First, I am running a COIBot. Primary tasks at the moment are to match username with links (domains) added or match username with page edited. Reports are, atm, going to m:User:COIBot/COIReports (per day). COIBot has black-/white- and monitor lists, which give it the capability to either link users/IPs to certain data, to whitelist certain users on certain pages/links (some usernames have accidental overlap with the edits) and to monitor link-additions (handy when spammers use many IPs or sock-accounts). COIBot is still under development, though it is working quite well. I am waiting for a toolserv account, and will then also do a BRFA on this wiki to make it report here.

A second tool we have is a link-watcher, which alerts us of users adding many links. That alerts us to have a look into these edits.

I don't see (yet) how to create an algorithm for your proposal. Catching COIs on basis of editing patterns is going to be difficult. E.g. in mainspace my main interest is chemistry, and I edit all through chemistry mainspace. If I would have a COI that would not be noticed because of the broadness of the edits. On the other hand, a Madonna fan who only edits the page Madonna and her albums would not have a COI, but a very select editing pattern, which might give a false positive. Also new editors tend to make a lot of small edits without using preview (heh, even I do that), which again would be diffcult to catch.

So the problem here is that almost every wikipedia-editor has a focus on a certain set of articles, and most will make edits within a certain set of categories relating to their favourite subject. That pattern is the same as a company/IP that wants to promote their own interest.

If you have any ideas on how to do this, and we could have a look how far that would lie within the scope of COIBots capabilities. For more information, you can reach us on IRC (where also the edit-feeds can be found), talk-channel irc://chat.freenode.net:8001/wikipedia-spam-t or via the normal wikipedia channels (talkpages, email). Hope this helps, have a nice day. --Dirk Beetstra ^{T C} 14:35, 15 April 2007 (UTC)[reply]

I think it's very simple to do, all it does is look at major edits per day, major edits being something like 3 words changed or more, and finds a spike e.g. 1, 0, 5, 0, 1 and 1, then it looks at the 5 articles and finds out how closely related they are and finds out how closely related those articles compared to the other edits i.e. 1, 0, 0, 1 and 1(that, I hope, gets rid of the article focus thing) and comes up with a score(based on how closely related the articles are, those articles compared to the other days edits(edit size, etc.), and the size of the edits, i.e. the bigger edit the more suspicious) and adds it to the list for someone else to check. 80.109.79.136 15:20, 15 April 2007 (UTC)[reply]

PS: It should at least add people who, as their very first edits, make 3+ major edits in related articles to a list. Look at http://en.wikipedia.org/wiki/Special:Contributions/83.132.128.252 80.109.79.136 15:25, 15 April 2007 (UTC)[reply]

PPS: I just got myself an account I'm now Jeffrey.Kleykamp 15:53, 15 April 2007 (UTC)[reply]

(edit conflict)I see. I think this lies outside the scope of COIBot, completely different algorithm (though this may be a plan for a COIBot2). I am still afraid of a relative large number of false-positives with this, but that remains to be seen, and maybe can be tweaked when the bot is running (IPs making major edits may also be experienced users who are not logged in at that moment, or who are on a different IP every time and hence have more edits than expected). I'll let this idea mature a bit first. Please inform us when you start programming it! See you around! --Dirk Beetstra ^{T C} 16:01, 15 April 2007 (UTC)[reply]

I thought about this idea a bit, and it seems like a bot would have to:

Find a source of IPs
- Load contribs for each
  - Sort by date
  - Calculate mean, standard deviation, and determine the days with are more than X standard deviations above mean
    - Compare article titles using COIBot's algorithm
    - Report to IRC
- Load next IP

ST47Talk 16:49, 15 April 2007 (UTC)[reply]

I just tried to create simple stuff code in C# to see my skills with web pages and bots and it was way to complicated and I failed. Is there a website with some sample bot code written in C# or Java or even a bot tutorial because I learn fast but I need help with a start. Jeffrey.Kleykamp 17:33, 15 April 2007 (UTC)[reply]

I don't think so, but I can give you a sample of a framework written in Perl - perlwikipedia - or post a link to your code and someone can look over it. ST47Talk 17:45, 15 April 2007 (UTC)[reply]

I don't like the idea of comparing article titles using COIBot's algorithm (I don't really no what it does), I think it would be better but more complicated and slower to find out how far one must travel in the categories to find the other article, and I don't understand what you mean by "Calculate mean, standard deviation, and determine the days with are more than X standard deviations above mean" Jeffrey.Kleykamp 18:17, 15 April 2007 (UTC)[reply]

IntelliBot

Would it be possible to create a bot which would be able to perform calculations and approve of calculus on recent changes/new pages?, and then further explain what the calculus is intending to get across. This might help people who are doing maths, and are unable to see why the calculation may be the solution to their question. Just a thought. Doesn't matter if it doesnt 'come alive'. Radio_Orange (talk • contribs) 18:30, 15 April 2007 (UTC).[reply]

I'm not sure exactly what you mean, can we have an example? ST47Talk 18:41, 15 April 2007 (UTC)[reply]

GA candidates count

Would it be possible to have a bot that organizes/sorts/counts GA candidates? This hasn't been discussed extensively at GA; it just came up as an idea and I thought I'd ask if it were feasible. It would read the GA candidates page and count the number of nominees in each category, figuring out if they were on hold or under review.

Possible outputs would include:

A count of total GA noms outstanding, including those held or under review
An exception list, showing GA holds over 7 days, GAs out of sequence in the lists, GAs under review more than 7 days, noms > 1 month
A list showing nominator names and the number of noms, for anyone with > 1 nom
A list of categories with totals for each category -- e.g. "Military History: 12 noms"
A similar list showing untouched noms -- e.g. subtracting ones which were on hold or under review
A list of suggested categories for the backlog template to mention

I guess a subpage of the candidates page would be the place to put it. Historical data could be saved; might be interesting to look at trends.

As I said, these are just ideas. I'll post a note at the GAC talk page and ask people to comment here. Mike Christie (talk) 15:40, 16 April 2007 (UTC)[reply]

Simplest COI bot

I was thinking and I found out a very simple algorithm for finding conflicts of interest, all the bot has to do is search through the contributions of the current IP and look at each change(number characters different) to an article(counting multiple changes in a row as one) and compare it to the average change per article to all articles excluding the current selection, and depending how much different these two numbers are it adds it to a list of suspicion, i.e. the higher the difference the more suspicious and higher on the list, it also has to have a minimum change level so it doesn't catch people who just got into wikipedia and made their first edit(e.g. added a space). This wouldn't find people who make multiple changes to articles related to a subject and it might catch some people who didn't do anything wrong but it should only add those people to a list for someone else to look at so there isn't much damage to be done. Someone else has to make this bot because I can't. Jeffrey.Kleykamp 18:34, 16 April 2007 (UTC)[reply]

I don't think that would work too well. What if someone is just interested in an article/subject, but doesn't necessarily have a COI with it? ^demon^{[omg plz]} 18:54, 16 April 2007 (UTC)[reply]

That's why it looks at the average change to all other articles edited because it is unlikely that a person will only make huge edits to a single article related to a subject that they just like on a website that they know little about(i.e. they must have made more edits) so if the changes are small or medium then they are probably just fans but I'm talking about catching people who make huge changes to a single article without ever making any other significant edits, but I do see your point. PS: did this make sense to you because I'm writing funny today and I'm too lazy to check my work. Jeffrey.Kleykamp 19:22, 16 April 2007 (UTC)[reply]

Newest bot

I guess I'm in a bot creating mood because I have another idea, how about a bot that looks at the editing patterns of the edit that got people warned/banned and compares it to normal people to find people that are most likely to be vandalizing. The different patterns that it can look at is words/characters/phrases added, words/characters/phrases deleted, words/characters/phrases replaced, time between edits, size of edits, etc.

PS: In a sort of unrelated bot idea create a base bot that takes the commands of other programs written in any code as a sort of interface to Wikipedia so that people have to spend less time writing code the bots. Jeffrey.Kleykamp 13:23, 17 April 2007 (UTC)[reply]

AfD bot

I was looking through AfD, and I noticed that there were quite a few deletion debates with Wikipedia:Articles+for+deletion instead of WIkipedia:Articles_for_deletion. This would surely be easy to sort out using a bot; if anyone is interested then I think it would be a good idea. -- Casmith_789 (talk) 15:55, 17 April 2007 (UTC)[reply]

Does that even matter? I'm not entirely sure, but I think MediaWiki recognizes them both as one and the same. Jayden54 16:33, 17 April 2007 (UTC)[reply]