User talk:Jimbo Wales

(Manual archive list)

The User:Qworty matter

I have cited this statement of yours on Qworty's talk page [1], from January of this year, in proposing an indefinite ban/block of the editor at Wikipedia:Administrators'_noticeboard/Incidents#User:Qworty. You therefore may wish to comment. Hullaballoo Wolfowitz (talk) 16:25, 18 May 2013 (UTC)

On the Internet, nobody knows that you are being sarcastic. Robert McClenon (talk) 16:27, 18 May 2013 (UTC)

As Carrite says, this was good investigative journalism in the real world, that should be used to identify a Wikipedia editor or editors who have been deliberately and systematically misusing Wikipedia. ArbCom proceedings take a while. Can this editor, and possibly his sockpuppets be banned by Jimbo Wales, either indefinitely, or until the ArbCom can decide? Robert McClenon (talk) 16:27, 18 May 2013 (UTC)

Ironholds has indeffed for willful BLP violations, which seems reasonable. Carrite (talk) 18:09, 18 May 2013 (UTC)

A more serious concern is that Qworty was confirmed by a checkuser to be using sockpuppets in 2007. That check was initiated by a user who claimed that Qworty (who we now know is Robert Clark Young) was cyberstalking them. None of the accounts were blocked. Members of ArbCom, specifically User:Fred Bauder (although probably others as well) investigated another user's claims of cyberstalking at around the same time. Again, Qworty was not blocked. Jimbo's recent warning to Qworty has been mentioned elsewhere, but Jimbo made a similar warning to Qworty back in 2010. This is not a case of someone being sneaky and flying under the radar, this is a case of someone blatantly using sockpuppets, violating policies, coming to the attention of WP's most powerful and yet being allowed to continue unabated. This is perhaps the most concerning aspect of this case. Delicious carbuncle (talk) 21:10, 18 May 2013 (UTC)

Really? All of the dots were connected now, and yet it's somehow arbcom's fault that they didn't connect the dots (some of which did not exist back then, of course) 6 years ago? --Conti|✉ 21:28, 18 May 2013 (UTC)

The dots were connected 6 years ago. The user who initiated the checkuser request also posted this request for help in which she said "To whom do I report Robert Clark Young for Wikipedia violations? He has archived the talk page on his site so that none of the discussions are visible, re-edited, and deleted sources Under the name "Professor Ron Hill" (a name from our shared undergraduate days) ... Bob has a very personal grudge against me and I do not feel it is right that he is basically able to manipulate whatever article he wants on Wikipedia with the administration doing nothing about it". Elsewhere, she wrote "User Qworty is an editor who was a former boyfriend of mine against whom I have had a previous restraining order. In 2001, he sent me a Trojan horse computer virus. He has, under the name Qworty, repeatedly followed my edits on Wikipedia and maliciously changed them". The checkuser confirmed that Qworty had two other sockpuppets and identified a long list of possible socks. None of the accounts were blocked. All of this was known in 2007. Delicious carbuncle (talk) 21:53, 18 May 2013 (UTC)

Yes, it's surprisingly easy to judge situations correctly in hindsight, isn't it? I have no idea why the account was not blocked back then, given the positive checkuser results, but if anything, I doubt it would have been an indefinite block. First time socking usually doesn't get you an indef block. And we know now that the accusations you cite were true. But people make accusations of all kinds all the time around here, and it's usually pretty hard or downright impossible to determine the validity of such accusations. It would certainly be nice to have more people that thoroughly investigate every accusation made around here, but I'm sure you know as well as I do that that's downright impossible. Making this somehow arbcom's fault is just silly. --Conti|✉ 22:16, 18 May 2013 (UTC)

There's absolutely nothing silly about it. You have a case where an editor is making the assertion that Qworty is Robert Clark Young, their ex-boyfriend, against whom they claim to have had a restraining order, and that Qworty is editing against policy with that account and with sockpuppets. The editor filed a checkuser request. It was considered credible enough that a checkuser was done and sockpuppets were identified. Even back in 2007, I think it was common practice to block sockpuppet accounts, even if the main account remains unblocked. That was not done. For abusive sockpuppetry like this, an indef block is quite normal. That was not done. At almost the same time, another editor, who happens to be an admin, made similar claims about Qworty and stated they were in contact with members of ArbCom about it. Fred Bauder appears to have done some investigation. It is unlikely that Fred was unaware of the checkuser result and the claims it was based on. I have alerted him to this conversation and suggested that he comment here. I believe that people knew in 2007 that the allegations were credible and very likely to be true and did nothing about Qworty. No hindsight is required. Delicious carbuncle (talk) 23:31, 18 May 2013 (UTC)

I agree with Delicious carbuncle. Based on what is being said to have been reported in 2007, there has been a six-year lapse in sanctions. I would suggest that this episode illustrates that Wikipedia does a better job of dealing with "more obvious" problem editors, that is, blatant POV-pushers, uncivil editors and flamers, tendentious editors, and outright vandals, than it does with devious editors. Since editors are required to assume good faith until proven otherwise, devious editing stays under the radar, while it is easy to spot vandalism, and other obviously disruptive editing becomes apparent soon. In this case, the policy of assuming good faith is what allowed bad faith editing to go on for a long time. What I don't understand is how sock puppetry went unchecked. The process did not work, and should have worked at the time. The established Wikipedia processes are not designed to deal with long-term devious editing. They are designed to deal with sock-puppetry, when sock-puppetry is reported (which it was), and it isn't clear why they didn't work in that regard. Robert McClenon (talk) 01:08, 19 May 2013 (UTC)

The foundation is to blame for any libel this man may have perpetrated on this site. It is responsible for any libel that appears on this site. It hosts those pages and negligently allows anonymous editors to say anything they want to about living people, while knowing full well that our processes are (however well-intended) inadequate to prevent this sort of abuse happening. In a just society, it would be liable. --Anthonyhcole (talk · contribs · email) 02:05, 19 May 2013 (UTC)

While libel and responsibility are complex legal issues, I agree with the basic sentiment that Wikipedia and its editors exist in a curious limbo, where the project is seen as vitally important to society as a whole, and yet the typical checks and balances we require of central social institutions do not exist, or at least do not have the procedural clarity and strength we typically ask of them. This is a deeply important matter and we should all take it to heart and work to come up with more responsible ways of dealing with many kinds of problems that have come up recently. That we can read several people defending anything about the processes by which Qworty was allowed to keep operating on this site only digs the hole deeper. This is serious trouble for Wikipedia's reputation, by no means the first trouble we've encountered lately, and we cannot afford to have more incidents like it--reading some of the UK Parliament "Joint Committee on Privacy and Injunctions" proceedings , it is now conceivable that governments may start to get involved with this project at a regulatory or statutory level, and we should do whatever we can to put our own house in order to make that as unlikely as it would be unwelcome. Wichitalineman (talk) 02:57, 19 May 2013 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘I know we have a reasonable array of software tools for looking at content. I would humbly suggest that the available toolset for examining editor behavior is not as well developed. Personally, I think that is a problem, and one which is contributory to the problem being discussed here. -- [ UseTheCommandLine ~/talk ] # _ 08:19, 19 May 2013 (UTC)

The difficulty with Qworty's behaviour was that it was expressed over a pattern of edits, rather than in much that was itself sanctionable. There were times when the edits could have resulted in actions, such as when Jimbo placed his warning mentioned above, but generally what happened was that the warning would come from a different editor each time, without connecting the dots. This is something that software has trouble picking up. If someone had the full picture the community could have responded, but building that picture is tricky, and needed someone driven to do it. I'm glad that someone was, and perhaps off-wiki was the only way it could happen. (Which, of course, doesn't speak well of ourselves).

If the community does respond to this, and I hope it does, then I guess what I'd like to see is an effort to ask if our processes are sufficient for identifying difficult problems which aren't encompassed in a single edit. - Bilby (talk) 09:52, 19 May 2013 (UTC)

One of our guidelines may clash with this. If I read it correctly we are not allowed to use contribution histories to 'stalk' other editors. I came across a similar editor recently that is a borderline case so consensus would need to decide if their edits are POV. The edits aren't a violation but seem POV. Even mentioning the editor's name may be a violation. I may be wrong in this and if so let me know a way to bring editors like these to the attention of the community. One article did have other issues that I brought up at a notice board but that is as far I went with it.--Canoe1967 (talk) 11:16, 19 May 2013 (UTC)

That's part of the problem. Where a pattern of edits is the issue, you need to watch the editor carefully to connect the dots. Which creates a risk of harassment - whether real or just perceived. Or, in this case, if you believed that there was a connection between Qworty and Robert Young, you would have had a very hard time drawing it, as it would probably have required outing to make the case. Anyway, I'm not sure if the processes are wrong, or if they weren't used properly, or if the conflicting issues just prevent things from working, or if there are just things we can't do, but I'd like to know how to handle this better. - Bilby (talk) 11:57, 19 May 2013 (UTC)

I can see it creating a huge backlog at ANI if we change guidelines to allow contrib checks. Perhaps an IRC reporting method? I have noticed that when an editor is taken to ANI then their contribs are looked into. I don't even know if this is proper use of them. They are taken to ANI for one issue and everyone starts posting diffs that have nothing to do with the issue they were taken to ANI for.--Canoe1967 (talk) 12:13, 19 May 2013 (UTC)

I keep a close eye on the contributions of a few editors. I do not believe that I am violating any guideline by doing so, but I don't care if I am. Anyone is welcome to try to get me sanctioned for this. Good luck. Delicious carbuncle (talk) 14:19, 19 May 2013 (UTC)

There is no plausible suggestion that we do not "allow contrib checks". Policy says "The contribution logs can be used in the dispute resolution process to gather evidence to be presented in requests for comment, mediation, WP:ANI, and arbitration cases". --Demiurge1000 (talk) 15:06, 19 May 2013 (UTC)

I am probably just reading it differently. I read it as they should only be used in reviews of editors as listed above. If an editor is taken to ANI for POV then do we start listing all the diffs that aren't POV like 3RR, copyvio, or civility, etc? Or should we just take them to ANI for 'general violations' and list them all in the lead of the ANI?

Bilby, did you read my statements just above yours? In 2007, two separate users (one of whom was an admin) complained about Qworty. A checkuser confirmed that Qworty was using sockpuppets and ArbCom members were aware of the allegations of cyberstalking. All of that was sanctionable. Delicious carbuncle (talk) 14:15, 19 May 2013 (UTC)

I did. The problem is that there were a lot of similar events, all treated as separate issues. In each case, he could have been sanctioned, but people seemingly unaware of the previous cases decided to drop it back to a warning instead. My concern here is that the problems were seen in isolation. Once someone has a block log, or has a series of clear escalating warnings on their talk page, we can see the picture, but where the problem is scattered across a few years, with different people raising it each time, it isn't always the case that they recognise that this time something stronger needs to be done. - Bilby (talk) 16:04, 19 May 2013 (UTC)

They may appear to be separate incidents, but perhaps I should have mentioned that they were happening at the same time. I have no doubt at all that the case which sparked the checkuser was mentioned by the admin who discussed their own case with ArbCom members. User:Jpgordon, who ran the checkuser, was on ArbCom at the time. So here are two ArbCom members in 2007, Jpgordon and Fred Buader, with knowledge of alleged cyberstalking and proven sockpuppetry and abusive editing. It would be out of character for anyone associated with ArbCom to give a straight answer about why this resulted no action. Delicious carbuncle (talk) 18:36, 19 May 2013 (UTC)

Careful of jumping to conclusions there. I have no recollection of discussing the case with other ArbCom members, but it would have been entirely within my work at what's now SPI to simply take an open case, get the results being asked for, and pay no further attention to it. I wouldn't have blocked based on the SPI result -- I would have left that to another admin, as was my procedure at the time. I don't know why the checkuser request was closed when it was (before action had been taken on the findings); that situation should have led to an immediate block of the master. Straight enough answer? --jpgordon^{::==( o )} 19:09, 19 May 2013 (UTC)

Thanks for your answer. I hope that Fred Bauder has a better recollection of these cases. Perhaps someone could check the ArbCom archives? Or maybe those involved could look through their own emails from that time to refresh their memories? Delicious carbuncle (talk) 12:12, 20 May 2013 (UTC)

The claim of stalking has certainly been used as an attempt to avoid scrutiny. (Fae comes to mind.) But if WP is going to look after its own house rather than rely on people going to Wikipediocracy and/or the press, then this so-called stalking has to become accepted custom and practice here. As it is, not only do we have someone banned for linking a Wikipediocracy blog that names one of the Commons admins who Jimbo regularly clashes with here, but people are even suggesting that saying that Qworty=Young can still be seen as outing here. Indeed if Young had not decided to out himself here under pressure from Salon, multiple users in this thread would be subject to blocking for outing the squirt. For this site to live up to its obligations as the dominant English-language reference source, then we need a radical re-alignment so that our responsibilities towards our readers and the subjects of our articles are put at least on a par with those to our contributors, and active gaming of policy by contributors in order to avoid scrutiny should become a blockable offence.--Peter cohen (talk) 18:51, 19 May 2013 (UTC)

Qworty

"For those of us who love Wikipedia, the ramifications of the Qworty saga are not comforting". That sums it up for me. More thoughts soon.

I would have banned him outright years ago. So would many others. That we did not, points to serious deficiencies in our systems.--Jimbo Wales (talk) 20:34, 19 May 2013 (UTC)

This can't be the sum total of your response to this issue. Please weigh in Jimmy. InconvenientCritic (talk) 23:23, 19 May 2013 (UTC)

What about "the system" stopped you? Why didn't you?Dan Murphy (talk) 20:40, 19 May 2013 (UTC)

In past cases, he did: Often, when Jimbo has used his admin rights to block users or delete nonsense files, then several people have launched endless complaints, tirades, diatribes, potty tantrums (etc.) to flood message boards with numerous insults (even on other websites) to almost shutdown meaningful work. Meanwhile, you might think other admins would instantly come to block the haters, issue warnings, or redact insults, but no, its like no one left will help in a visible way, as if they, too, would become likewise hounded, as part of an imagined "Jimbo conspiracy" of tyranny to suppress anyone who is badmouthing "questioning" or insulting "correcting" the horrible intelligent, experienced, accomplished, and generous people in the world who "do not have consensus" to oppose the ideas of the uninformed. In fact, some admins move quietly to help (very quietly), but when Jimbo walks away from insults and ridicule seen as weak, he is probably providing the strongest leadership to avoid the hate-mongering that would lash back against his admin actions. In earlier years, Jimbo was directly involved, writing articles, inspiring massive fundraising, and setting major priorities to keep Wikipedia from total failure, and then people had the gall to refer to him as "Co-founder of Wikipedia" (just too funny). Extremely intelligent people can see how Jimbo has won amazing victories against what could have been extended petty battles with bad karma to waste his time, which he used instead to spread positive news worldwide, and hence, thousands of people see Jimbo and Wikipedia for those big, worldwide accomplishments, rather than trivial mudslinging fights. It might take a while to understand the social dynamics of those actions, but I hope I have offered some insight into the process. -Wikid77 10:40, 20 May 2013 (UTC)

jpgordon's comment above suggests that, for whatever reason, procedures that were (and are) normally followed, were not followed in one instance. --Demiurge1000 (talk) 20:48, 19 May 2013 (UTC)

I look forward to hearing your thoughts about this in greater detail, Jimbo. Even more so, I look forward to substantive actions. I was exceptionally disturbed by Qworty's intemperate and crude talk page rants and revenge editing a few weeks back, and if anything, more disturbed at the willingness of several other editors to defend and protect this gross misbehavior. Hand wringing is not enough, and disruptive editors must be identified and have their editing privileges removed if they are unwilling or unable to comply with our policies, with a special emphasis on BLP. Cullen³²⁸ Let's discuss it 21:49, 19 May 2013 (UTC)

The "fight the Man!" instinct around here is strong. No matter whom is banned, there is a vocal group of people who will argue against it. If action is taken by ArbCom, they will be accused of star chamber-type justice--with side accusations of favoritism, capriciousness, or ideological motivation to boot. Qworty's problematic behavior was brought to Arbcom's attention six years ago... yet here we are, only now having given him the boot. Jclemens (talk) 22:32, 19 May 2013 (UTC)

The last time Jimbo acted to ban an editor with a conflict of interest, ArbCom overturned his action with extreme prejudice (see Wikipedia:Arbitration/Requests/Case/TimidGuy ban appeal). That context may explain how "the system" constrains him from acting. MastCell ^Talk 22:50, 19 May 2013 (UTC)

Jimbo honestly has two choices—create stronger institutions on Wikipedia such as an elected assembly, content review boards, and professionalization of the admin force (with an admin review subcommittee of ArbCom hiring, sanctioning, or dismissing admins), rather than handing the encyclopedia over to the mercy of corrupt, self-serving cabals; alternatively, he can reassume the role of a god-king. I would prefer the former. Wer900 • talk 23:06, 19 May 2013 (UTC)

That would be because Jimbo's ban was wrong on a number of different levels, MastCell. You'll note divisions in the committee over whether and how Will Beback was to be sanctioned for his behavior in the matter, but there was not a single dissenting voice among the committee of the time who said that Jimbo's ban was correct. Jclemens (talk) 23:40, 19 May 2013 (UTC)

• Mr. Wales contacted Qworty at his talk page on two occasions (both times in regards to severe BLP violations): in 2010 and in 2013. So, Mr. Wales, could you please explain what serious deficiencies in your systems have prevented you personally from banning Qworty ? Thanks. 76.126.142.59 (talk) 01:49, 20 May 2013 (UTC)

We have policies and institutions to handle these sort of situations, insufficient and flawed though they may be. Jimmy Wales has enough on his plate not to be expected to run around as judge, jury, and executioner. Moreover, none of us should WANT a system in which one individual has such power. The failure to stop Qworty from malicious editing is a failure of all of us. And this failure was exacerbated by Wikipedia's unhealthy worship of editing secrecy and its failure to install mechanisms to halt the ability of one person to start and use multiple accounts. Along the latter line, one good idea I heard recently was that WMF should unilaterally begin including IP addresses in the signature of each post at WP. This would serve as a red flag on multiple accounts being used in close proximity to one another by a single editor. In the long run, Wikipedia needs real name registration and sign-in-to-edit mechanisms to further limit the use of multiple accounts and to make sure that content can be attributed to a real life individual — which would incidentally slice vandalism massively. Carrite (talk) (Tim Davenport, Corvallis, OR, USA) 06:00, 20 May 2013 (UTC)

That is certainly one step in the right direction. However, I think a broader issue is that the rules have been created by editors for editors and they therefore receive more consideration than either the readers of our articles, who want balanced and accurate information, or the subjects of our articles, who do not want to be libeled or to be represented unfairly. Consideration for Qworty in his capacity as an editor ahs meant that our readers and the people he hates were ill-served. Of course, now he has been declared a non-person, he will himself be receiving the same loving care which his enemies did from him. Commons is particularly extreme in looking at ways to ban those who expose multiple copyright breaches by members of the in-crowd and in banning the whistleblowers, but the whole span of Wikimedia projects need a whistleblowers' charter that protects those who seek to protect stakeholders outwith the editor/admin core. --Peter cohen (talk) 15:30, 20 May 2013 (UTC)

(ec) I do (want a system where one individual has such power). Essentially 100% of other successful organizations that produce reference works have such a person. They are called Editor-in-Chief. Having one here would not impinge on the community aspects of creating content and deciding most issues. Jimbo decided not to have one, so for better or worse we don't and won't. Instead he have a rather nebulous and very vexing and labor-intensive system. We have to make it work as best we can I guess. We need to accept that situations like this will likely arise, again and probably forever. It doesn't mean the Wikipedia as a whole doesn't work, though.

Requiring editors to use or link to their real identities would mean the immediate exit of me and many editors like me. I have standing in the community, vulnerabilities, dependents, and so on. I can't get into a real-life pissing match with someone who lives in his mom's basement and has nothing to lose. Phone calls to my employer and so forth are not part of any deal I want to be a part of. Requiring editor identification would basically allow the participation of two groups: the truly strong (who have resources to engage lawyers and PR men, tenure or independent means or other secure income or position, many friends, a secure pubic reputation, and whatnot) and the truly weak (who have little to lose, no reputation to tarnish, no assets to seize, no job of any importance to lose, and whatnot). Many many everyday people like me would go, and that at once. Whether that'd be worth I don't know. I don't think so. Herostratus (talk) 15:42, 20 May 2013 (UTC)

I wonder what a pubic reputation might be... Pgallert (talk) 21:45, 20 May 2013 (UTC)

By "public reputation" I mean a reputation with at least some segment of the general public, rather than just with one's friends, neighbors, colleagues, and so forth. A whispering campaign, or letters to his hometown paper, or calls to his employer, or emails to his neighbors, or anonymous threats to his relatives and yadda yadda, by some small collection of random mooks, is not going to much bother say Paul Krugman. They would me. Herostratus (talk) 01:49, 21 May 2013 (UTC)

Let's fix edit-conflict problems

We know how to fix them, and edit-conflicts in section-based editing are already auto-corrected. So, let's fix more problems, and the more we discuss the issues, then perhaps we can reach consensus on how to merge multiple changes. The rationale is simple: as more people join Wikipedia, then more edit-conflicts are likely, and we need to prepare to reduce them before the crowd arrives. I think we should just start talking about various scenarios, and suggest patterns to auto-fix the edit-conflicts. Some scenarios (numbered as "Sn") to consider:

• S1: Two replies after same message: I think this might be the most-common fear, and I suggest to insert the 2nd reply after the 1st reply, so the logic would be to insert an edit-conflict reply not exactly after line n, but rather, insert the 2nd reply before old line n+1. In practice, it is often easy to spot the original line n+1 after the message, and hence to auto-correct, then insert reply_2 after that message+reply_1.
• S2: Two edits change the same line: A simple fix would be to treat a line as two halves, and simply combine changes to each half deleting/adding text, else the latter change overrides the first change. However, I think, with a little analysis, the removal/addition of text could be combined in subtle cases, as perhaps matching 10 characters either before/after the changed text, else use the line contents from the 2nd edit (as replacing that line).
• S3: Line(s) deleted during 2 edits: When the 1st edit deletes the lines, then changes to those lines in the 2nd edit should be ignored, but conversely, lines deleted by the 2nd edit take precedence and should be removed. As in S1, new text lines (or replies) should be inserted before line n+1, after the deleted lines. To re-sync with the original line count, then deleted lines should perhaps be considered as empty text at those old line numbers, until the page is saved. The software might need to have extra internal line-counters to skip the deleted lines after combining the texts.
• S4: Numerous lines differ: In cases where perhaps, 50-100 lines differ, then I am wondering if the software should issue a warning, where the text might actually be editing a different page.
• S5: Changes could be interpreted in 2 ways: In some cases, the software might judge a change as 2 different possibilities of changes, and in such cases, perhaps look-ahead and choose the simplest path which avoids multiple differences after that point. Often the easiest solution is to merely count the number of "unchanged" lines after that point, and the "right" path will have far fewer mismatched lines.

In the initial analysis of edit-conflict problems, it can be easily seen that many talk-pages involve mainly the accumulated additions of multiple replies, as perhaps the easiest type of auto-correction during an edit-conflict. So then, merely retro-insert the 2nd reply, as inserted before line n+1, to appear after the 1st reply. In fact, the auto-correction of talk-pages might use some different (simpler) rules, as compared to the auto-correction of article edit-conflicts. The more we discuss these various issues, then the more we can find a consensus where the 2nd edit should override the first, or perhaps the auto-correction should warn of numerous changes, as if trying to edit after vandalism, or hack edits, which should most-likely be halted, to allow totally reverting the prior edit, not auto-merging of the changes which might conceal the prior hack edits. The auto-correction could be kept simpler, at first, then improved later. For example, by treating long lines of text as "auto-lines" or sub-lines of 50-character segments, then the auto-correction could be treated as simple line-for-line recombinations of changes, even with thousand-character lines, if that were easier to verify. Start with simple ways to auto-correct the text. Anyway, let's fix those many, many simple edit-conflict problems.

Conclusions so far: Most of the talk-page edit-conflicts are very easy to auto-correct, and I see no excuse for having edit-conflicts in busy talk-pages. It indicates a serious failure to correct a very common, easy problem, as perhaps due to fears about more-complex edit-conflicts within articles. So, let's fix them separately, not let fears of complex cases then paralyze the fixes for simple cases. -Wikid77 23:56, 19 May 2013 (UTC)

Further considerations about edit-conflicts: Some areas of Wikipedia have numerous edit-conflicts, such as the wp:Help_desk, where I have tried to assist in answering questions several times, but the edit-conflicts seemed to happen in "85%" of all help-replies (although edit-conflicts in 55% might be more accurate). Anyway, it did not take long to realize how answering the questions at the Help_desk involves redo for numerous edit-conflicts. Now it might seem the edit-conflicts merely show, hey, "that question was already answered" so just focus on the next question, but in practice, very many Help_desk questions could have 5 alternate answers, and so edit-conflict does not really stop duplicate answers, it typically rejects alternate solutions to those same questions. In short, the "only good edit-conflict is a vandal being thwarted". As for new articles about a major event, or recent tornado, or shipwreck, then I dread trying to help the editors update those articles, and I feel sorry for their edit-conflict suffering. -Wikid77 04:11, 20 May 2013 (UTC)

Further analysis of edit-conflict pages: Fortunately, for years, many people have been logging the actual incidents, of each edit-conflict event, by embedding the "{{ec}}" template (which says: "(edit conflict)") into many of the affected pages. So now, there is ample edit-conflict data to analyze, in WhatLinksHere/Template:ec as "9,918" links (10 thousand pages). A quick breakdown of the first 3,000 edit-conflict pages: 31% article-talk pages, 22% user-talk (like this page), 13% Wikipedia-talk (about policies or guidelines), 5% help-desk, 5% reference-desk, 2% village-pump edit-conflicts, but I will check more edit-conflict pages and refine the counts somewhat. Anyway, 31% (as article-talk conflicts) then means 69% of edit-conflicts were logged in user-talk (this page), policy-talk or other forums. Of course, people cannot mark live articles with "{{ec}}" even if the article main-namespace hid the message "(edit conflict)" then other editors would continually remove the "ec" tags cluttering the text. I think to simulate the actual edit-conflict of article "xx" then check for "Talk:xx" in the WhatLinksHere, because the edit-conflict levels are likely to be similar, where conflicts in talk-pages would tend to reflect conflicts in the related article text. More later. -Wikid77 04:11/10:40, 20 May 2013 (UTC)

• Current support does Edit-merge but also Edit-clobber: To double-check the current handling of edit-conflict scenarios, I have re-tested as two editors concurrently changing top or bottom of the same page, where edit-conflict does auto-correction to edit-merge different areas of a page. So, the good news is that potential edit-conflicts were properly auto-corrected as edit-merge when the changes were in separate portions of the page, so a prior edit to page-top (infobox format) was then edit-merged with changes to page-bottom, as combining both edits. However, trying to re-edit the page-top (infobox) was reported as "Edit-conflict" until those attempted changes were removed, and then the SAVE was accepted as auto-correction to edit-merge the top/bottom of the 2 conflicting editors. However, when the same username ran 2 edits together for the page, then the 2nd edit (in conflict with infobox changes) caused an implict override, or "edit-clobber" of the first saved edit and removed all prior changes to the infobox, with no warning. So, that is a condition to beware, if a user edits the same page from multiple windows, then a 2nd edit might override the first edit (not warn edit-conflict), so be very careful to re-run a 2nd edit using the current revision. -Wikid77 16:23, 20 May 2013 (UTC)

• • I agree but as long as long as the priority is on fixing things no one cares about like the Login page, forcing the notifications or killing off the Orange Bar of Doom there is little chance at this happening. I find it to be a real pain on active discussions when I have 3 or 4 Edit conflicts in a row and sometimes I just give up and go away until it slows down. Kumioko (talk) 19:34, 20 May 2013 (UTC)
    • If more people keep noting the edit-conflict problems, then the developers could see them as a priority, such as noting "3 or 4 edit-conflicts in a row". -Wikid77 22:22, 20 May 2013 (UTC)

I expect that part of what is going on is that the developers don't want to spend much time improving things that will be made obsolete in any case when Flow comes online. One of the central goals for Flow is to complete eliminate edit conflicts on talk pages. Looie496 (talk) 19:58, 20 May 2013 (UTC)

The clear advantage of fixing current edit-conflicts, before installing Flow, is the fixes for article edit-conflicts not just talk-pages only. -Wikid77 22:22, 20 May 2013 (UTC)

I admit I just read the link to Flow just now and so am new to the idea of what is being planned. However, based on what I read there I see the following being a rallying call in the future "We aren't Facebook!"; as that was a common compliant with what replaced the orange bar (which I always thought was a shade of yellow personally...). And so is Flow really going to be a solution that the Community can or will accept? We shouldnt put off improvements because a comprehensive solution to ALL is "around the corner". If scientists said they would stop working on the issue of what dark matter and dark energy were because they were SURE that a unified theory of EVERYTHING was around the corner and therefore solve what those issues were anyways, well those scientists would be waiting for a long time twiddling their thumbs.97.88.87.68 (talk) 20:15, 20 May 2013 (UTC)

• Old stuff remains years longer as people oppose change: The value of fixing the old talk-pages can be seen by problems in other new-talk products, such as wp:LiquidThreads which is slow, rambling and had lacked the view-history of talk-pages. Hence, many people will likely support fixes to edit-conflicts now. A common complaint about LiquidThreads was inability to reply to multiple comments at once, or search within replies, where users were forced into an "opinion forum" where each message replied tediously to only one post per edit, as laboriously stretching a 5-person discussion into hours of cumbersome chained replies, many of which were likely not read as chained too far from original comments. However, unless people knew how much faster the old talk-pages were, where a single edit can post replies to 7 comments, then the hours of tedium of chaining replies in LiquidThreads was not immediately understood as boredom mired in a lengthy snail trail. A typical 20-line talk-page dialog with 12 replies could ramble across 3 or 4 screens of message-reply boxes. Truly, hours and hours were needed to post all replies, and then re-post to cross-update the replies, chained across a bloated network of multiple separate message boxes. -Wikid77 (talk) 22:22, 20 May 2013 (UTC)

Fix the problem not the blame for edit-conflicts

I am adding this edit-conflict topic, as a separate thread, to vent any ill-feelings about not fixing the trivial edit-conflict problems, as easily possible, all those many, many years ago. It can be very easy to get hopping mad, once truly realizing how very easy the talk-page edit-conflicts could have been auto-corrected. However, as the adage states, "Fix the problem not the blame" because we could spend more years finger-pointing who to blame for the misery, or trying to rationalize why the edit-conflict problems were "deliberately left unfixed" (or so it seems), for years and years, as thousands of editors (at least "9,918") were suffering the humiliation of recopying or retyping entries, like busywork for some antiquated school official demanding to complete the form as hand-written, in triplicate (no machine copies). Surely, numerous clever people had seen the edit-conflicts over the years, and then concluded this "Version 1.0" software would be updated soon, but later rolled their eyes and left. However, I will let others express whatever justifiable anger, below, but meanwhile, we should re-focus on emphasizing the time is now, the time has come, the angry mob is at the gates, and we need to fix the simple edit-conflicts soon. Yet I would like to hear other horror stories about past edit-conflict nightmares, such as most edit-conflicts in a row, or what people did after the umpteenth edit-conflict pushed them to the brink, and I think people's pain needs to be acknowledged, but if this thread gets out-of-control, then feel free to hat whatever parts of the discussion. Also, as a point of overall perspective, the editor community also shares the blame in not helping to resolve the scenario of 2 adjacent replies, where we need consensus to resolve the edit-conflict by deciding a 2nd reply should be appended after the 1st reply. -Wikid77 (talk) 04:11/16:23, 20 May 2013 (UTC)