Wikipedia:Administrators' noticeboard/Incidents/User:Jaguar

User:Jaguar

 * Live discussion moved from archive 732.

Before departing, retired User:Jaguar created many articles with malformed ledes and infoboxes, (as seen in a search for the diagnostic string "Jaguar/Sandbox/3" and ), presumably with a malformed script or bot. Over 100 (but under 250) exist. Those articles, and other, more recent examples without the aforesaid malformations, also include the text "(Chinese: ?)" as shown, including the question mark. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 20:01, 17 December 2011 (UTC)
 * I fixed forty, and there are 82 left to do. --Dianna (talk) 08:08, 18 December 2011 (UTC)
 * I fixed forty-eight, and can't find any more in mainspace. Rich Farmbrough, 20:24, 18 December 2011 (UTC).


 * Thanks, Rich. I did 34 more this morning, so it looks like the problem is resolved. --Dianna (talk) 21:21, 18 December 2011 (UTC)
 * I must apologise for my actions that I have done a couple of months ago. I'm afraid that I don't use Wikipedia anymore and I only will return for emergencies such as this one. By the way I didn't use a script or bot, I used to create articles manually. Anyway, thanks a lot for your help! Jaguar (talk) 17:06, 19 December 2011 (UTC)
 * I have looked at many of the stubs that User:Jaguar created after this discussion, and many of the ones I looked at have multiple issues such as: reference urls's that don't point anywhere, malformed reference url's, reference url's that point to a website as oppossed to pointing to the page inside the website that talks about the subject, internal links that are wrong, reference titles that are wrong.
 * Also I don't know if the (Chinese: ?) thing is an issue or not, but they all have this.
 * In my opinion, there is no point in replacing a red link with a stub that doesn't say more than the title and contain things that are wrong. Let alone doing this 10,000 times. Azylber (talk) 10:25, 17 May 2012 (UTC)
 * Can you provide examples of articles where there is still a problem, please? Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 11:07, 17 May 2012 (UTC)
 * Yeah, please could get me an example so I can look at it and hopefully fix it? I've checked many of my new articles and references work just fine. Thanks, Jaguar   (talk)  15:13, 17 May 2012 (UTC)


 * An example? Let's go to List of township-level divisions of Heilongjiang and start from the very top: the Tongcheng Subdistrict link in Acheng District. It takes us to the page that reads, in its entirety: "Saiqi (Chinese: ?) is a township-level division situated in Ningde, Fujian, China". So is it Saiqi or Tongcheng, is it in Ningde or in Acheng (part of Harbin Prefecture), is the province Fujian or Heilongjiang? A few more items look "OK" (as in, "no useful info, but no absolutely misleading info either"), but then in the 3rd line we have Daling Township whose article has a link to the List of township-level divisions of Hainan in its "See also" section. Obviously I am not going to inspect more than a few stubs - I usually run into them when I need to do something useful - but a good round of quality control seems to be in order here, before more stubs are to be created. Again, I am not against the creation of a large number of township articles per se, but I'd like them to be generated at least at the minimal information level that one can see at zh.wiki. Over there, they had a a bot create them all, and the bot was doing it based on some kind of CSV file with quite a bit of basic information, such the correct county assignment (with the appropriate county-wide category), the list of villages within the township, geographic coordinates, and even the national identification number (zh:中华人民共和国行政区划代码 - something that each township apparently has). -- Vmenkov (talk) 17:08, 17 May 2012 (UTC)

Thank you for your concerns. I will do my best to address them later on in the week as I am busy for the next few days. I would like to point out that I simply start these stubs so that any user with the knowledge of that area of China can expand them and contribute to them. There has been a mass creation of red links and naturally red links cannot sit there forever, so I took up the task of making those red links blue. It's a feat that improves the encyclopedia, adding some base articles, as of all, we're here to build an encyclopedia, not to finish it. Many thanks Jaguar   (talk)  15:19, 17 May 2012 (UTC)


 * Ok, the first thing I'm going to say is: I'm going to list lots of errors here that affect thousands of articles, so I hope nobody takes this personally, ok? I'm just concerned about the quality of the encyclopedia. Please don't take this personally.

For example, look at this stub: http://en.wikipedia.org/wiki/Jinsha_Subdistrict

Here are some of the errors present in this stub, which are also present in hundreds of other stubs Jaguar created:

This error exists in a large number of articles. Does this break the policy on creating lots of unreferenced stubs?
 * 1) URL references that are wrong. For the stub we're looking at, the URL for the reference is http://www.xzqh.org/html/gu/ which does not exist and as far as I know never existed.

This error exists in a large number of articles.
 * 2) Internal links that are wrong. For example, in that same article, look at the link that says "township-level division". Instead of taking you to the list of township-level divisions of Guangdong province, it takes you to the list of township-level divisions of Fujian province.

So again, introducing information that is wrong. This error exists in a huge number of articles, ranging from March to right now, for example this one created yesterday: http://en.wikipedia.org/wiki/Guanfang_Subdistrict
 * 3) Cite titles that are wrong. For example, in that same article, the reference given (which by the way, takes you to a page that doesn't exist) also has the wrong title. It says "福建省", which means Fujian province, when it should say Guangdong province.


 * 4) The article says "(Chinese: ?)", which I don't know if it's against the policies or not, but some people have complained. In my opinion, a stub that says nothing more than the title doesn't say much. If you could at leave give us the Chinese name, you're adding something that's not on the title.

http://zh.wikipedia.org/wiki/%E9%87%91%E7%A0%82%E8%A1%97%E9%81%93
 * 5) No interwiki to the Chinese wikipedia, even though the article exists in the Chinese wikipedia.


 * 6) He was told about some of these errors in December at WP:AN and numerous times since February on his talk page and he didn't fix them. Instead, he chose to go on to create thousands more stubs, with the same errors.

http://en.wikipedia.org/wiki/User_talk:Jaguar#Jiangwan where he mentions a province and calls it a city, a county and calls it a district and so on. He blames these ones on errors that exist in other pages, but obviously when you create a new article you have to verify what you're writing, right?
 * 7) Errors like the ones pointed out here:


 * 8) He was asked on numerous ocasions by numerous users to slow down and check the errors in his existing stubs before creating thousands of new ones. I think it's important to listen to that advice.

I think I'm probably missing a few other errors in some batches that I haven't reviewed, but this should be enough to show what the situation is.

Whether or not creating thousands of stubs is a good idea or not has been debated many times and I don't want to enter that discussion, but I think a one line stub that contains errors is definitely a minus and not a plus, because it's misleading and also because it takes longer to fix it than to do it right at creation.

Finally, if you look at the notice at the top of Jaguar's talk page, it says that if you report these issues he will give you one of these: http://en.wikipedia.org/wiki/Finger_%28gesture%29 I think this is not constructive.

Again, I hope nobody takes this personally. Azylber (talk) 15:13, 17 May 2012 (UTC)
 * It's OK don't worry! I didn't take any of that personally. Can I point out to you that there are actually fewer mistakes than you think:


 * 1) These URLS are broke because the Chinese website went down at the time and that is entirely not my fault. I will find a new link and will correct them using AWB if you want.
 * 2) Yes, those are plainly my mistakes that I have made when creating these articles and I knew that I have done them. I fixed a lot of links in the past when I had found out that I had made typos in User:Jaguar/Sandbox/3. A few more might exist, but not as much as you think! :)
 * 3) Again, a typo. Like above I speedily corrected some of them when I found out that I had forgot to copy and paste in extra words.
 * 4) That is there for a reason. The question mark is fine! If I were to look up every single one of those Chinese symbols it would take me half a century to start these articles!
 * 5) I will add a interwiki soon.
 * 6) That's misleading. I did fix any articles I found problems with in December, before I retired.
 * 7) I just follow the lists on what I'm creating on. If there is a province, I put it in the article expecting if it is correct. I had no idea that they could be anything else like prefecture-level cities and so on!
 * 8) I didn't create thousands more, I've stopped right now.
 * 9) I've removed that from my talk page.


 * I will be busy for a few days, which means that I can't correct them just yet. I've just left school for the final time today and said my goodbyes to everyone, so I'll be busy at the moment. I can say that I feel guilty about all this. Please don't look at me like I'm selfish or not considering Wikipedia. I will do anything to put myself in ANI's good books, but I can't today. Thanks, Jaguar   (talk)  15:48, 17 May 2012 (UTC)


 * Yes, you did create a very large quantity of articles containing errors after you were told on numerous occasions. So please don't say you didn't know.
 * I'm glad that you have at least removed the "fuck you" gesture at the top of your talk page threatening anyone who reported these issues. Azylber (talk) 15:56, 17 May 2012 (UTC)
 * It was intended to be a joke and not taken seriously. Please, I'm getting the impression that you're trying to get me into trouble. Jaguar   (talk)  15:58, 17 May 2012 (UTC)
 * Like I said 3 times, this is not personal. I'm not trying to get you into trouble, I'm concerned with what you're doing, despite having been told many times by many people.Azylber (talk) 16:06, 17 May 2012 (UTC)
 * Good, perhaps we should continue at Jaguar's talk page? We can resolve this fairly easily I'm sure, there are a few more wrinkles that need smoothing out. Assistance from someone with strong Chinese reading skills might be an advantage.  Rich Farmbrough, 16:29, 17 May 2012 (UTC).


 * I'm not sure that continuing in his talk page is enough. Many have told him about these things for months, and what he's done is make up excuses, leave all the errors there, and create thousands more stubs with the same errors.
 * I think perhaps some policy could come out of all this, because all this mess will take a lot of work to fix.Azylber (talk) 16:49, 17 May 2012 (UTC)
 * WP:MASSCREATION? --92.6.200.56 (talk) 17:38, 17 May 2012 (UTC)
 * Azylber, I am listening to all your concerns and I am taking in the comments. I am not ignoring them or making up excuses. There would be no need to go off creating new policies on stubs because there is already enough! If I'm creating stubs for a good cause and if they have at least one suitable reference, then there should be no problem. We are here to build an encyclopedia, not to finish it. Jaguar   (talk)  18:05, 17 May 2012 (UTC)
 * I'm more interested in fixing up issues than worrying about policy. If Jaguar is keen to do as much of that as he can (and I understand that motivation) then his talk page seems a good place to coordinate resolution. Rich Farmbrough, 19:12, 17 May 2012 (UTC).

Do you have any idea of the scale of the issues—is it as big as this, or this? I clicked on the "Jinsha Subdistrict" example above; the amount of pages Jaguar created in the following minute alone is eleven. That's a new one every 5.4 seconds. I have no idea if that was a particularly slow minute. The single reference on each is a googletranslate link. --92.6.200.56 (talk) 19:48, 17 May 2012 (UTC)


 * I fixed the "Jinsha Subdistrict" — Preceding unsigned comment added by Eskimopie300 (talk • contribs) 01:22, 23 May 2012 (UTC)


 * I'm getting here a little late to the party apparently, since we have timestamps from 2011 up there... perhaps some formatting considerations (and a descriptive title) would be called for in future notices.


 * Anyway, I'm one of the editors that suggested jaguar slow down. He indicated on the talk page he's made over 10,000 of these stub-type pages... and the creation rate is astounding. I'm not doubting that copy-pasted into chrome and did it that way, but whether we wikilawyer over what semi-automated means or not, the Bot guidelines are very clear for large semi-automated article creations, and this is a textbook version of that. We have policies on hand. Let's please use them.


 * Massive stub creations in batch (and i mean massive) are not helpful, and they create way more work to our editors than they provide knowledge to our users. I don't think jaguar means ill in any of this, but it needs to be clear that there's no glory in making hundreds of pages generated out of a table.


 * What I would like to see is a consensus that this sort of mass creation, particularly when it's so full of errors (that thankfully people have caught... I shudder to think how many we don't catch), needs to be limited in the least, and that the BAG guidelines are followed, in Jaguar's case specifically, but also more generally. Shadowjams (talk) 22:08, 17 May 2012 (UTC)
 * Shadowjams, I agree with everything you say, it's exactly my same opinion.
 * What I would like to know is who is going to fix all this mess. Thousands of articles without references (a URL that points nowhere or that points to the wrong place is not a valid reference), with internal links pointing to the division list for the wrong provinces, with cite titles that are wrong, without the interwiki link etc etc. It will take a very long time to fix all this, much longer than it took Jaguar to mass-create all these stubs. Are we going to spend the time it would take to fix all this? Is it worth it? We could simply mass-delete them. Or, we could leave them there, trashing the quality of wikipedia.
 * It's also worrying to think of how many we don't catch.
 * I also want to know what is going to be done to prevent other people doing this in the future.
 * Azylber (talk) 22:29, 17 May 2012 (UTC)
 * Well for systematic problems, like the ? in infoboxes, I can help Rich do those with AWB if he wants (because rich is under a bit of a restriction on that I think), but Rich has been very helpful in offering advice about fixing those. If Rich wants to contact me about some of those things I can run I'd be happy to. I have a high level of experience with regular expressions.


 * My bigger concern is accuracy related. I don't know anything about the subject of those articles, and I certainly can't dig deeply through those lists. But, if there's stuff that just needs a hammer to do in order to fix it, let me know on my talk page. Shadowjams (talk) 23:04, 17 May 2012 (UTC)
 * I was involved in a similar situation about a year ago, though on a much smaller scale; an editor was attempting to provide similar information about localities in India (though in aggregate articles rather than individual ones), and they were similarly unsourced or undersourced. One of the ANI reports can be viewed at Administrators' noticeboard/IncidentArchive696. The end decision was to delete most of the articles he had made. The most relevant Afd is Articles for deletion/List of villages in Haryana. The argument I made there, and would probably make here, is that these articles, if unsourced, are actually harmful, and not a part of the incremental step of building the encyclopedia. If we know that a reasonable number of them are wrong, and have no reason to believe that they rest are correct, then it's actually more work for an editor who wants to make these articles to edit these than it is to start from scratch. That's because first they have to look into the existing article, and get confused (wait, is this about a different village with the same name?); then they may have to backtrack to the list articles and fix those. I haven't researched the details above, but if this is a regular, wide-ranging problem, mass-deletion is actually probably a better fix than anything else, unless there is currently another editor who has an accurate almanac who is willing to commit to fixing them relatively shortly. Qwyrxian (talk) 03:06, 18 May 2012 (UTC)
 * Mass deletion is not the answer. That is the most upsetting thing I've ever heard. That would mean hours of my work would be gone, all for nothing. Listen, I can fix most of those issues. Rich Farmborough is doing the right thing by making a list of solutions and I will use those solution! I would also like to point out that the whole issue everyone has made here is not as serious as you think. Everyone in this ANI discussion has just pointed out every single bad detail of my Wikipedia career, to be honest. Also, the number of Chinese townships I created is actually not 10,000. It's probably around 8,200+. 10,000 is the total number of articles I've created. And to be honest I know that it sounds a lot, but in truth it isn't. Other uses have created much more the 10,000. Say Dr. Blofeld has created 80,000!


 * Please don't take this discussion too far. I am going to do everything I can to fix these issues. I expect every single article to be kept as they are each notable enough for its existence - it's a Chinese town somewhere in the world! Jaguar   (talk)  08:56, 18 May 2012 (UTC)


 * The difference between Dr. Blofeld's stubs and your stubs is that yours are full of errors and therefore do more damage than good.
 * And let me remind you that this discussion wouldn't be taking place if you hadn't ignored the warnings that many people gave you for months on your talk page and welcomed us all with a fuck off gesture that you have removed now that this came to light.
 * If you're going to sit down and fix your 10,000 full of errors stubs then it's fine. Otherwise they should be mass deleted because like several people pointed out, they do more damage than good. And it doesn't matter how much work you put into it, what matters is Wikipedia. If you chose to continue working for hours making more stubs with errors after you were told many times, that is only your fault.
 * I think you should stop making all these excuses and start fixing. Azylber (talk) 11:43, 18 May 2012 (UTC)
 * Better yet, someone revoke their autopatrolled rights. Blackmane (talk) 13:21, 18 May 2012 (UTC)

You say there's "no need to go off creating new policies on stubs because there is already enough", but it doesn't appear you've taken notice of the existing ones. You had Autopatrolled status revoked in late August for creating dozens of unreferenced stubs, then asked for it back 3 weeks later "I have mass created over 200 articles and each and every one of them has a suitable reference". If you've mass created 10k, that's 9,800 since last September; 90% of them on Chinese townships. You say you'd been authorised to do the mass creations, as is required, yet when asked for a link to the discussion you gave a link of you re-asking the admin for autopatrol. That isn't soliciting community input nor a proposal of any sort.

Your userpage has an ANI comment linked where you say you created over 100 pages in six minutes. Faster than one every 3.6 seconds. It's directly above: "To do list: 1. Create every township in China, 2. Get to #10 on List of Wikipedians by articles created". A current WP:BON discussion has highly experienced admins & members of the Bot Approvals Group (see WP:MEATBOT) saying even the simplest bot shouldn't exceed 1 edit every three seconds because sometimes bad edits are made and it can take some time to fix/check. And that's talking about approved bots doing a minor activity.

Problems with the substubs containing temp sandbox titles were raised in late November ; you continued creating en masse, the last one six days later - Hongxing Township, placing retired shortly afterwards.  You unretired in the new year with the first edit summary "Nobody's gonna push me about", adding "I have returned - but only for a limited time. This time no crackpots at ANI [shortly after changed to nobody] are going to push me about, I'm gonna get this job done once and for all." Your very first edit outside userspace was to resume mass creating with Chengbei Subdistrict, Beijing—which still contains "ENTERHERE". Two in that same minute, fourteen in the following minute continuing that day, and the next and so on, into the several thousands.

The downplaying the issue as "not as serious as you think" (How can you know?) or pledge to do everything you "can to fix these issues" (Suddenly learn to read Chinese?) is what's troubling. Despite you saying this morning "There are no more errors. That's the last of them.", the Chengbei article alone shows this is untrue. The rate at which they're made means mistakes, yet inability to understand the foreign-language source hoping on gtranslate of an Asian language seems the fundamental problem as Azylber and Vmenkov showed above. --92.6.200.56 (talk) 14:07, 18 May 2012 (UTC)

I've removed Jaguar's autopatrolled (again). That is the bare minimum that is required here given what evidence suggests is an ongoing inability to trust that his stubs meet the bare minimum requirements for content level and correctness. That doesn't mean this should be closed quite yet. Chris Cunningham (user:thumperward) (talk) 14:46, 18 May 2012 (UTC)
 * Would anybody else like to point out anything bad about my Wikipedia career? How about taking this matter that didn't start off so serious much higher? I'm going to fix these myself since this situation can't get any worse. To be honest I think everyone's jealous that I can contribute to Wikipedia by expanding knowledge and not sticking around ANI all day bullying people into self pity. Jaguar   (talk)  15:36, 18 May 2012 (UTC)
 * No one is trying to crucify you. You were creating hundreds of stubs that had errors. You have the view that creating hundreds of error-filled stubs and then leaving it to others to clean them up and expand them is not a problem. Consensus here disagreed with you and an admin removed your autopatrolled rights. Other editors are merely telling you to slow down and focus a bit more on quality rather than quantity. Chillllls (talk) 15:53, 18 May 2012 (UTC)
 * That IP definitely was. I just don't like it when I try my best to solve issues but I'm being accused of "ignoring them" and "making up excuses" which is not true. I don't appreciate Azylber highlighting the words "fuck off" in bold which is trying to make it look like that I'm being uncivil, but I have never been uncivil around here. I am fixing some of the problems now. I estimate that around the 8,000 Chinese townships I created, only 30% or a little more have errors in them. Do people have the joy of running me down? Jaguar   (talk)  16:03, 18 May 2012 (UTC)
 * No, people are not taking joy in "running [you] down." Think about this for a second: you estimate that 30% of 8,000 stubs have errors. Looking at it from another perspective, that's twenty-four-hundred errors that you've inserted into the encyclopedia. You're creating these stubs at roughly the same rate as a bot, and a bot with a 30% error rate would never ever be approved. You should realize that there are editors on this page who have said nothing about your civility but have a problem with your stubs. No one is calling for you to be blocked, so please stop playing the victim and fix your contribs. Chillllls (talk) 16:11, 18 May 2012 (UTC)
 * That IP was not. Your talkpage includes comments from two users experienced in high-volume page creation, one of whom mass created the politican stubs highlighted above as AlbertHerring then four days after the bulk AfD closed renamed to Ser Amantio di Nicolao (not all he does & he's done a lot for the site), and Dr. Blofeld - who wrote he's also counseled you in email.
 * It pushes credibility imo, that they wouldn't be aware of the policy. It became policy not long after that incident. At worse, it can be argued the editor(s) knew or could reasonably be expected to know that you hadn't proposed it, perhaps considering policies don't have to be followed and/or it's better to ask forgiveness than permission, yet didn't bring it up to you in passive encouragement to avoid following policy. The reasons it mandates tasks must be approved are twofold: to help ensure projects that ought to go ahead go well and to ensure editors are not demoralised. You wrote above "Mass deletion is not the answer. That is the most upsetting thing I've ever heard". Had it been proposed help could've been given. Instead a result has been to make an editor, and a young editor at that, feel like crap.  This is exactly why DGG said what he did in the community discussion  linked from the policy. People are not taking joy in this at all. --92.6.200.56 (talk) 17:30, 18 May 2012 (UTC)
 * How do you know so much about me!? And I guess I would have felt like more crap if the '10,000' of my articles got deleted. Jaguar   (talk)  20:43, 18 May 2012 (UTC)
 * Don't worry none of us here know anything personal about you. I was going by your upset comments above. Nobody here wants to make you feel crap, or crappier. I wrote young because you use the school template on your page. --92.6.200.56 (talk) 21:10, 18 May 2012 (UTC)
 * Also, Jaguar has a "this user is a teenager" userbox. Quite frankly Jaguar, you've created an enormous workload on others now. All of your articles need to be checked for errors. Even if by yours reckoning 30% of your articles have errors, it makes no difference to the fact that someone is going to have to go through all of them to work out which ones have problems. In fact, I just sampled the last 29 stubs you edited and every single one used the same link as a reference, to the wrong page. All of them link to the Anhui province page except you created 29 stubs about township level divisions in Beijing. Honestly, I see some serious competence issues here. If you can't be bother to check your reference then you shouldn't be creating articles. I propose that Jaguar be banned from creating any more articles until they've sorted out the mess they've created. Blackmane (talk) 01:00, 19 May 2012 (UTC)


 * Since I've been mentioned: we have a great need to properly advise new editors, more carefully and consistently than we do, but even if we always did it properly, it can only work with those editors willing to listen to advice. When they do not listen to advice, the next step is enough of a warning that they realize. And then if they finally learn, mistakes at the beginning will in fact be forgiven. Creating mass articles is dangerous. It can be done right: a few very experienced and skillful and careful editors have done excellent jobs of it in both geography and biology and to a certain extent in biography also. But some pretty good editors in each of those fields have  also gotten overconfident and let things go too quick to control, and have shown sometimes they did not realise all  the potential problems. WP is a live & very visible database, and testing any automated process on a live database is dangerous. The way to do mass anything is to start slow and small, increase the numbers and speed gradually, test the output yourself at every stage,  and pay attention to the results and the comments. And then decrease the speed if problems develop. New editors especially need to do this: the number of things that can go wrong with an article here is beyond what anyone can possibly realise at first. The difficult of fixing them, especially when there are few qualified experts except yourself because of language or subject, is very considerable.  You cannot expect the people who have to do the work not to resent it. When you start again, and I suggest you wait a while before that, please go very slowly. I'd suggest 5 or 10 articles a day at most. I'be been here five years, and I  never would even try  to make articles any faster than 5 a day. I might write a great macro process, but i would fell obliged to check everything I did, and that cannot be done quickly.  DGG ( talk ) 04:33, 19 May 2012 (UTC)
 * I could not agree more with what DGG has said. He and I have differed on views about notability, but I think one consistent theme is an emphasis on accuracy. The above is excellent advice. People have been talking to Jaguar about this for a while now, and I don't think he's getting the picture yet. As I said before, I don't have any belief Jaguar's acting with any mal intention, however I think there's a serious problem with some of these stub creations by their sheer volume alone. I don't have much to add I haven't already said, but I think Jaguar needs to understand that this is a serious issue. Shadowjams (talk) 04:54, 19 May 2012 (UTC)

Yes, I am a very young editor. I fear that if I ever revealed my real age people would be surprised at me. I can assure all of you that there are not as many errors in my articles as you might suspect; I will correct all the ones I can find soon. I too could not agree more with what DGG has said. I will of course take that advice and use it; firstly, instead of going through some of my articles and correcting them, I could rewrite them using User:Jaguar/Sandbox/3, just filling in all the appropriate details. Once I have corrected my errors and redeemed myself at ANI, I will start slowly creating the Chinese articles, doing at least 10 a day at the most. I am over halfway through creating every Chinese township in the world. I will correct them - I've got to do it since it's all my fault really.

By the way there would be no need to ban me from creating articles, I'm not exactly an evil vandal who can't be trusted. Jaguar  (talk)  10:49, 19 May 2012 (UTC)


 * You're not a vandal and we can all see your efforts to try to put things right. By the way Jaguar, in all the talkpage/email comments to you did Dr. Blofeld mention the mass creation policy that's been talked about? --92.6.200.56 (talk) 16:24, 19 May 2012 (UTC)
 * No, we did not mention any mass creation policies or not that I can remember of anyway. How come you ask? Jaguar   (talk)  17:46, 19 May 2012 (UTC)
 * Curiosity. It's interesting to know more background sometimes. It would be good if Dr. Blofeld could come to this section, he might be able to help. --92.6.200.56 (talk) 18:15, 19 May 2012 (UTC)

I endorse the creating of articles about Chinese townships, infact I started and encouraged the creation of the lists by province. I believe China geographically and in terms of population is the most sparsely covered country on the planet on wikipedia and I believe we should have articles on all of the townships in the long term. However, I too have frequently spotted errors in Jaguar's stubs and if you check his talk page history you'll see I contacted him numerous times. The concerning thing is that the ones already created were not corrected after I spotted them. Technically I really think these articles would be better started with a carefully planned bot and given a trial run to look for errors. It als would be good if they could be started with a population figure. I believe there is also a website which lists subdivisions and postcode etc. I think in the long term we'd be better off having a bot create them. The problem of course is few people are expanding them but I believe we should be covering them. But its finding the most efficient way to start them.. When I started stubs in the past I always double checked to see there were no errors and if I did spot errors I'd contact Rich or Ser Amantio to AWB correct them and sort out any mistakes. I think the most productive thing out of this would be to organize a bot to fix all errors. Some of the dead ref links with the wrong code could simply be fixed with a bot after finding what province is what, you just run a bot through the whole province fixing the ref link.♦ Dr. Blofeld  11:46, 20 May 2012 (UTC)
 * Dr. Blofeld, thank you very much. You're right, it's apparent from the page history you've spoken to him more than once about errors in his stubs. One thing I wondered about, Jaguar said you hadn't talked about wp:masscreation policy. How come? --92.6.200.56 (talk) 12:08, 20 May 2012 (UTC)
 * Because he's not a bot. They are generated manually. And I have no problem with mass stubbing provided they are accurate without errors and with a fact or two. but as I say in regards to Chinese townships i think a bot should be used.♦ Dr. Blofeld  12:24, 20 May 2012 (UTC)
 * I think that's an example of this. The policy's about mass page creation and the page says whether they're human‑generated manually or not is irrelevant. --92.6.200.56 (talk) 12:38, 20 May 2012 (UTC)
 * Well whoever amended MEATBOT is violating one of the most important principles of wikipedia, WP:AGF. "The disruption must be stopped" does not apply to every stub. It is possible to generate a lot of valuable sourced stubs manually without errors which are useful as a start.♦ Dr. Blofeld  13:46, 20 May 2012 (UTC)
 * If I remember correctly User:Ganeshk has a process and instructions for creation using AWB if the relevant data is available in csv format, if there is a database to provide that, then it shouldn't be a problem. Most of the India village stubs created through this process are quite better than user generated stuff (primarily newbies who don't know the policies and guidelines). &mdash; Spaceman  Spiff  13:58, 20 May 2012 (UTC)
 * The disruption's plain to see. The thread length and amount of editors trying to clear this up alone attest to that. As others observed it would be wikilawyering to keep to the letter but not the spirit of policies. However, in this case it is the letter. MEATBOT is policy and has been for over two years, Dr. Blofeld. Going back even earlier, principles on higher speed editing or assisted human editing have been established policy for at least four. In any event I was asking about mass creation. Policy requires any large-scale creation task must be pre-approved and further strongly encourages (and may require) community input be solicited at WP:VP/PR. --92.6.200.56 (talk) 15:34, 20 May 2012 (UTC)

How bad is this?
How many articles are we talking about, in total (ballpark figure)? And approximately how many of them have serious problems (like where they say they're in one province, but they're linked to from a totally different Province article)? Anyone have an estimate? Qwyrxian (talk) 01:36, 20 May 2012 (UTC)
 * As I said above, I estimate that around the 8,200 Chinese townships I created, I say around 25% or 30% might have mistakes. It's not that bad to be honest. I could overwrite all the errors I can find. Jaguar   (talk)  10:13, 20 May 2012 (UTC)
 * I recently made 600+ beetle stub article, and every single one has MOS and Category errors. I fixed 'em all &mdash; 4 hours work. (account renamed – tomtomn00) Thine Antique Pen (talk • contributions) 11:53, 20 May 2012 (UTC)
 * It's good you fix up after yourself, TAP. That situation's probably a little different since they're all English-language though. --92.6.200.56 (talk) 12:08, 20 May 2012 (UTC)
 * Your estimated error rate makes me think we should rather delete them all and start over. Mass-creation with a more than 1% error rate just screams "nuke from orbit". Wrong info that isn't easily visible as such is worse than obvious vandalism IMHO. Also, in the substubs that do not even give the township's names in characters (making it hard to research and expand them), essentially nothing is lost by deletion. —Kusma (t·c) 12:07, 20 May 2012 (UTC)
 * I think it would be counterproductive to delete them. Override them, maybe, if somebody can sort out a bot and finish off the rest.. ♦ Dr. Blofeld  12:25, 20 May 2012 (UTC)
 * I agree, just imagine deleting 8,000 Chinese townships! I see no point - Like Dr. Blofeld has said, China is one of the most sparsely internet-covered nations on this planet, and having every Chinese township on Wikipedia has a huge potential of becoming a major article one day. China is the most populous nation, so it even has a bigger potential. These need to be kept. Jaguar   (talk)  12:30, 20 May 2012 (UTC)
 * The end (which is a long way off) does not justify the means ( talk→  BWilkins   ←track ) 12:32, 20 May 2012 (UTC)
 * This is an encyclopædia. If we can't be sure that something is accurate, it shouldn't be in article-space. I realise that rote editing and mass-creation of geographical stubs is very important to some people, but I would prioritise quality over quantity. bobrayner (talk) 12:42, 20 May 2012 (UTC)
 * The only alternative to mass-deletion at 25% error I see is to topic-ban the creator so that they would not be able to create anything until the existing errors have been fixed.--Ymblanter (talk) 12:53, 20 May 2012 (UTC)
 * He's not presently creating any. Also, with the best will in the world it's still unclear if he'd be able to fix or even detect all problems due to the language barrier. --92.6.200.56 (talk) 13:12, 20 May 2012 (UTC)
 * Don't worry, the language barrier is not the problem (or our biggest concern anyway). The issues are the errors in the articles (simple broken links and links that take you to different places etc). And Ymblanter, please, just assuming that this is an ANI discussion concerning me doesn't mean I'm a criminal who needs to be banned! Jaguar   (talk)  13:25, 20 May 2012 (UTC)
 * I do not assume you are a criminal, it is just that 25% is way over the top, especially given the absolute numbers. The material is just not credible, and has to be either immediately corrected or mass-deleted.--Ymblanter (talk) 13:29, 20 May 2012 (UTC)
 * The only option is to correct the ones that need correcting. As of all, it's a Chinese town somewhere in the world. I have seen some of them expand since after a few days I have created them. Trouble is, China is a big place and nobody might have travelled that far. Jaguar   (talk)  13:33, 20 May 2012 (UTC)
 * Do you have a list of the 2000-3000 that need correcting or do we have to go through the 8000 to find out which ones do a disservice to our readers? And if you are going to correct this, how long is it going to take? &mdash; Spaceman  Spiff  13:44, 20 May 2012 (UTC)
 * Then please start correcting them, Jaguar. In your contribution in the last couple of days I do not see any edits in the article space. These are your mistakes, and this is you who is primary responsible for correcting them, not anybody else.--Ymblanter (talk) 13:53, 20 May 2012 (UTC)
 * I will make a start correcting them tomorrow on in two days as I've got a mock exam tomorrow. I don't know how long it will take me until it's 100% clear that no more typos or errors exist but I should give it a week by myself, or longer if I get disrupted by another test. Jaguar   (talk)  15:38, 20 May 2012 (UTC)

According to my calculations, 5.4 solid non-stop days of editing. --Thine Antique Pen (talk • contributions) 15:43, 20 May 2012 (UTC)
 * I see about 30mins before this comment you added a userbox saying you're able to understand/communicate in Chinese at an advanced level—one step below near-native. I don't know why that talent'd be left out up to now while basic-ability German/French was highlighted on the userpage. Oh well it doesn't particularly matter. --92.6.200.56 (talk) 16:14, 20 May 2012 (UTC)
 * Bit confusing, eh? --Thine Antique Pen (talk • contributions) 16:17, 20 May 2012 (UTC)

I think everyone is overlooking this issue too much. I am the only one here who knows what I'm talking about, since I've started these articles, I know that in reality I haven't created that many mistakes. When I did spot a mistake, I corrected the error immediately and corrected my previous articles I created. All the mistakes you see in my articles are probably the leftovers of all the mistakes I have tried to fix in the past but I missed out. I might have even overlooked how many mistakes there are, there might even be less than 25% of 8000. It shouldn't take too long to fix once I start tomorrow or in two days. Jaguar  (talk)  15:47, 20 May 2012 (UTC)
 * Jaguar, I just had a look at some of your "fixes". They're nowhere near enough to be able to save the articles. You've directed the reference to the correct province page, but that's still way too lacking as a reference. It needs to direct to the actual township page. If this is all you can do, then I suggest you stop now and give it up as a lost cause. Blackmane (talk) 06:07, 22 May 2012 (UTC)

A sample
Just so we're clear about what's under discussion, I looked for some examples. I pulled these five off the bottom of Jaguar's contribs list (feel free to provide other examples if these are not a representative sample) Each seems to be, well, a neat assembly of templates and links and stuff but based on a single datapoint; that some placename exists. I realise that in the past we've often turned a blind eye to the use of an unreliable listing to create masses of geographical microstubs which fall far short of the GNG, but if the entire article hinges on a single fact that "this place exists" and our only source is a Google translation of a Chinese forum... surely we have to draw a line somewhere? (Google Translate isn't working very well for me at the moment but I can't even find some of these placenames on the page supplied - are these real places?). Sadly it's not the first time I've seen an argument that it's OK to mass-produce this kind of crap because in principle somebody else might be able to fix it - which, in reality, causes maintenance headaches for everyone else further down the road. I have no ill wishes against the creator, and I hope they get past this episode and make a lot of good contributions in future, but I think these articles as they stand are a net negative for the encyclopædia - shouldn't they be deleted, or sandboxed, or incubated, or something? bobrayner (talk) 14:42, 20 May 2012 (UTC)
 * Xinghuaying
 * Liangyuan Subdistrict
 * Shuidao Township
 * Beidaomen Subdistrict
 * Beishudian Subdistrict
 * They should all be kept, getting rid of them in any form is counter-productive, just like Dr. Blofeld has said. By the way those five examples you gave are 100% fine! Jaguar   (talk)  15:42, 20 May 2012 (UTC)
 * I don't see any s' filled in, do I not? --Thine Antique Pen (talk • contributions) 16:48, 20 May 2012 (UTC)
 * I don't doubt that Dr Blofeld told you it's fine. However, other editors may take a different stance on the value of a huge pile of microstubs which appear to fall far short of the GNG. Surely, removing flawed content (some badly-sourced, some outright wrong) isn't counter-productive, it's improving the encyclopædia. Insisting that articles are 100% fine despite specific problems being pointed out is part of the problem, not part of the solution, and does not bode well for the possibility of fixes being made in article-space. If thousands of articles are left in article-space even though we can't trust their content, doesn't that undermine the encyclopædia? bobrayner (talk) 17:23, 20 May 2012 (UTC)
 * Contrary to your statement Jaguar, they are not fine. The main problem here is that you have found a single source to reference your stubs. Normally, I wouldn't say that is a problem, but your obvious inability to read chinese means that you have no idea what to do with that source. Here's a list of what's wrong in just those ones you've sampled

1. Each of those townships is part of Kaifeng city in Henan province but his reference is labelled Fujian province (yes I can read chinese).

2. Clicking on that link takes you to the google translated main page of the source. The very least they could have done would have been to link to the city or even the province page, which given that it's been translated would have been a simple task

3. I pulled Liangyuan Subdistrict to see if I could find some info on it. I dug down into the reference page to see if I could find it. Jaguar wrote that it's in Kaifeng city and the List of township-level divisions of Henan also has it listed as part of Kaifeng. After 20min of poking around, I find that Liangyuan is part of Shangqiu, which a search in Google maps will tell you is 150km east of Kaifeng. Somewhat concerned, I had a look at the other 4 articles bobrayner linked to and those ones were at least placed in the right city. Taking this as a first order approximation, there is a possible 20% error of locating the place, with a 100% failure to properly reference the stub. Blackmane (talk) 17:01, 20 May 2012 (UTC)
 * Additional comment I would think that goes with out saying that there are possibly 2000 stubs which aren't located properly and 10,000 or more that have to have their refs checked. Blackmane (talk) 17:06, 20 May 2012 (UTC)
 * Past projects reliant on google translate have gone poorly. Quoting Eloquence: Engaging in large scale translation projects has its very own problems. See, for example, Sodabottle's scathing criticism of Google's translation efforts in Tamil Wikipedia. --92.6.200.56 (talk) 20:07, 20 May 2012 (UTC)
 * According to List of township-level divisions of Henan there's a Liangyuan (梁园) District in Shangqiu and a Liangyuan (梁苑) Subdistrict in Kaifeng. Google maps does know of a neighbourhood of that name in the right part of Kaifeng, so it could be right, though we can't be sure because unfortunately the township list is unreferenced too.  Kanguole 16:53, 21 May 2012 (UTC)


 * As I said before, these pages should be deleted. I can also read Chinese, and the "references" given are not referencing the article. It is possible to find references on XZQH, e.g. this about Xinghuaying, but the substubs link elsewhere instead. However, before mass-importing data from a single source, we should check what kind of source this is (copyright questions aside). Start over from scratch and ask people who can read Chinese to help (e.g. at the relevant WikiProjects). —Kusma (t·c) 17:21, 20 May 2012 (UTC)
 * And certainly any mass creation done here should be interwikilinked to the Chinese Wikipedia, which seems to have at least Xinghuaying. —Kusma (t·c) 17:26, 20 May 2012 (UTC)

I'm not understanding why Jaguar repeatedly reports that he does not have the time to begin fixing the errors in the sub-stubs he created, but has the time to post multiple entries in this thread. Please, Jaguar, stop talking and start fixing. Beyond My Ken (talk) 22:19, 20 May 2012 (UTC)
 * I've just fixed over 50 refs this morning. Will do more. Jaguar   (talk)  10:13, 21 May 2012 (UTC)
 * I am sorry, but what you are doing is not nearly enough. If you go through the articles, you have to check them for accuracy. I just checked one of the things you "fixed", and Xingfeng appears to be to be a subdistrict (街道), not a township. I am led assume none of your articles is correct, so I guess deletion and starting from scratch is probably a faster way towards covering Chinese townships than expecting you to deliver on your promises. —Kusma (t·c) 10:43, 21 May 2012 (UTC)

At this rate that Jaguar is going at, it will take him around 97 days to complete these &mdash; not what I was looking for. Now, at the speed that I corrected my bad stubs: 3 days, 4 hours it would turn out as. I did 600 in 4h, Jaguar did 50 in 11h, 49m (BST UK). --Thine Antique Pen (talk • contributions) 10:49, 21 May 2012 (UTC)
 * If we can't be sure that something is accurate, it shouldn't be in article-space. We can't be sure that any of these stubs are accurate unless/until verified by a third party. So, they should not be in article-space. bobrayner (talk) 11:02, 21 May 2012 (UTC)
 * I know that everyone is so keen on deleting my articles, but I have to say that I created them by using all the red links in List of township-level divisions of Henan etc. I copy and paste the header in the article as its province assuming it is correct. I would not know if it isn't correct, so it's probably the list's fault. Jaguar   (talk)  12:16, 21 May 2012 (UTC)
 * And there we are. If an unsourced wikipediapage is the source of thousands of articles, we risk to multiply our own errors. I appreciate the work done, but don't see how we can change this easily without deleting. Or is there a way to properly source(+interlang) all articles, while correcting the 500-2000 erroneous ones by bot? If the latter is possible, that seems the only non-deleting way forward to me... L.tak (talk) 12:59, 21 May 2012 (UTC)
 * Personally, I'd be happy with incubation or sandboxing as alternatives to deletion. Incubation was one of the more popular options in this case so there is some precedent, I think. bobrayner (talk) 13:41, 21 May 2012 (UTC)
 * Over a year on every single one in the nom statement is either deleted, redirected, unsourced and/or untouched since. Similar suggestions of templating or just leaving them were made in the other two mass problem cases; both were deleted. --92.6.200.56 (talk) 15:06, 21 May 2012 (UTC)


 * Three factual comments (in response to some comments above): (1) Province-wide pages such as List of township-level divisions of Fujian (and similar ones for other provinces) are actually pretty good. I don't know what the source for them was, but whenever I happen to work on a particular county, and look at that county's divisions' list in List of township-level divisions of Fujian (or another province's list), it checks out very well against high-quality printed atlases. Occasional discrepancies that I see sometimes are probably due to renamings/splits/merges that local governments carry out every now and then. (2) Most prefecture-level governments have web sites with at least a brief information page about every county, including the list of township-level units; this always can be used as a fairly reliable "official" source, if we can't find one at some kind of the national Census Bureau or some such. (3) On occasions, a unit called a "township" (乡 xiang) in one source may become a "town" (镇 zhen) in another source, or a subdistrict (街道, jiedao) in a third. This, per se, is not a reason to claim that a particular source is unreliable: it is very common for provincial authorities to convert a township to a town (as it becomes bigger and more urban in nature), or to a subdistrict (if it becomes more integrated into a city's main urban area; this is China's counterpart to the Municipal annexation in the United States, although the mechanism is quite different). -- Vmenkov (talk) 05:28, 22 May 2012 (UTC)


 * I don't advocate the deletion of all these recently created township units; certainly, not those which have already been worked on by other editors since they have been created. But perhaps we can "Anglicize" and "upgrade" zh:User:Liangent-bot, which has been quite successful in creating thousands of township articles in zh.wiki, presumably based on some master CSV file. One can add to it some functionality that will look at if the township's article already exists and is completely trivial (we know what I am talking about), and if so, overwrite it with a more content-rich one. -- Vmenkov (talk) 05:28, 22 May 2012 (UTC)
 * If that works, that would be a vastly more suitable solution. How would that work with incorrect references as is the case here? I'm probably going about this question the wrong way, but a more detailed description may prove the difference between wiping out all these articles and finding some way to salvage them somehow. Blackmane (talk) 06:01, 22 May 2012 (UTC)
 * @Vmenkov, all of my new township content comes from translating (by hand) the ZH Wiki's counterpart articles.
 * @L.tak. Nope. At least three of the lists are sourced to the prefecture by XZQH, which sometimes has holes. Once I complete moves in accordance with the new township disambiguation guidelines at WP:NC-ZH, my priority, as Dr. Blofeld discussed with me ages ago, is to build these lists, and not to find a government source listing all township-level units, which is in itself a challenge because finding them is not a trivial task! Also, Vmenkov is right. The only real concern occurs when there are mergers or upgrades. Otherwise, apart from the very rare typo or usage of the wrong particular reading of a character, these lists check completely correct against atlases. GotR Talk 19:14, 22 May 2012 (UTC)
 * So, do I understand correctly that the source is either zh.wikipedia.org or xzqh.org? L.tak (talk) 19:50, 22 May 2012 (UTC)
 * There's a website Bob CC Guide in which lists the village development committees and towns of each township. I used it for the article on Xainza which a few years ago was a sub stub..♦ Dr. Blofeld  12:46, 23 May 2012 (UTC)

Proposed deletion of all these pages
Several people have suggested deletion of all these pages as the best solution to this. I don't think anyone suggests tagging them all for AfD would be really useful, so we can just as well have the discussion here and now. What do people feel about the proposed solution to delete all pages created by User:Jaguar from 26 October 2011 on? Note that this would eliminate the need to clean up e.g. the 111 articles which start with the identical line "Saiqi (Chinese: ?) is a township-level division [...]", but that this doesn't mean that we shouldn't eventually have articles on some or all of these. Fram (talk) 13:44, 21 May 2012 (UTC)
 * I would support that proposal. If anybody wants to remove an article from the list which has been checked & fixed by a third party, I think that would be reasonable too. Alternatively, we could take them to a separate AfD page, like Dr Blofeld's geographic.org articles, but that would mean splitting discussion across different pages... bobrayner (talk) 13:57, 21 May 2012 (UTC)
 * With regret (this could all have been avoided), yes it would seem deletion is necessary. It'd be on the basis of all the significant problems identified. I think a method used in such cases in the past is to use a single example and refer to the rest. --92.6.200.56 (talk) 14:03, 21 May 2012 (UTC)
 * Oh no you don't! You're not deleting my hours of hard work! I won't let you! Jaguar   (talk)  14:11, 21 May 2012 (UTC)
 * Well now I'm curious, how would you stop us? --Golbez (talk) 14:14, 21 May 2012 (UTC)
 * I don't know but it's going to be really upsetting if somebody deletes them. I'll have to create more outside of China. Jaguar   (talk)  14:16, 21 May 2012 (UTC)
 * If you create additional content which is accurate and meets wikipedia standards, that's great. If you create more content which is inaccurate and fails wikipedia standards, it would probably get deleted. Your choice. bobrayner (talk) 14:19, 21 May 2012 (UTC)
 * Everyone is implying that all 10,000 of my articles have errors. That is by far not true. I estimated that around 20% of the 8,000 Chinese townships have errors, maybe under 1,000 articles. It's not that much and it is fixable by a bot or manual users. Jaguar   (talk)  14:21, 21 May 2012 (UTC)
 * Point is: NONE of them have a reliable source according to WP:RS, so that needs to be changed. If you have a proposal how to do that (and improve all errors), then be my/our guest. However, I am afraid that this takes a lot of time (I myself have 5000 edits in total or so, so the mere suggestion to find and correct 1000 errors without a clear bot/plan sounds very ambitious). L.tak (talk) 14:24, 21 May 2012 (UTC)
 * A User:JaguarBot might be in the question. Jaguar   (talk)  14:28, 21 May 2012 (UTC)
 * Oh Jaguar. :-( You do not understand Chinese. A bot won't either. --92.6.200.56 (talk) 14:30, 21 May 2012 (UTC)
 * I guess so, but a bot can at least correct the errors! Jaguar   (talk)  14:31, 21 May 2012 (UTC)\
 * Great! Please make an action plan on User:Jaguar/ChineseTowns of what the bot should do. By which mechanism it would find wrong names, what would be the basis for interlangs and which reliable source it would be based on. If that is credible, I am willing to reconsider... L.tak (talk) 14:34, 21 May 2012 (UTC)
 * Jaguar, read my support below. We're not saying all of your 10,000 articles have errors. It's the fact that we don't know which of the 10,000 articles have errors and regardless of whether some do or don't all of the articles would have to be checked. Blackmane (talk) 15:51, 21 May 2012 (UTC)


 * Support, but also support sandboxing or the like for a limited time period (let's say: 1 year, which means 30 articles per day) if someone is willing to "adopt" them L.tak (talk) 14:27, 21 May 2012 (UTC)
 * Support; while we should have articles on all of these places, these are not even useful stepping stones in that direction. —Kusma (t·c) 14:32, 21 May 2012 (UTC)
 * Support owing to the terrible state these are in, especially with what Blackmane and Kusma have unearthed. I'm also open to moving these all over to userspace, to be returned only after source verification and content verification is done on an individual basis. &mdash; Spaceman  Spiff  14:39, 21 May 2012 (UTC)
 * Support Neutral if  can fix all of them up within a month, then it's fine. However, if he cannot Nuke the new pages Jaguar made. --Thine Antique Pen (talk • contributions) 15:13, 21 May 2012 (UTC)
 * Support - I don't have any expectation based on the conversations above that these are going to get fixed. Or if they can be fixed. There's a lack of understanding by Jaguar that even if most are accurate, the extremely unacceptable error (which appears to be randomly distributed through the set) rate introduces too many errors with little or no notice to readers that there are errors. Shadowjams (talk) 15:15, 21 May 2012 (UTC)
 * Support, and I would also suggest that Jaguar does not attempt to "create more outside of China" as such mass edits of a similar nature could be considered as further disruption. GiantSnowman 15:19, 21 May 2012 (UTC)
 * Yup. I can grudgingly accept geospam when it is properly prepared, extracted from data we know to be reliable and then carried out in an automated manner (which means that if there are any obvious screwups that these can be resolved with a bare minimum of drama). None of that is true here. "Turn all redlinks blue" is not an end to itself, and general community consensus is that the acceptable error rate goes down rapidly with an increase in editing rate. And whether or not Jaguar was privy to the debates or not, we've been down this road before, and the pattern is the same as always: the editor responsible repeatedly denies the scale of the problem until it is revealed to be wholly unmanageable, and yet still insists that the positives (of having more articles) outweigh the negatives (that the articles contain either no information at all, or objectively false information). The worst thing that can happen is that Wikipedia becomes the canonical source for information on a subject (due to our huge PageRank) while that information is wrong. Chris Cunningham (user:thumperward) (talk) 15:32, 21 May 2012 (UTC)
 * Support. This popped up on my watchlist, and I thought, "Oh, it can't be THAT bad." It can, and it is. We now have thousands of articles of which we have no way of ascertaining the accuracy. And we have a user (Jaguar) who left a comment that basically amounts to a threat if these are deleted. I think the time for action is here on this issue. LHM 15:37, 21 May 2012 (UTC)
 * Support because we are encyclopedia first and playing ground for whiny pouty children second last. / ƒETCH COMMS  /  15:47, 21 May 2012 (UTC)
 * Shouldn't that be " second last never"? GiantSnowman 15:51, 21 May 2012 (UTC)
 * (ec) Fetchcomms, I think that's needlessly personal. --92.6.200.56 (talk) 15:54, 21 May 2012 (UTC)
 * Unfortunately, given the header box Jaguar has seen fit to place on their talk page, this is somewhat accurate now. Blackmane (talk) 16:01, 21 May 2012 (UTC)
 * He removed 'not' from the longstanding header box in response to a complaint it was no longer accurate. When I saw it it seemed more like wry humour used in a stressful situation than obstructiveness. Anyway ymmv. --92.6.200.56 (talk) 16:31, 21 May 2012 (UTC)
 * Blackmane - you can do exams at any age, and that certainly doesn't justify describing another editor as "whiny, pouty" - their response, however, perhaps does. GiantSnowman 16:35, 21 May 2012 (UTC)
 * Theirs is an unacceptable response certainly. Although as an aside User:Geogre/Comic did spring to mind. --92.6.200.56 (talk) 16:44, 21 May 2012 (UTC)
 * Now you're at it! Jaguar   (talk)  16:43, 21 May 2012 (UTC)
 * Jaguar, go read the poem "If—". You don't have to read much less respond to all comments in this section, and sometimes it's better not to do that. --92.6.200.56 (talk) 16:49, 21 May 2012 (UTC)
 * Thanks, I've read it. I think you're right. Jaguar   (talk)  16:51, 21 May 2012 (UTC)
 * I had been meaning that I was going to revert their attack on fetchomms. As for the childishness, I was referring to the info box that was below the exam one. oh well, moving along then Blackmane (talk) 20:07, 21 May 2012 (UTC)
 * By "whiny, pouty", I was referring to Jaguar's outburst of, "Oh no you don't! You're not deleting my hours of hard work! I won't let you!" above. Regardless, the "Go fuck yourself you prick" was amusingly ironic. (And as an aside, thanks for mentioning If—, 92.6.200.56; I don't think I've read that one before, but it gives me an idea for a poster.) / ƒETCH COMMS  /  01:02, 23 May 2012 (UTC)


 * Multiple Sadly, I will have to Support this motion to delete. However, if there is an option to put them into some sort of incubator outside of article space I would prefer that. Jaguar, it's great that you chose this set of articles to create and I applaud you for that. It's also great that you agreed to go back to correct errors and that shouldn't take all that long if you get stuck into it. However, the problem here now isn't just the article errors, it's the referencing. It took me almost 25min to look for the correct page in the reference (I don't know how long it took Kusma) when clicking on a link that should take me straight there, but can you see the problem with doing that for 10,000 articles? If I had to do that for all 10,000 articles, that's almost 170 days worth of editor man-hours. The other problem is that you are more concerned with the rate of your article creations rather than the accuracy of your article creations, this you freely admitted to. The very fact that you failed to check the accuracy of the location information prior to creating the article is also very concerning. The example I highlighted before I easily checked by entering the name into Google maps. You could have done this before creating the article and corrected the list before creating your article, but instead you chose to take the information at face value and prized speed above accuracy. None of us here want to delete your articles out of spite or malice but out of consideration of the huge expenditure of editor hours required to clean up. Blackmane (talk) 15:49, 21 May 2012 (UTC)
 * Changing them based on Google maps would be so great either – the example you pointed out might not be an error, as noted above. But I agree with you that the real problem here is the lack of referencing, which has been inherited from List of township-level divisions of Henan and similar articles, and multiplied several hundred-fold.  Kanguole 17:26, 21 May 2012 (UTC)
 * That's why I said it was a first order approximation. Given the scale of the problem, sampling merely a handful of articles is hardly statistically relevant. I was aiming for a hand waving estimate on what might be the upper limit of the problem, but as we see now, I most likely have vastly understated and underestimated the scale. Blackmane (talk) 21:41, 21 May 2012 (UTC)
 * Support, there is a bunch of problems that leave no other option, IMHO. 8000+ stubs with barely one sentence and one basic infobox. An unknown percentage contains errors or is totally mistaken. Not based in any reliable source, so they are impossible to verify. The creator fixed a few after a lot of prodding, but the fixed articles still contain basic errors and no reliable source (example of one "fix" made, only the glaring error in the name is fixed, the only source is still a Chinese forum with lists of names, it links to a list of articles not to the specific page that holds the information, the name is not translated to Chinese so I can't even search its Chinese name in the forum to find the correct page and translate it, never mind that the whole article still seems to be based on the unsourced article List_of_township-level_divisions_of_Fujian). The whole thing should be deleted as salvageable without complete rewriting, and future mass creations should be based on a reliable database that is cited in the created articles. --Enric Naval (talk) 16:11, 21 May 2012 (UTC)
 * 20-Mule Team Support: Oh hell yes. "This is an encyclopædia. If we can't be sure that something is accurate, it shouldn't be in article-space," given above by bobrayner is .sig-worthy if Wikipedia had .sigs.  Something I've often said is that Wikipedia is not a race - we don't hand out door prizes for the most new articles (deletions, prods, redirects, AfDs) thrown up in an hour's time.  There is never such a burning need for an article that WP:V and WP:IRS has to be set aside.   Ravenswing   16:21, 21 May 2012 (UTC)
 * Support - doubtless Jaguar means well, but these are unreliable (because they're unreferenced) and hard to check (because they lack such basic information as the Chinese characters, pinyin and district). They're a net negative – best to start over.  Kanguole 17:26, 21 May 2012 (UTC)
 * Support - 8,200 incorrectly sourced articles about locations that may not be where they are listed, which may be named incorrectly, and which may not even exist is a serious concern, and there's no easy solution other than starting from scratch. —Torchiest talkedits 21:02, 21 May 2012 (UTC)
 * Support per Torchiest and others. 8200 completely unreferenced pages is a serious problem. Under the circumstances, I suppose we might be able to give someone a few days to copy them to disk, but nothing more than that. John Carter (talk) 00:08, 22 May 2012 (UTC)
 * Support It's probably not necessary to pile on, but for future events with different editors it may be desirable to get a clear consensus that the mass creation of junk stubs is not helpful to the encyclopedia. Sure, the pages look pretty, and if someone else cleaned them up, it would be handy to have been the creator of many articles. However, as described above, this is an encyclopedia and content-by-guesswork is not acceptable. Mass creation means there must be sources that are ultra-reliable—the content must be known to be accurate. Johnuniq (talk) 02:10, 22 May 2012 (UTC)
 * Support - I am not convinced by his contribution to this section that Jaguar understands what he did wrong, or has the capability (or time) to fix them. Correcting misinformation is much harder than starting from scratch with new information.  It is better for an encyclopedia to have a lack of information rather than misinformation, since the misinformation could be seen as authoritative. Beyond My Ken (talk) 02:19, 22 May 2012 (UTC)
 * Support: If Jaguar had made 10K articles all based off a reliable source (say, an official government census, if China has such a thing), and even if all that source verified was the exact location of the place and its governmental hierarchy (township, city, etc.), I would say that these could be kept, even if they had lots of template errors, missing info, etc. But right now, we don't even have evidence that these places exist, other than that some previous Wikipedia editor added them to a different Wikipedia list. As Johnuniq said, consensus is already clear but the overall message needs to be more clear: mass article creation must be at least a little bit reliable, and there appears to be a relatively high chance that any given article in this series isn't just incomplete or mal-formatted, but actually completely wrong about its most basic fact. Qwyrxian (talk) 02:52, 22 May 2012 (UTC)
 * Comment — since when was ANI the new AfD? I appreciate we're talking about a large number of articles here, but I don't believe it's wise to set a precedent that articles can be deleted based on ANI consensus. —Strange Passerby (talk • cont) 03:37, 22 May 2012 (UTC)
 * I believe a similar action occurred in the past when an editor mass created articles on Pakistani townships. However, I think what we're aiming at here is an agreement by the community to put forward 10k articles for deletion. How that would be properly managed without having ANI look like AfD (except in the case of blatant copyvios) is another question. Blackmane (talk) 05:56, 22 May 2012 (UTC)
 * AfD is designed for single articles. However, as long as editors have the ability to create multiple articles, we'll need the ability to delete multiple articles sometimes - and trying to shoehorn them into AfD can be rather disruptive. I don't want to get hung up on bureaucracy - AfD is just a way of gauging the community's stance on whether an article should be kept or deleted, and we're getting the same community discussion here. Considering that it all came about of an AN/I dramathread, discussing it here is probably the best option. However, in future we might want to explore better ways to deal with mass-deletion proposals. bobrayner (talk) 07:13, 22 May 2012 (UTC)
 * Support As I've said on Jaguar's talk page, just fixing up the Chinese characters in the stubs is a potential nightmare because of the ambiguity of pinyin. It would be better to start again. ► Philg88 ◄ star.png 07:36, 22 May 2012 (UTC)
 * Support and don't waste any more time discussing this - absolutely obvious need for mass deletion. Not fair on other editors. Jaguar can always recreate them properly. That they were his hard work is irrelavant, I could put an enormous amount of effort into had creating an obscene piece of ascii art to illustrate auto-fellatio but I can't imagine that it would be kept. Alternatively if anyone wants to write a bot to do it, I don't see any problem with userfying all of them for him to fix over the next 97 days or whateveer...  Egg   Centri  c  13:52, 22 May 2012 (UTC)


 * Support: Per Phil88. It would be much easier to insert Chinese, which is perhaps most important in verifying locations, if we started over with an article-creating bot rather than instructing a bot to do that now with the existing articles.
 * Comment : I was the one who amassed together much of the lists of these township-level divisions. As Vmenkov said this morning, the errors are actually quite sparse, so generally the issue is not whether they exist or are located correctly. Also, I have occasionally checked through Jaguar's creations and have noticed that if an error (such as linking to Hebei in an Anhui location) is present in one article, it is prevalent across the other township articles in that province. If deletion proceeds, I suggest, beginning this December at the earliest, that a bot with the ability to download all relevant information from these lists and use a template from which to create articles is implemented. GotR Talk 19:28, 22 May 2012 (UTC)


 * I would be happy with a bot doing this, if the bot were able to write substantial content about notable topics. Which sources would the bot use? bobrayner (talk) 22:58, 22 May 2012 (UTC)
 * Either the prefectural page at XZQH is used, or someone actually takes the time, which I certainly don't have this year, to find government sources for each county-level division. GotR Talk 23:04, 22 May 2012 (UTC)
 * As far as I can tell, the pages at XZQH are basically just a list of placenames. Maybe I missed something. Are there pages there which describe each settlement directly and in detail? If you have some other source that could be used to build substantial content about notable topics, that would be helpful. bobrayner (talk) 23:11, 22 May 2012 (UTC)
 * I had a read of some of the pages in the XZHQ website and it's more of a general intro into that locale. Population, extent, that sort of thing. Nothing that really makes the place stick out. Blackmane (talk) 23:48, 22 May 2012 (UTC)
 * XZQH does occasionally provide town descriptions, but you'll need to look wide to find them. As stated by Vmenkov above, some prefectural/county government pages fulfil GNG. However, you cannot reasonably expect a bot or a human mass creating to meet GNG on the first edit of every entry. As a start, the Chinese name of a place, and correct sourcing affirming the existence of a location is a requirement, which almost all of Jaguar's creations have failed to meet. GotR Talk 07:07, 24 May 2012 (UTC)


 * Support. For all the reasons I stated above when I started this discussion.Azylber (talk) 11:55, 23 May 2012 (UTC)
 * Support No other way to deal with them, as a point of procedure should they be treated as if PROD'ed i.e. a request from a user in good standing should be enoutgh to get the article back ? Mt  king  (edits)  07:26, 24 May 2012 (UTC)

Is the problem limited to China?
I looked further back through Jaguar's article creations (long list on the toolserver here) and noticed a bunch of other placenames. I took a sample of some Spanish ones created in September 2011: Each of those was created with a simple copypasted '' "  is a village in..." '' (yes, that translation warning tag is backdated, and no, there's no evidence that they were actually translated from anywhere). Each claims this is the source, but that is the front page of an (english-language) wiki; searching it for those placenames returns zero results. I haven't counted, but there seems to be a large number of Spanish geostubs wilfully created with a single "source" which doesn't actually mention the subject. bobrayner (talk) 10:54, 22 May 2012 (UTC)
 * Callús
 * Bigues i Riells
 * Badia del Vallès
 * L'Ametlla del Vallès
 * Alpens, Barcelona (A duplicate of the pre-existing Alpens which already has some content and sources)
 * Vita, Spain
 * Sotalvo
 * That does appear very concerning, and it's also concerning that Jaguar doesn't seem to have owned up to it. —Strange Passerby (talk • cont) 11:20, 22 May 2012 (UTC)
 * Jaguar was sanctioned for some reason other (I'll have to find it later) which limited them from creating articles without at least one source. However, this looks like just plain gaming of the system. I think Jaguar has just done a google search for something like "China census" or "Spanish census" then simply copied the link of whatever was the first search that popped up and used it as the source without checking it. Blackmane (talk) 11:46, 22 May 2012 (UTC)
 * Followup Yep, Googling "Spanish census" returns this source as the second hit. First hit is the National Statistics Institute of Spain. Ironically, Jaguar could have punched in Callús and gotten quite a bit of information from the NSI which they could have used to populate an infobox rather than use an uninformative wiki. Blackmane (talk) 11:53, 22 May 2012 (UTC)
 * Your assumption about e.g. China although reasonable is wrong, Blackmane. The sanction you refer to is his having autopatrol temp revoked last summer by the way. --92.6.200.56 (talk) 12:10, 22 May 2012 (UTC)
 * I remember about the autopatrol rights being removed the first time. There was another restriction imposed, but I'll strike my first statement anyway as it's not relevant at the moment. Blackmane (talk) 13:21, 22 May 2012 (UTC)


 * Yep, this looks like a pattern for his Spanish edits as well. We should probably add everything from San Clemente, Cuenca to Santa Olalla (another 531 articles) to the list. His Serbian entries (544 of then, which run from his last Spanish creation to his first Chinese one) are equally suspect: Stančići (Čačak) is sourced to this, for instance, which from my best guess is a collation of census figures which doesn't mention the settlement in question at all. Chris Cunningham (user:thumperward) (talk) 13:23, 22 May 2012 (UTC)
 * That looks to be the French portal for accessing various census data from around Europe, but again Jaguar has linked to the main page which is useless for anyone who doesn't understand french. Blackmane (talk) 13:28, 22 May 2012 (UTC)
 * The autopatrolled bit was returned on 12 September 2011 with "per discussion with editor - Jaguar is expected to create articles with at least one reference". The Spanish articles predate that so I also sampled some more Serbian ones which are later creations. These also appear to have been copypasted without sources. (Even the transliteration was copypasted - there seem to be many different placenames which are all written as "Јеловац" in Serbian - so we can be sure that there are a lot more factual errors out there waiting to be found). I shouldn't clutter this subsection with more lists/examples - can we focus on resolving the deletion question? bobrayner (talk) 13:38, 22 May 2012 (UTC)
 * At this stage, aside from the deletion question, I think ANI needs to be considering some sort of community sanction on Jaguar. —Strange Passerby (talk • cont) 13:45, 22 May 2012 (UTC)

I can see that some of you are arising concerns about my other articles. I must tell the truth that at the time of creating the Spanish articles I had a hard time trying to find a good enough census, so that Spanish census was the only one I could find. The Spanish articles I created back in August/September were my first bunch of mass creations and I didn't create much. I don't see any errors with the Spanish articles and they are nothing compared to this Chinese articles, I must admit. I then went to create a couple of thousand Serbian articles, again I really can't find any errors at all with them. I think it's just the Chinese townships that have errors, but then again, not all the Chinese townships have errors, only a proportion of them do. Jaguar  (talk)  14:21, 22 May 2012 (UTC)
 * You say "that Spanish census was the only one I could find", but that source doesn't actually mention the settlements. The best source you could find was one which didn't even mention them, but you went ahead and created 531 Spanish articles anyway, each using that link as the "source"? Wow. You're already in a hole; I'd urge you to stop digging. bobrayner (talk) 14:28, 22 May 2012 (UTC)
 * "I don't see any errors with the Spanish articles". Apart from the sourcing problems, e.g. Bocos de Duero has the wrong province category (should be Valladolid instead of Albacete). But then again, San Clemente, Cuenca, Ribatajada, Manzanillo, Valladolid, Castrobol, Castrillo-Tejeriego, ... all have the exact same problem, so I assume that every single Spanish stub you created used the same province category, no matter what the actual province was? I suggested above to delete all Chinese stubs, but wholeheartedly support an expansion to a deletion of every single article you created, and a one-year ban on creating new pages. You clearly can't even be trusted to find problems in your own articles even when you are under close scrutiny. Fram (talk) 11:28, 23 May 2012 (UTC)
 * Could I suggest a compromise?
 * As far as the existing articles are concerned, there are new problems everywhere we look; so I would recommend deleting all Jaguar's mass-produced placename stubs (mostly Chinese, but also some Spanish, Serbian &c). I could create a list; happy for individual articles to be removed from that list if they've been fixed/expanded by other editors. If any have passed DYK or been assessed as Start class or better &c then I would automatically remove them from the list. Delete whatever's left on the list because it simply cannot be trusted, does not demonstrate notability, and manual investigation & repair would be a huge drain on the community's resources.
 * As far as future editing is concerned, I think that this thread has been an eye-opener for Jaguar; they may have defended the existing articles but I'm going to go out on a limb and suggest that there's a reduced risk of recidivism (especially if they know that a relapse is likely to bring severe sanctions) and despite the earlier excuses I'm willing to give Jaguar another chance. So, instead of a 1-year article creation ban, something a little milder like mentoring or parole or what-have-you, as long as it doesn't involve a great deal of oversight by other editors. Jaguar has done some genuinely good editing in other areas; surely we'd all like to encourage more of that.
 * I hope that compromise could clean up the existing mess, reduce the risk of future mess, encourage Jaguar to do more of the good stuff, and reduce the cleanup burden on other editors. What do y'all think?


 * Don't worry about Spain I've just checked the categories and seeing where he starts one province and ends another, only a few needed correcting. As for adding infboxes data and sources, don't know if I have the steam for that but I might expand a few.♦ Dr. Blofeld  12:06, 23 May 2012 (UTC)
 * I agree with this measure, especially for the Chinese articles; but have not enough info to judge salvagability of the other. I do have the feeling the Spanish ones are salvable (based on the links in the interlangs), but think they do require a source (like verifying the statistical source (which I guess is not deep linkable to the villages) in some interlangs). Also as a sign we're not against jaguar and all his stub work has been for nothing, I am willing to source, check + infobox 100 before June 15. Who follows? L.tak (talk) 12:51, 23 May 2012 (UTC)
 * Thanks, D. Blofeld. What to do about articles created "from the Spanish article" which don't have a corresponding Spanish article? Setaigües and Titaigües are two example of this (also of HPBot errors, but that is another discussion). Fram (talk) 12:55, 23 May 2012 (UTC)
 * These articles' (backdated) Expand tags usually point to another unsourced article on es.wikipedia - that doesn't solve our problem any more than the ones with no es.wiki counterpart. If somebody is going to write new articles (and find sources) from scratch, that's fine, but it can be done even more easily if the existing unsourced microstubs are deleted. In the meantime, those articles are a net negative to the encyclopædia; a lot of errors have been found despite the minuscule content, and who knows how many other errors are yet to be found? bobrayner (talk) 12:59, 23 May 2012 (UTC)
 * Let me be a bit more clear. With the census site, I meant this site, from where reliable basic data (population and province and name) can be taken. Actually this shortcut gives good info. The ones without interlangs are immediately suspicious and should be checked. In your example, Setaigües is -I guess- the Catalan name for Siete Aguas and should be deleted (/merged, but there is not much to merge)... L.tak (talk) 13:09, 23 May 2012 (UTC)

As I said I checked them and the problems are eliminated. Just need to import the infoboxes and some data now.♦ Dr. Blofeld  13:04, 23 May 2012 (UTC)
 * You mean "and one problem is eliminated", surely? The articles at incorrect names (perhaps valid redirects at most), often duplicates of existing articles (e.g. La Pobla de Sant Miquel for Puebla de San Miguel) still exist. The poor sourcing remains. The basic layout issues (the disambiguators shouldn't be repeated and bolded at the start of the article) remain. What is the benefit in keeping these articles and fixing all the issues, compared to starting them again from scratch? Fram (talk) 13:20, 23 May 2012 (UTC)
 * Because nobody is likely to start them all from scratch and the deficit of article coverage will remain. At least with the interlinks the content and data is there to transwiki as a start and references can be sought to support them. There is much Catalan vs Spanish difference inevitably but neither are "wrong", just different variations. Sure there are a few errors but I see none for Spain which are too serious. Wikipedias not going to end because of a couple of Spanish village stubs. There are far worse problems in unsourced developed articles wih more serious implications.♦ Dr. Blofeld  13:24, 23 May 2012 (UTC)
 * Even apart from the complete lack of notability and content, there are errors. As with previous controversies over mass-produced stubs, each time an editor says errors are fixed, we can go sample a random article and find another error. And so on. This is why it's best to delete the articles. If somebody is willing to spend time writing proper articles with proper sourcing, we can wait until then. I realise that some people take a more inclusionist stance, but filling wikipedia up with placeholder microstubs which only say one thing and are often wrong about that is not the way forward. This is supposed to be an encyclopædia. bobrayner (talk) 14:29, 23 May 2012 (UTC)

Thoughts
Weighing up several factors, such as intention, maturity, age, chaos caused and so on, I propose that Jaguar should come back to ANI after the first five articles he creates, the first twenty after that, and then the first two hundred after that to get the community's approval, for the avoidance of doubt creating no more articles after each "milestone" unless he has received approval. I don't think any kind of ban is appropriate yet but it may be after we see what he does next. He should be on note that he has caused a great deal of trouble, failure rates of 30% are simply not acceptable, and that it is only his age, enthusiasm, and willing to engage in discussion that has prevented worse consequences. He is also recommended to seek out mentorship. Egg  Centri  c  14:07, 22 May 2012 (UTC)
 * Thank you, I think that's a step in the right direction. I've been on Wikipedia since 2009, I don't really think I need mentorship! Coming back to ANI seems like a good idea. I've had my Autopatrolled taken away anyway so if I create any articles NPP could review them. Jaguar   (talk)  14:23, 22 May 2012 (UTC)
 * You absolutely need mentorship or at least someone to hold your hand, because you are not quite getting the enormity of what you did. Egg   Centri  c  15:38, 22 May 2012 (UTC)


 * For a variety of reasons Egg Centric, several things about this section are ill thought through. --92.6.200.56 (talk) 14:29, 22 May 2012 (UTC)
 * And I've changed the title. --92.6.200.56 (talk) 14:33, 22 May 2012 (UTC)
 * I'm in favour of some kind of community sanction but I agree this is fairly ill considered at this point. We need something that would work, rather than arbitrary points at which Jaguar would get a "report card". Not to mention how ANI would get filled with pointless "reports" about how Jaguar was doing. I would like to see see some sort of conclusion to the above discussion, then for some fresh discussion on what kind of sanction, exactly, Jaguar should have (probably an editing restriction), rather than people needing to !vote on a proposed set of arbitrary sanctions. —Strange Passerby (talk • cont) 14:36, 22 May 2012 (UTC)
 * I'm currently working on this very thing (conclusion/discussion/recommendations). --92.6.200.56 (talk) 14:40, 22 May 2012 (UTC)
 * Jaguar should be topic banned from creating any new article in mainspace, and should instead demonstrate in user space how they have taken on board all of the above. GiantSnowman 14:42, 22 May 2012 (UTC)


 * x2 No, no and no. Jaguar should be placed on an indefinite article creation ban, except under the guidance of a mentor, leading nicely to the next requirement. Jaguar must seek adoption to understand how to source articles properly. Assuming all the good faith in the world doesn't help justify approaching 15,000 virtually unsourced stubs. The failure rate is already incalculable without including the problems with the Spanish and Serbian stubs. Blackmane (talk) 14:43, 22 May 2012 (UTC)
 * I'm sympathetic to Egg Centric's position, but I must agree with 92.6.200.56 here - this particular proposal is probably not helpful right now. Personally, I would argue that a key problem with Jaguar's previous edits is that they need so much adult supervision; a solution which centres around even more supervision by other editors is probably not optimal. Further, if we're still getting denial and avoidance from Jaguar, a solution which allows even more articles to be created is just saving up more drama for the future. bobrayner (talk) 15:05, 22 May 2012 (UTC)
 * ( non admin suggestion ) As Jaguar and other seem to be opposed to adoption; an alternative option could be to remove autoconfirmed status, which would enable most editing, but would require a pair of eyes extra for article creation. There seems to be no disruption in any other field than new articles, and this won't stop him from making articles so he still has the potential to show for a few months that he is capable of creating articles with sources conforming to WP:RS. In view of his presumed age, I don't think it is fair to give him much more rope and to have to end up with stricter solutions in the end... L.tak (talk) 16:45, 22 May 2012 (UTC)
 * That's already been done. The focus here should be on how to repair the extant damage: Jaguar is guaranteed to get blocked if he dpes this again regardless of sanction, so this whole subsection is just cluttering up an already-busy ANI. Chris Cunningham (user:thumperward) (talk) 17:02, 22 May 2012 (UTC)
 * Then I am mistaken, I thought "only" autopatrolled had been removed and not (auto)confirmed. Indeed, then this section is not necessary at all... L.tak (talk) 17:59, 22 May 2012 (UTC)
 * No, no, no. Everyone has got this all wrong! I've been on Wikipedia for three years, I don't need "adult supervision" or an "article creation ban". I'm not going to mass create any more articles now. I think everyone is misunderstanding me for a good for nothing mass creator. I've brought up for GAs, done three GARs and created 2011 England riots and Operation Ellamy. Those articles have been expanded so much.


 * To be honest I think I've ran out of time. I would rather edit with full rights than being kept under a close eye by ten different people. If it is going to carry on like this, and all my articles, all my hours of hard work, get deleted, then I will leave. I have tried but I failed. Jaguar   (talk)  17:27, 22 May 2012 (UTC)

First of all I must say I think the way Jaguar is being treated here is quite appalling. Failing to adhere to WP:AGF is by far the worst problem on wikipedia than somebody creating stubs or understanding why they do so. I'm witnessing an extreme reaction here to subjects which are all verifiable and indeed combined cover several million square kilometres of the earth's surface, even if in Chinese and indeed verifiable on the reference provided in the articles, although there are clearly quite a few which don't even link to the region on the website given which I must say was very careless of Jaguar and I prompted him many times to check the links before starting a new batch. I'm failing to see this as a huge issue, given the lack of traffic the articles get and I think a bot could reasonably be programmed to fix the ref error links and if possible the article soveridden with something similar to that bot on Chinese wikipedia, hopefully with more data. Yes, the articles are placeholder stubs and the extent of work needed to expand them is enormous but if wikipedia is to even attempt to achieve anything near even world coverage one cannot ignore millions of square kilometres, especially when there is already information on the web in Chinese to make them all encyclopedic and valuable. Contrary to what Bob Rayner thinks, I don't continue to endorse sub stubs on geo places, especially as more and more sources are becoming available which can produce some meatier stubs with useful facts. However, in the long term of wikipedia what Jaguar was attempting to do is to try to plant seeds for growth and wants to see them flourish which is highly commendable. I still sense a moral panic and overreaction here so something which could quite easily be fixed by an automative tool and build upon what we've got.♦ Dr. Blofeld  18:06, 22 May 2012 (UTC)
 * Would you please answer my question below? Thank you. --92.6.200.56 (talk) 18:11, 22 May 2012 (UTC)

If you look at a global map coverage of major world settlements on wikipedia in relation to the country is generally not bad, although most African nations of course still pretty sparse but we generally have most sizeable towns covered. But if you examine a map of China and Russia in particular the coverage on here is shocking beyond belief. In relation to the population of China representing 1/5 of the world population roughly and the sheer number of towns with sizeable populations, some with millions still missing on wikipedia I personally see improving coverage of China one of the most important priorities for the whole website. Whilst Jaguar's new stubs are far from ideal as a start, I appear to be the only one who can really see the importance of having encyclopedia articles on them. We could potentially have hundreds of articles on topics within each township. Some townships cover tens of thousands of square kilometres. China and much of Russia is a blackhole on wikipedia in terms of knowledge. We should really be trying to organize something which can productively and efficiently produce articles on them and overide the articles. Nuking them all in my opinion is not a move forward, problematic or not. They could be recreated by a bot, as indeed could most articles on species. I agree wikipedia articles are best manually written and researched but in areas where we literally have a frightening lack of coverage we need something more powerful to at least give us a decent start. People here seem perfectly happy to delete the subs which are rendered useless for their lack of content and their minor errors of formatting/sourcing but nobody here is really thinking encyclopedically apart from one editor who mentioned the Chinese bot in just how important it is for the project to be improving coverage in this area. We should be spending more time discussing a way to cover Chinese townships adequately rather than continuing to attack and threaten the editor.♦ Dr. Blofeld  19:11, 22 May 2012 (UTC)
 * No one has "failed to AGF" Jaguar, he seems honest in what he was attempting. The problem is that he mass-created a bunch of articles that we don't have the information to verify, and likely do not pass WP:N. Mass-creation isn't a good idea anymore, due to issues like this. There's no rush to cover a subject. If you really want to expand our coverage of China, WikiProject China would be the place to get more people interested. &mdash;  The Hand That Feeds You :Bite 20:47, 22 May 2012 (UTC)


 * This is a little OT, but in reply to Dr Blofeld, the biggest problem is sourcing. My ancestry is chinese and my family comes from such a "township", but you'll find thinking in the western sense of a township is where this all falls down. "Townships" near major urban areas are effectively suburbs, my family's ancestral village is one among a thousand or so that has been amalgamated into the city. You'd be lucky to find any sources about such places at all. The best you'll likely get is a bunch of articles barely out of stub grade with minimal information beyond geography, location, population. However, that being said, the best way to cover it wouldn't be to create a series of articles about each individual township, but to amalgamate them into an omnibus style per region. Blackmane (talk) 00:07, 23 May 2012 (UTC)

I beg to differ. If I can produce an article on a rural township in Yunnan then any township could be adequately written about. Some sources exist for most in Chinese.♦ Dr. Blofeld  07:08, 23 May 2012 (UTC)


 * I'm sure there are. What I meant was that many of the ones that Jaguar has created are part of cities already and there is little point in creating thousands of stubs about "townships" which are now little more than a suburb. External to a major city, sure, I agree that an article could be created. Blackmane (talk) 10:07, 23 May 2012 (UTC)

Proposed Resolution
We give Jaguar a specified timeslot, in which Jaguar has to fix every stub. Let's say 2 months randomly. In that time he would have to: Fill in Chinese, correct references, add coordinates, add more sources. That would judge if Jaguar actually wants to fix them, and thinks they mean something. Of course, other editors could help. Just a proposal, as always. --Thine Antique Pen (talk • contributions) 19:31, 22 May 2012 (UTC)
 * Oppose, he should have been well aware by now that the articles he created were less than optimal to say the least (I e.g. asked him in October to provide a correct, direct reference in the Chinese stubs, not the generic one he kept on using), and even yesterday he claimed that there was nothing wrong with the Spanish stubs, which has been shown to be wrong. He's had ample time to change his ways, he has had the opportunity to show that he can be trusted to check and correct his stubs thoroughly, and he hasn't done either. Fram (talk) 11:48, 23 May 2012 (UTC)

In all honesty I doubt a ban will be necessary as if you delete those articles I doubt Jaguar will ever edit wikipedia again. you'll have scared him off. Emirjp another sub stubber as disappeared too. If you look back he's a decent editor capable of nice articles on hundreds of Somerset etc. I think its the systematic bias thing he got wind of when he reviewed the Adar oilfield and wanted to help towards it. ♦ Dr. Blofeld  12:13, 23 May 2012 (UTC)
 * [Non-Admin] Comment [If in inappropriate place please move]. This is the first comment I have made, I'm not familiar with any prior discussion other than reading the above. I selected a stub at random, Xianrenzhuang Subdistrict, and my immediate reaction was (a) without Chinese characters it'd be quite difficult to identify - but I took a guess at "wizard village road" 仙人庄街道, and it was. (b) a rural subdistrict or "road" 街道 is a very tiny administrative unit, not covered by zh.wp, though it may be by Baidu. In this case the zh.wp article for the admin unit one level up, zh: 鼓楼区 (开封市) has these 7 or 8 "roads" only in black html, as names, the English Gulou District, Kaifeng doesn't even mention them (no reason why it should) and Googling "[Shanrenzhuang Gulou"] picks up zero. (c) the actual source in the stub is a html link to heaven-only knows what, and with "Fujian Province" pasted over the link in Chinese - which is the wrong province. Okay, so that's the comment based on a sample of 1. I would suggest that a helpful encouraging project for Jaguar would be to work only on fixing stubs with interwiki winks to zh.wp articles which may be none. That I didn't check. There's no value fixing roads and subdistricts with only a few 1000 people at a level lower than zh.wp. In ictu oculi (talk) 13:44, 23 May 2012 (UTC)
 * That's partly what I mentioned earlier, and what Dr Blofeld and bobrayner were discussing in the section I hatted below. Blackmane (talk) 14:18, 23 May 2012 (UTC)
 * Yes, I see I scrolled through it and missed it, well at least my findings agree with yours and Bobrayner.
 * @In ictu oculi. 1) A 街道 is certainly not rural and not a road. Subdistricts are instead urban administrative units. 2) I wonder why Googling ["Shanrenzhuang Gulou"] turns up zero. Apparently, either you made a typo, or somehow 仙 (Xian2) = 山 (Shan1). The failsafe alternative is to search in Chinese: i.e. "河南开封鼓楼区仙人庄街道", because Google Maps does not translate "乡" into "Township", but instead appends "xiang" at the end of a name. 3) "Gulou District, Kaifeng doesn't even mention them (no reason why it should)". Appalling. By the same reasoning, no District article should contain at least a list of the constituent township-level divisions. Also you should know EN-Wiki is nowhere near ZH-Wiki's level when it comes to nearly anything Chinese. 4) In other words (Blackmane included), don't deem something a problem when you have not double checked yourself. GotR Talk 06:56, 24 May 2012 (UTC)
 * Hi GotR. maybe I've misunderstood, do you want Xianrenzhuang Subdistrict to have an article on en.wp? Apart from postcode changes and names of villages in the subdistrict the sum total of government information about this tiny district is this "Area of 35.3 square kilometers, a population of 10000":
 * 1) yes sure these subdistricts called "street" or "road" aren't all roads or rural, but they are a tiny admin unit. 2) You're correct, duh duh duh, I was tired and searched Shanren for Xianren, so, duh, yes [Xianrenzhuang + Gulou] turns up a few English addresses imbedded in Chinese pages. 3) I can see the value if Jaguar would add into en.wp Gulou District, Kaifeng the 7x names of the 7 "road" subdistricts mentioned in zh.wp Gulou District, Kaifeng but creating 7x sub-sub-stub level "road" subdistrict articles at a level more detailed than zh.wp seems pointless. What can anyone possibly say about an admin unit this small. 4) Blackmane and Bobrayner seem to have grasped the situation in relationship to zh.wp very well without being able to read Chinese, I find that pretty impressive. In ictu oculi (talk) 08:31, 24 May 2012 (UTC)
 * Actually, I can read chinese, also something I mentioned further up :p @GotR I've no issue with articles however small (although I find one line stubs an eyesore, but that's just me). Let me re-iterate the problem. It's not lack of information, it's lack of properly sourced information, which is the crux. Language wise, "xiang" is a difficult word to properly translate into english. Some are mere villages while others are large towns, which throws a spanner into the works when juggling notability. For example, where my family comes from, it's a fairly small village but it's been absorbed into the city as it's expanded, so it's more like a commune or small suburb than it is a town/village anymore. I'm a copy-editor primarily and at the moment I'm working on an article, once I'm finished up with that, I'll start poking about on the list that bobrayner has set up and see what can be done with it. Blackmane (talk) 10:44, 24 May 2012 (UTC)
 * Actually, I can read chinese, also something I mentioned further up :p @GotR I've no issue with articles however small (although I find one line stubs an eyesore, but that's just me). Let me re-iterate the problem. It's not lack of information, it's lack of properly sourced information, which is the crux. Language wise, "xiang" is a difficult word to properly translate into english. Some are mere villages while others are large towns, which throws a spanner into the works when juggling notability. For example, where my family comes from, it's a fairly small village but it's been absorbed into the city as it's expanded, so it's more like a commune or small suburb than it is a town/village anymore. I'm a copy-editor primarily and at the moment I'm working on an article, once I'm finished up with that, I'll start poking about on the list that bobrayner has set up and see what can be done with it. Blackmane (talk) 10:44, 24 May 2012 (UTC)

Prevention
Continued from above. We've made progress in the proposed deletion debate. It's important to know what failed, if we're to achieve real progress and resolution. We've existing policies and measures (regarding mass creation tasks) designed to make them go a lot smoother than this did and to prevent such problems in the first place. N.B. I've asked to return to this thread.

Dr. Blofeld, above I asked how come you and Jaguar hadn't talked about wp:masscreation policy. I didn't really get an answer; you replied he wasn't a bot, the policy writers didn't assume good faith and you felt stubs valuable or useful. How come you didn't talk about the large-scale creation policy? --92.6.200.56 (talk) 17:37, 22 May 2012 (UTC)

Get yourself an account and we can talk. Above all I find it amusing I continue to be implicated in mass stubs... ♦ Dr. Blofeld  18:39, 22 May 2012 (UTC)
 * You're clearly not going to answer. You've avoided doing so twice above and now again. OK. --92.6.200.56 (talk) 18:49, 22 May 2012 (UTC)
 * No. Why would I be compelled to discuss an invented purported policy over something which I'm not involved in? Editors have their own brains to think for themselves. Contrary to me being a criminal mastermind I'm not some puppetteer directing on here. If there was really some strict "policy" restricting manual article creation, why I ask has Jaguar been able to generate articles for near 6 months without even a glimmer of a mass complaint?? Why is it now? ♦ Dr. Blofeld  19:00, 22 May 2012 (UTC)
 * Oh I think you're rather underplaying your role. --92.6.200.56 (talk) 19:04, 22 May 2012 (UTC)
 * (I've not reached the scale of mass creation you have, Dr. Blofeld yet! I don't think I do bad against WP:MASSCREATION, I make up to 25 articles at once, check them, and make more.) I think Jaguar might have been thinking that more articles would be better for his stats, and maybe wasn't thinking about the encyclopedia that much, thoughts? --Thine Antique Pen (talk • contributions) 18:46, 22 May 2012 (UTC)

Rate/speed and number of new articles is irrelevant if the stubs are correctly sourced and contain some sourced useful facts as a start. This is why I oppose "policies". If you're going to complain about that then the whole fabric and way in which we approach wikipedia and its very openness needs major reform as let's be honest the average article is utter crap, even if we're gradually improving. And frankly I see much worse problems in stale unsourced articles than could ever be created by shortish sourced new stubs. That said, a bot really is needed for generating initial articles on populated places and if programmed correctly could do it far more efficiently and quicker than a human could. The problem of course is lack of editorial interest in expanding them, like the Rambot articles on the US.♦ Dr. Blofeld  18:51, 22 May 2012 (UTC)
 * Yeah, that's one thing. All the stubs do is just tell you what something is, and they mostly have the lowest notability possible that's why people don't edit them! --Thine Antique Pen (talk • contributions) 18:56, 22 May 2012 (UTC)
 * So you don't like or 'oppose' a site Policy. A Policy created to prevent precisely these problems. From what I see, you act to disrupt Wikipedia to make a WP:POINT, and worse, you involve a young guy to do it. --92.6.200.56 (talk) 18:58, 22 May 2012 (UTC)
 * If you were commenting on what I said, I'm just stating what the reality often is with my stub creations, they don't get edited by many people. --Thine Antique Pen (talk • contributions) 19:02, 22 May 2012 (UTC)
 * Nope, wasn't commenting on what you said. 92.6.200.56 (talk) 19:04, 22 May 2012 (UTC)


 * Jaguar's contribs are interesting. For a long time he quietly worked on local interest articles – even helped get one to good article status, and the occasional film/game. Any article creations were in keeping with that and involved nothing large-scale, unreferenced or high-speed.
 * [ 21 August 2011 was the day that everything changed.] That day he ga-reviewed an article, on an oilfield in the Sudan. Your article. That very day, he began to create microstubs on Spanish villages. Unreferenced stubs, containing a trans-interwiki tag. They bore a startling resemblance to articles you've created and argued for in the German politicians substub incident. Many were and still are unsourced substubs on the Spanish-wp as well. He was quickly stripped of autopatrolled for creating the unref'd stubs. That's when you told him "no point in adding a reference", it was "redundant" and an interwiki tag was enough. Moments later adding  "sometimes this is the easiest way to get English wikipedia working towards such content at all costs ".
 * The sea-change in focus and its timing is striking. We'll probably never know if you advised the mass-creation by email or otherwise. But we certainly know what prompted the Chinese stubs. --92.6.200.56 (talk) 19:10, 22 May 2012 (UTC)

I encourage anybody to create articles on Chinese townships, for sure, so long as they are properly sourced. It really is a desperately lacking area of wikipedia which I believe needs major attention. You've also taken what I said out of context. I said it was pointless adding a source to a website for the sake of it without directly linking and sourcing actual data. The link to a website with no verification is pointless, you may as well just create them unsourced if you're not going to bother to add actual information. You continue to try to blame me for this incident. Its also amusing to me you have nothing better to do with your time than fish around on things I said. Quite pathetic in fact and the fact you can't reveal your true identity by logging into your account.♦ Dr. Blofeld  19:29, 22 May 2012 (UTC)
 * The highest amount of contribs to his talkpage after him come from you, Dr. Blofeld. In the first or so comment you encouraged him to violate WP:V saying "No point in adding a reference" to an article he makes if it had a translation tag. ...This was in a thread called "Unreferenced stubs" where he'd just lost autopatrolled for creating stubs without a reference.
 * The one who enticed him to ever start creation of the Chinese nanostubs was you. At the time he wasn't working on anything like the scale of this or on Asia period. Until you advised: "could next create stubs on List of township-level divisions of Hebei and the others", "just reference them to www.xzqh.org",  later mentioning you couldn't speak Chinese.
 * Everybody is responsible for the edits they make. But let's not be distracted by this simple truth. You've been a constant whisper in the ear of an impressionable young editor. You told him sources don't matter if there's an interwiki-trans tag. It was you who told him to just reference to xzqh.org, a key problem we've seen. You directly urged him to create the stubs we've been discussing. Your part in this isn't trivial: You've been a driving force. --92.6.200.56 (talk) 19:20, 22 May 2012 (UTC)


 * mmmm delicious drama Egg   Centri  c  19:32, 22 May 2012 (UTC)
 * More diversion attempts. You know about and "oppose" the mass creation policy. Or as you call it "invented purported policy". You've been repeatedly told about the mass creation policy and referencing, such as here where you wrote "As for a lecture about referencing, p-lease, tell me something I already don't know. Occasionally, just occasionally WP:Ignore all rules is neccessary". Playing the part of an experienced knowledgeable editor you've approached multiple contributors, often young. Anticipating a desire for approval and desire to make a difference you tell them they're doing good works. Despite being aware of the Policy, you deliberately neglect to bring up the requirement of pre-approval & consultation for mass-creations, in your drive to get such stubs produced "at all costs". The cost is plain to see. Massive disruption, heaps of people spending time trying to clear this up, the editor saying all this is "most upsetting thing I've ever heard" along with us having to watch as he rushes around like a Brave Little Toaster trying to fix problems we know are beyond all his efforts. --92.6.200.56 (talk) 19:39, 22 May 2012 (UTC)

So you're saying I should be vindicated and burned at the wiki stake (or wiki toaster) for my crimes? Aren't we, eh, getting a little personal? Its quite amusing to me you continue to paint me with the sub stub brush. My editing no longer revolves around mass stubbing on the scale it used to be in the days where wikipedia needed a huge boost in trying to work towards systematic bias and it might just be possible my whole project outlook has changed quite considerably. Referencing is dramatically improving, to the point it is is now possible to create decent new articles about quite a lot of non anglo places. If there are quite a few editors who find me inspirational, they must be seeing some merit in my work on here and vision for the project which far exceeds the small minds of people such as yourself who don't even have an editing account on here. Big words for an IP and wiki nobody. ♦ Dr. Blofeld  19:47, 22 May 2012 (UTC)
 * There is no witch hunt here. I don't find this amusing and it's real clear Jaguar hasn't either. You've said a few times you don't nowadays get involved in stubbing. This comes across as gamesmanship when you "significantly encourage" others to do so. Similarly in the older discussion just linked you claim you never run a script or bot, yet said above you do so by proxy i.e. commission others to make your AWB edits. The thread isn't about decently-referenced accurate articles about places. I would support sanctions against you specifically designed to prevent you directing anyone else to undertake mass-creation stubbing without letting them know it must be pre-approved and possibly require community input through WP:VPPR/WikiProjects. A repetition of this is no good for anyone. --92.6.200.56 (talk) 20:04, 22 May 2012 (UTC)

Arbitrary Break 1
Dr. Blofeld, you are easily the most prominent advocate of (and indeed participant in) the mass-creation of geographic stubs. You pretty much single-handedly forged an agreement that, in principle, Wikipedia should endeavour to mass-stub settlements around the world. This is extraordinary enough in itself in that in the case of literally anything else except for human settlements the community does not consider existence to equal notability. Your exceptions were hard-fought and predicated on very strong guidelines: specifically, it must be proven that these settlements exist by reliable sources in every case, and the stubs must be accurate. We cannot afford to weaken those rules, given the extreme lenience under which they are afforded this type of article. Surely you must understand this given the drama which surrounded the significant effort which went into est ablishing these rules in the first place? Chris Cunningham (user:thumperward) (talk) 20:15, 22 May 2012 (UTC)

You're completely missing the point. Rate/speed and quantity are utterly irrelevant if the articles are of a reasonable quality and are accurate without errors. Its the quality/accuracy of the stubs which is an issue and only when there are a mass of faulty stubs does it become a major cause for concern. Editors like Ruigland and Nielsen create a huge number of sourced stubs with some info, its utterly irrelevant how fast they create them, if they are accurate and sourced then they should feel free to create what they want. Why should they be hampered by imaginary laws restricting the very core principle of the project? As for me. Almost half way through 2012 now. Have you even seen how I operate now?? If you truly study my edits this year in comparison to previous years you would notice a marked change and see my vision is changing largely due to the developing information situation on the web which means the hit and hope method is no longer appropriate when in past times it seemed the most plausible way to try to attack systematic bias at all costs. The reality is few people are expanding them. Add to the fact I've gradually accepted wikipedia is massively underdeveloped in parts with millions of missing articles and however hard I try I will never be able to cover a small fraction of it in the way I'd like to. You no longer get placeholder stubs from me and you no longer see me specifically asking people to create sub stubs without content. You ignored the fact that most of my comments to Jaguar were negative in regards to the township articles and asking to be more careful. I had a discussion with Aymatth2 and DGG quite some time ago and I fully agreed with them and you've not seen me create like that since; if I have once or twice it was to rid of red links in an article... There are enough editors doing the mass stubbing now anyway. Am I to be held accountable for User:Doma-w, User:Hockeyben, User:Nielsen, User:Ruigoland, User:Carlossuarez who currently create hundreds of times more stubs what I produce daily?? I encourage people to produce fleshy stubs with a few sources if possible sure but the honest reality is that I am completely different in function to how I used to be and your view of me is incredibly dated. Just look at the new pages, editors with their own brains are mass creating, and if they're properly sourced with some info why stop them?.. There is a considerable difference between faulty, unsourced sub stubs and well sourced useful meaty stubs. You must really view me as a powerful force on here with a lot more respect for me than you let on if you genuinely think I am accountable for all editors creating new content enmasse. I'm incredulous that I still continue to be lured into debates about sub stubbing despite the fact I've clearly distanced myself from the placeholder factory method for quite some time and anybody who has been regularly keeping tabs on what I edit for over a year will be the first to note this. My desire is for wikipedia to have a full length sourced article on every settlement on the planet ultimately yes, but that doesn't mean I'm happy with any geo article being an empty stub. You're so wrong! ♦ Dr. Blofeld  20:57, 22 May 2012 (UTC)

We agree on a surprising amount. I agreed with you earlier that you'd pulled Jaguar up about errors &c. in his township articles. I definitely agree with you systematic bias is an issue we struggle with. I've no real objection to geographical articles, not even to small stubs! It's completely true you're not responsible for all the stubs that get created. As I've said, I think in this case you did have larger input than might have been the case elsewhere. You wrote imaginary laws as a reference to the mass creation policy. It's perfectly alright to disagree with policies or notability or how they get applied sometimes, we *all* do. Personally I think wp:masscreation is one that got stronger consensus than most, it was re-examined last year and extended to all namespaces such as categories. It might help if I explain how I think this would've gone if it'd been brought up and followed so to speak. The articles very likely would've been made, and very probably by Jaguar. Rate/speed and quantity might well have been nearly identical. The difference would be, the endeavour would've got closer supervision, a small trial run allowing that batch to be checked before making more, and consultation at brfa/wikiprojects before and during the trial run so hopefully experts like our Chinese-speaking friends above could've said, and before huge amounts were created, hangon this source isn't great but here's a great alternative or lay it out this way instead and wikilink to that instead. Something like that. The end result would've been simply a healthy amount of new accurate articles combatting systematic bias. Your desire to address systematic bias and encourage new content is admirable. At the same time, being only one guy you can only help so much. The consultation means more people who might know lots about the topic can give thoughts or advice early on, helping it to go smoothly and well. --92.6.200.56 (talk) 21:45, 22 May 2012 (UTC) A glimmer from you in terms of understanding my approach on here and the actual me I hope. I have never thought one line unsourced stubs really acceptable. And its very frustrating to go through a category and find them all flooded with empty stubs and anything decent unfindable amongst them. However, in the past I've mass created stubs I was in the "wikipedia is empty" frame of mind at a time when I saw massive inbalances in coverage and given that I did not have the time to write them I was convinced that the best possible thing for the project would be to highlight as many subjects which appeared notable as possible and start them and think about the overall picture in the long term. And I always made sure the errors were minimal and any major ones were corrected. That was what I meant by the "break all rules" approach in that at times it seemed very important to stub missing rivers or towns which I knew had information which could easily expand it by anybody. I saw it as planting seeds for growth and very necessary for trying to spread knowledge as quickly as I could at the time, especially with sourced info already on another wiki. And if you examine the vast majority of my created articles and indeed AFDs of my articles, almost always kept because I made some sort of assessment on notability before starting them. Unfortunately it often backfired and as a result I still continue to be implicated in the mass stubbing thing and seen as its Number 1 adherent and still seen and disliked by many who still think of me as an irresponsible clown who doesn't care about building a decent website, as if I really want wikipedia to be full of useless sub stubs. The idea is that they were all expanded and knowledge is spread, a pretty desperate attempt to try to get things covered more evenly. Times are changing now and I now view wikipedia as an actual encyclopedia rather than purely a development project in which so much information is becoming available in google books, JSTOR, Highbeam etc that my time is better spent writing content, which as I always said I prefer and we have a duty to be nurturing existing articles.
 * Why are places supposedly exempted from the GNG? Why is it that a mere mention of a placename in some listing or gazette is considered sufficient to create an article, even though we have nothing else to say about the place, and this deviation from the rules serves only to enable rote creation of thousands of crappy unmaintainable stubs from the flimsiest of sources? (And, as we've seen, from sources which don't even mention these subjects). If somebody tried this trick with articles about people or books or businesses, they'd have a talkpage full of CSD warnings before the end of the day. Why are settlements supposed to be exempt from this? Who approved this? bobrayner (talk) 21:53, 22 May 2012 (UTC)

@Bob. Not really a believer in "inherent notability" but geographical settlements and features usually have something substantial which can be written about them and some coverage in sources, although admittedly much of Asia and Africa still very poor. Gwebin for example, would have been your xxx is a village a few years back. Increasingly so we'll get more coverage in books. but my approach now is lets work with what currently exists. There is such a vast amount in google books which could be used to source missing topics it seems the best way to move forward. As for notability of Chinese townships, some of them have millions of people and may cover a few thousand square kilometres. In terms of land area I'd say they are very notable, and indeed one could write about numerous landmarks within them in the sort of coverage we get in the UK, but may not be reflected in sources. There appears to be enough info right now though in Chinese to make fleshier stubs on Chinese townships with population/basic crops, area etc. I'm having connection difficulties right now but I'll show you some Chinese townships I have expanded or started to start class tomorrow. ♦ Dr. Blofeld  22:20, 22 May 2012 (UTC)
 * I've read the same TLDR Blofeld drama at previous discussions, and the same flailing around for different excuses to keep the fait accompli whilst refusing to acknowledge problems (example); but it keeps on happening and we still get more and more crappy one-sentence stubs based on any source which has a list (or which the editor.pretends has a list) of settlements. Creating them isn't countering systemic bias; if that were the case we could solve our CSB problem by creating an article for everybody in the Nairobi phonebook (which would be pretty similar to what you created and to what you encouraged Jaguar to create) . Building actual encyclopædic content on notable subjects is the way to counter systemic bias. I've already tried to do this - but only a few articles, not ten thousand, because looking for sources, reading them, and writing content takes much longer than copy & pasting one string. If these settlements are as notable as you say, why haven't you brought more sources and written more content? bobrayner (talk) 22:30, 22 May 2012 (UTC)

Excuse me me, why haven't " I brought more sources and written more content? The articles are not mine!!! I'm not Jaguar and regardless of the fact you think these people are brainless morons controlled by me I'm not responsible for them. I've been the one actually expanding/starting a few like Luoji Township, Shangri-La County and Fuzhou, Wafangdian and expanding articles like Tunchang and Xainza County. Muang Sing on a small Laotian town on the Chinese border (and yes I went out of my way to make a flick agremeent and upload the pretty pictures), still complaining?? Give me a break! I don't need to "flail around for different excuses" to excuse what I produce these days!! Perhaps if you read every article in User:Dr. Blofeld/DYK for instance and many thousands more decent new articles which I never bothered to nominate you'd see that yes I've been more productive than anybody in producing   actual encyclopædic content on notable subjects and most people know this here and respect me for it. Aside from the fact I have no obligation to add anything at all to wikipedia. However, I still continue to have grudges held against me by people who just cannot move on.... If this doesn't convince you I don't know what will, a proposal to ban empty placeholder sub stubs from being created. ♦ Dr. Blofeld  05:44, 23 May 2012 (UTC)
 * We already have a rule which prevents all this crap; it's the GNG. We just have to start enforcing it again. Write an article which uses substantial discussion of a topic by multiple sources? Great, it passes the GNG. Use a list as the basis for copypasting thousands of  " is a settlement in..." ? That fails the GNG. bobrayner (talk) 08:12, 23 May 2012 (UTC)

Clearly we don't. Because empty placeholder stubs with one source (even if just a website link) took 6 months before anybody complained. Jaguar's stubs are actually verifiable on that website if one finds the Chinese name and exact link and MASSCREATION currently makes no mention of manual creatiyon, in fact it is under Bot policy.♦ Dr. Blofeld  08:36, 23 May 2012 (UTC)
 * Dr Blofeld, perhaps there has been some misunderstanding. I pointed out a rule which requires substantial discussion by sources. You repeated the canard that Jaguar's stubs are verifiable because those settlements are probably in a list somewhere (although many aren't verifiable at all because Jaguar used a "source" which doesn't actually mention those places, and many have no source at all; looks like Jaguar was given a bum steer). Notability and verifiability are not the same thing. It's quite simple. bobrayner (talk) 08:44, 23 May 2012 (UTC)

Its pointless adding a reference when you are not using it as a reference and it isn't immediately verifiable. They may as well be unreferenced if you can't verify the article in the external source. That was my point. Not that it is really acceptable producing unsourced stubs. It is more irritating to have a source linked which isn't actually there than none at all. Stop trying to blame me for everything. The fact it took 6 months for anybody to complain and Jaguar was actually given barnstars by several people complimenting him on his stubs says it all. "Thanks for your run of new China-related articles. A significant contribution that truly improves the encyclopedia." Northamerica1000(talk) 12:03, 11 February 2012 (UTC), "Great Job creating stubs, Keep it up" Solomon7968 (talk) 16:19, 20 April 2012 (UTC), Special Barnstar: "Thanks for your contributions to Wikipedia! " for his efforts... ♦  Dr. Blofeld  09:07, 23 May 2012 (UTC)
 * Yes; it's pretty unfortunate that people praised Jaguar for mass-producing crap. If proper feedback had been given earlier, much of this drama could have been avoided. This is why I feel Jaguar has been given a bum steer. It's unfortunate that Jaguar's stubs are still defended with arguments that are impossible to reconcile with our existing rules. However, I fear we're going off-course here - and I have no interest in sniping at you, Dr Blofeld - would you mind if we hatted this section and let AN/I focus on the original question of what to do with Jaguar's articles? You've already created a thread elsewhere to discuss possible rule-changes in future... bobrayner (talk) 09:18, 23 May 2012 (UTC)

Yes, I would appreciate that. Me I'm off to enjoy the sunshine for a bit and will return later.♦ Dr. Blofeld  09:44, 23 May 2012 (UTC)

Just a little followup... Chris Cunningham suggests that "(Blofeld) pretty much single-handedly forged an agreement that, in principle, Wikipedia should endeavour to mass-stub settlements around the world. This is extraordinary enough in itself in that in the case of literally anything else except for human settlements the community does not consider existence to equal notability". I've been trying to find where this is documented. Can anybody assist? Where is this exception from WP:N written down? bobrayner (talk) 14:12, 27 May 2012 (UTC)
 * Notability (geography), although that only has essay status, rather than guideline status. —Strange Passerby (talk • cont) 14:15, 27 May 2012 (UTC)

Hello from a voice of experience
As someone to whom cleaning up large messes is not new, I have some questions: Uncle G (talk) 13:22, 23 May 2012 (UTC) indent=1 |
 * Where did this 8000 figure come from? It seems to have come from Jaguar xyrself.  Have any of you checked it?  Blackmane said "10,000 or more" above.  Where did that figure come from?  Where's a sub-page with an article list?  We need one, whatever we decide to do.
 * Where did this 20%–30% error rate figure come from? That also seems to have come from Jaguar xyrself, with everyone in the preceding discussion accepting it uncritically.  An independent check by someone else is a very good idea.
 * You can look at Jaguar's entire list of created articles here. Scroll down and you'll see where the names changed to Chinese at #8166.  If you go back to the very top of this discussion, you will see that it was initiated by an independent check. Scroll down a bit for a numbered list of issues.  Here is an independent check of five articles, explaining their problems.  Here is another analysis of the non-China articles.  That 20–30% error rate was a guess, but independent checks indicate it is much higher. —Torchiest talkedits 13:53, 23 May 2012 (UTC)
 * I had already read the discussion to this point, and the fact that basic concrete data are missing and everyone is getting more and more worked up as the figures being bandied about go up by almost a thousand per day, is why I asked. As I said, I've helped deal with several of these situations before over the years.  You'd be wise to heed my repetition of what Qwyrxian asked but never got an good answer to. There is no independent check in this discussion.  There's a few people mentioning a few tens of pages.  That's nowhere near the 20%–30% error rate that everyone has blithely accepted without checking from the creator of these articles, even though from the discussion its clear that the creator doesn't see the errors that others see.  To actually check you need to take a sizeable sample and count.  No-one seems to have done that.  Using five articles as the basis for an estimation is ludicrous. And your vague handwave in the direction of a toolserver query is not an list of the articles under discussion.  Jaguar created Tom Clancy's Ghost Recon: Shadow Wars, for example.  Create a proper on-wiki list of the articles that need dealing with, in whatever manner is decided.  That's how we always approach these things: methodically.  A list will give us accurate information, rather than the different figures that are several thousands apart that different people are waving about in this discussion. Uncle G (talk) 15:00, 23 May 2012 (UTC)
 * That Ghost Recon article is from much earlier than the mass created Chinese settlement articles. I don't understand where you are getting the notion that "the figures being bandied about go up by almost a thousand per day", when there are only two numbers: 8200, which is in line with the actual figure of 8166 Chinese settlement articles, and 10,000, which is close to Jaguar's total creation count of 9576.  The reason the number was increased was because it was determined that it was not only the Chinese settlements that had issues, but also the Spanish and Serbian settlement articles, of which there are about 500 each. —Torchiest talkedits 16:12, 23 May 2012 (UTC)
 * I proposed creating an article list earlier, but nobody really seemed interested. It's not totally straightforward; one reason being that Jaguar has also created a few more substantial articles which I wouldn't want to delete. Will put in a little more work later today... bobrayner (talk) 14:43, 23 May 2012 (UTC)
 * Good idea, manual or bot though? GiantSnowman 14:58, 23 May 2012 (UTC)
 * I've done a quick count from Torchiest's link. Before late summer 2011 his focus was local interest and cultural topics. His edits were strong, including to good article level and without mass-creations; any created during that time are believed fine/improvable & cover those topic areas.

Total page creations: 9576.

Total mass-creations (MCs): 9250
 * Of the MCs: x8166 China (#1-8166)

x545 Serbia (#8167-8711)

x525 Spain (#8712-9240 (529, but I exc. four British Iron-age pages created during that time, which are fine))

x14 Burma (#9263-9276)
 * They were created consecutively so are easy to separate, with occasional unrelated creations as stated. --92.6.200.56 (talk) 14:56, 23 May 2012 (UTC)

List
I created a list at User:Bobrayner/StubList then discovered that wikipedia doesn't like tables with 10000 rows, each transcluding a template. :-) Will provide a Google spreadsheet or pass an Excel spreadsheet to anybody who needs it - whichever is most convenient.
 * The article creation campaigns weren't completely segmented; sometimes there would be a few articles on one subject in the middle of a few thousand articles on another subject.
 * After accounting for that, and a couple of other quirks, by my reckoning there are 8166 "China" articles, 525 "Spain" articles, 542 "Serbia" articles, and 343 misc (including some UK placenames, but I don't plan to investigate those at this point). bobrayner (talk) 17:34, 23 May 2012 (UTC)


 * Oh! I just this minute came to post here that I'd created one at MCGeoList. The initial creation was with all articles inc. the good-quality non-MC ones. --92.6.200.56 (talk) 17:42, 23 May 2012 (UTC)
 * Yours is a lot more sophisticated I see. I modelled mine on User:Anybot/AfD, which used a bulleted & wikilinked list. --92.6.200.56 (talk) 17:44, 23 May 2012 (UTC)
 * After hitting a few technical limits I've split it into User:Bobrayner/ChinaStubList and User:Bobrayner/StubList and got rid of the pretty tables &c so it starts to look more like your list anyway. Should have stuck with Excel... bobrayner (talk) 20:01, 23 May 2012 (UTC)


 * There are 8165 Chinese stubs (Bentworth Hall was created in the middle, and Lingbei Subdistrict is a redirect to Lingbei). 8 are dab pages containing a link to a renamed stub (or two).  50 others have been renamed since creation.  65 pages have interlanguage links (all the ones I saw were added by User:Pengyanan), from which some of the missing basic info (like names) has been filled in.  Kanguole 00:00, 24 May 2012 (UTC)
 * I noticed yesterday accidentally that I was editing a page which made it in the list (Gornji Matejevac). I will take care of this article, and it has already more info than the first version. Would it be a good idea to remove it from the list as I guarantee its quality in the near future? Or somehow mark it in the list, in the case a mass deletion decision would be taken?--Ymblanter (talk) 06:02, 24 May 2012 (UTC)
 * That seems reasonable. Sadly, it's not practical to have detailed tables on-wiki (due to technical/size constraints) so we just get a flat list that doesn't show nuances easily. I'm happy for editors other than Jaguar & Blofeld to take individual articles off those lists if they can vouch for improvements in content & sourcing. I'll remove a few, too. bobrayner (talk) 07:32, 24 May 2012 (UTC)


 * I've taken the liberty of adding some info to User talk:92.6.200.56/MCGeoList, which is easier to handle than Bobrayner's list. In my opinion we should focus on the Chinese entries.  Kanguole 09:38, 24 May 2012 (UTC)
 * Since others are editing User talk:92.6.200.56/MCGeoList, I'll work on that too, to avoid duplicating efforts. Will redirect mine there. bobrayner (talk) 20:58, 24 May 2012 (UTC)
 * Of the 8165 Chinese stubs, 67 have interlanguage links that have been added by other editors, and may be salvagable. Kanguole 22:36, 24 May 2012 (UTC)
 * I made a random sample of 92.6.200.56/McGeolist at User talk:66.127.55.46/MCGeoList sample. I looked at all 20 (i.e. at the content, I haven't actually examined the histories except for a couple) and none had any substantive content.  At least one had an apparent error.  Only Linbrook had obvious signs of other editors contributing in the text, though several had edits by bots adding geo coordinates etc.  One idea is to use a script to examine the article histories and select out the ones that have been edited by humans other than Jaguar ("humans" means ignore bot edits for this).  Mass-deleting the ones without additional human contributions should leave a much smaller set to figure out what to do with.  66.127.55.46 (talk) 01:10, 25 May 2012 (UTC)
 * Agreed. 3876 of the Chinese substubs haven't been touched since they were created.  3251 of them have only been stub-sorted.  I can make lists if an admin is ready to delete them.  Kanguole 09:04, 25 May 2012 (UTC)
 * That looks like a good move to me. bobrayner (talk) 09:26, 25 May 2012 (UTC)
 * On second thoughts, I suggest that the first step should be to delete the 8054 of these Chinese subdistrict stubs that have, native_name, translit_lang1_info1 and no cross-wiki link. Without any of that, we can't even look them up in Chinese databases.  I've listed them at User:Kanguole/ChinaStubs/NoNativeName.  Kanguole 00:45, 26 May 2012 (UTC)
 * Yes, that would be a good start. We should try to avoid getting distracted (I'm as guilty as anyone); there are lots of different problematic geographical stubs out there, some by Jaguar (in various groups) and further out there's a much wider variety by other editors. If we tried to solve all problems simultaneously, we'd fail; a dog which chases two rabbits will catch none. Let's focus on Jaguar's chinese stubs first ... bobrayner (talk) 16:08, 27 May 2012 (UTC)
 * I'm going to start deleting these stubs (listed on User:Kanguole/ChinaStubs/NoNativeName) in approximately 24 hours, since as far as I can determine there's a clear consensus to delete. T. Canens (talk) 16:46, 28 May 2012 (UTC)
 * I've contacted Plastikspork who might be able to fix the Chinese and we cold possibly arrange something later to replace the short stubs with more productive articles. Can you hold off on the deletions for a week to allow us to properly arrange something on this? A lot can happen within a week.♦ Dr. Blofeld  08:26, 29 May 2012 (UTC)
 * Plastikspork does not seem keen. The empty lang-zh is the most glaring omission, but they also lack proper sources (like the one Maculosae tegmine lyncis is using, though apparently not linking) and interwiki links, and the infobox shows several empty slots.  Even with that they would be minimal.  But if someone were using a bot to do that, surely it would be easier to create those slightly larger stubs from scratch than work around the idiosyncracies and errors of the existing stubs.  Kanguole 09:16, 29 May 2012 (UTC)


 * Yes, sorry, but I have to object. All of these articles have a massive potential of being expanded. I have looked at some of them and some of the townships have populations over 30,000! Most of those townships included in that list do not have errors and are correct. Even if some of them did have errors, they could be corrected by a bot or script. The consensus may have been decided to delete the articles, but Dr. Blofeld had pointed out that the articles should not be deleted, and he has given out some good points. I have noticed now that some people have changed their minds about deleting the articles. Jaguar   (talk)  17:14, 28 May 2012 (UTC)
 * I have just created 12 new articles, slowly, with Chinese included. That is what could become of 8,000 articles if they are not deleted. Jaguar   (talk)  17:25, 28 May 2012 (UTC)
 * Hey big cat, tidy up some of the others before adding new ones, please....
 * (edit conflict) - This reminds me of an old supervisor's second favourite joke, about a guy who loses his watch down a dark alleyway, then searches for it out on the main road - why? because he can see better there; it might well be easy to compile a list of pages with, but as I've said below, I've looked at hundreds of these which were absolutely fine and for which it was a mere formality to add the link; would be a huge/impossible task to recreate from scratch; this needs to be far more nuanced; if people are feeling gung-ho, how about first firing a bot that strips this field from some of these stubs, on criteria to be determined; or better still, working through them "one by one" as they say at a Chinese banquet, Maculosae tegmine lyncis (talk) 17:36, 28 May 2012 (UTC)

Heavy Objection &mdash; yes, they're all full of errors, but I don't care about that. They could be easily fixed. Per Blofeld. --Thine Antique Pen (talk • contributions) 17:43, 28 May 2012 (UTC)
 * Saidly, I have to say: go ahead. I have not heard a plan on how within a reasonable timeframe 1 source+ "Chinese=?+interlang can be solved and that is the absolute minimum (as it gives links to the info "in the rest of th eworld"). If Jaguar comes up with a plausible plan to solve it, then I'll be the first to withdraw... L.tak (talk) 17:48, 28 May 2012 (UTC)
 * A clear consensus was formed at AN/I; delete them. Jaguar's promise to create even more of them is hardly an alternative solution. bobrayner (talk) 21:27, 28 May 2012 (UTC)

Wait, there's more
I've just come across a batch of very similar stubs from last July that may have served as Jaguar's model. Kanguole 02:00, 26 May 2012 (UTC)
 * Ouch. It seems none of them carries any information that is not in List of township-level divisions of Hebei. This type of substubs should carry a -100 penalty at List of Wikipedians by article count. —Kusma (t·c) 07:18, 26 May 2012 (UTC)
 * Actually they have less information than the list, which includes the Chinese names of these places (a vital identifier). I just found another batch from August; I don't know how many there are in all.  Comparing say Hanzhuang Township and Quyang Township, the only difference is that Dr Blofeld's includes the district in subdivision_name3.  They were each created by pasting identical text "   (?)  is a ... " text into new articles for each of the subdistricts listed under a district.  The similarity in the wiki source is striking.  Jaguar even copied Dr Blofeld's 18 August 2011.  Kanguole 09:51, 26 May 2012 (UTC)
 * Has anyone informed Blofeld about this development in which his creations are questioned? —Strange Passerby (talk • cont) 15:29, 26 May 2012 (UTC)
 * Kanguole did that right when he posted here. Anyway, those that haven't been expanded yet (Anxiang Township was, by Blofeld) should just be deleted and can be created again when somebody wants to write an article there. As they are not useful to build upon in their current form, there is no point in keeping them (red links are more useful to point out that articles need to be written). —Kusma (t·c) 15:37, 26 May 2012 (UTC)
 * Blofeld doesn't give a damn; you'll get nothing but bluster if he responds to this. He says that he hasn't been mass-creating substubs for more than year, but there's still junk like the ones Kanguole points out, as well as like this—hundreds of Cornish place substubs created last September and "sourced" to Google Maps, of all things (without links to the relevant map displays; and Northward, for one, isn't even labeled on Google Maps, at any zoom level)—and this (hundreds of completely unsourced Welsh SSSI substubs from March 2011). It's impossible to rein in this activity by Blofeld and others, because some folks will always pop up to defend it. Deor (talk) 16:02, 26 May 2012 (UTC)

I've often wondered how many of the regular people commenting at ANI are autistic or suffer from Aspergers. A lot of people here seem to have great difficulty in working with other editors and talking in a friendly fashion to their fellow volunteers with a common goal and accepting them as equals on their intellectual level with similar interests. There appears to be a tendency to revel in slating the efforts of others whenever they possibly can. You actually enjoy the drama and the sadism that comes with saying nasty things about the edits of others. I've never seen such a bunch of socially dysfunctional people in one place before. You find it impossible to interact with others without there being animosity. I would never dream of slating the efforts of others on a public forum on what is a charitable unpaid website and I at least try to see the good in most people and not focus on their more negative aspects, but there is clearly something very wrong on a humane level with ganging up on a forum and only commenting purely negatively on the efforts of others and passing judgements as if they are superior contributors actually in a position to do so. You talk about me as if I actually owe you something or am enemy number one on here and couldn't give a damn about adding anything worthwhile to the website. It is bitterly disappointing that there are few people on English wikipedia beyond Rosiestep, Gerda Arendt and Sarah Stiech who actually treat others like fellow human beings with feelings and fellow contributors sharing a common goal. You people seriously need to take a long hard look at yourselves. ♦ Dr. Blofeld  20:49, 26 May 2012 (UTC)
 * Sheer bluster, as I said above. Everyone who thinks your substubs are valueless is autistic, really? The "common goal" here is to develop encyclopedic articles; you've always claimed that your one-line unsourced placeholders somehow contribute to that, but I don't see it. Does that make me autistic, or perhaps a sufferer from some other of these? As for "talking in a friendly fashion to their fellow volunteers", a review of your interactions with those who disagree with you might be enlightening. Deor (talk) 21:58, 26 May 2012 (UTC)
 * Have you looked in a mirror recently Dr? —Strange Passerby (talk • cont) 01:23, 27 May 2012 (UTC)

Just five minutes ago. Admiring my blue eyes. Anybody who knows me knows I'm very friendly and positive unless provoked with nasty comments. Which is why I detest the way you only focus on people's worst articles and make them feel bad about it. It is by no means a coincidence that ANI attracts people who largely simply do not know how to interact with other people and how to discuss things for the improvement of the project rather than taking swipes at people. Its not the fact that you think my sub stubs are useless, its the way in which you go about it and excessively and unnecessarily attack them which is indicative of individuals with a completely lack of human empathy. This is my last post ever at ANI as I'm far too decent to be commenting here. Until one of you can approach me in a pleasant manner and discuss possibilities for improving the articles, then I'll be glad to discuss things with you.♦ Dr. Blofeld  06:09, 27 May 2012 (UTC)
 * As ever, what Dr Blofeld says now is very difficult to reconcile with Dr Blofeld's previous edits. Never mind; let's stick the content. Do any other editors have a reason to keep these crappy placeholder microstubs? bobrayner (talk) 14:16, 27 May 2012 (UTC)
 * Yes; I've looked at several hundred of these pages by User:Dr. Blofeld over the past 24 hours and, bar half a dozen that have since had a minor administrative adjustment (such as the addition of "east" to their name), these are of course all perfectly good as they stand; I'm interested in Chinese cultural sites, of which there are several hundred thousand, and being able to locate these more precisely is great; I've encountered and/or added links to the ca, eo, no, and zh wikipedias; deleting these "crappy placeholder microstubs" would not only stop en readers heading out but these other readers heading in, and would replace the routine task of adding a link with the insurmountable one of recreating everything from scratch; the less familiar the topic the more value there is in even the scantest information; there's also a distinct lack of respect here for other users and their work, cut it out, Maculosae tegmine lyncis (talk) 23:45, 27 May 2012 (UTC)
 * The bulk discussed here are not, directly, by User:Dr. Blofeld. On the general point it may also be helpful to take a look at this. --92.6.202.54 (talk) 02:08, 28 May 2012 (UTC)(formerly: User:92.6.200.56)


 * User:Symplectopedia has recently created lots of township-level pages on zh wikipedia, based on the information published by the National Bureau of Statistics of the People's Republic of China (here's the link for the 2011 data); this looks like a great source and you can keep on drilling down to the village level; does anyone have anything better? This AfD may be of relevance too, Maculosae tegmine lyncis (talk) 23:45, 27 May 2012 (UTC)
 * I think these pages from the same give provincial-level population figures per the 2010 census; is anyone able to find more detailed figures? Maculosae tegmine lyncis (talk) 10:34, 28 May 2012 (UTC)


 * If there's a proposal to delete and recreate with a bot, what is the end result? Do all the wikilinks and interwiki links get automatically (re)created? And any links from the articles listed at tools - what links here? Maculosae tegmine lyncis (talk) 23:45, 27 May 2012 (UTC)
 * Only 67 of Jaguar's 8164 Chinese township stubs have interwiki links. Those stubs are probably salvagable.  Kanguole 12:45, 28 May 2012 (UTC)
 * If somebody has developed a bot which is able to create proper articles which satisfy wikipedia standards (including WP:N) and write some actual content, I would quite like to see the bot proposal. bobrayner (talk) 13:16, 28 May 2012 (UTC)
 * There's no such proposal afaik. Bots are generally handled by the BRFA people. This may address your query about individual pages that've changed. --92.6.202.54 (talk) 19:41, 28 May 2012 (UTC)(formerly: User:92.6.200.56)

Nobody delete the Spanish stubs, i'm making good progress in sorting them, but was slowed by a faulty modem power supply which should be fixed now.♦ Dr. Blofeld  17:47, 28 May 2012 (UTC)
 * As soon as some other editor says that the articles are fixed - ie. they pass WP:N and are accurate, sourced &c - I would be happy for them to be kept. Unfortunately, going on past edits, I don't think it would be appropriate to take Dr Blofeld's word on this; so I think the community's previous decision to delete still stands. bobrayner (talk) 18:40, 28 May 2012 (UTC)
 * The Spanish stubs I've spent several hours over the last couple of days gradually sourcing and will add infoboxes to afterwards. If you so much as delete a single article which is sourced with data I will kick up a stink so smelly you'd wish you were in a Portuguese fish market instead.♦ Dr. Blofeld  21:25, 28 May 2012 (UTC)
 * Without being overly procedural: the very clear delete consensus was exclusively on the Chinese stubs (as the others had not yet been unearthed....). I think those planning to expand the others (which are overseeable in quantity) should be supported and not be automatically included in the Chinese set... L.tak (talk) 19:18, 28 May 2012 (UTC)

Don't delete. I like to help
Please don't delete all the Chinese subdistrict stubs created by Jaguar. I like to help fix them. But I need time, probably several months, at least. --Pengyanan (talk) 18:00, 28 May 2012 (UTC)
 * Quite a lot of editors agreed to delete them, above. I'm wary of overturning that consensus. There's not much to "fix"; if you want to create proper articles with content and sources, that's great - but the existing articles would be more hindrance than help, so feel free to make nice articles after these existing stubs have been deleted. bobrayner (talk) 18:27, 28 May 2012 (UTC)
 * A lot of editors agreed to delete them because they did not have time or sources to edit them and did not see anyone else like to help. Now I like to help and will do it. The Chinese subdistrict stubs created by Jaguar provide basic information and I can improve them. Saving and not deleting them is to respect editors' serious work. If you guys really decide to delete those stubs, I hope at least those that I and other, if any, editors have improved will not be deleted. Please. --Pengyanan (talk) 04:26, 29 May 2012 (UTC)

A solution could be to incubate them...♦ Dr. Blofeld  18:06, 28 May 2012 (UTC)
 * Yeah, they shouldn't be deleted. It's understandable if the ones with serious errors could be deleted, but there are only around a couple of hundred of them. We can work on a solution. I can help fix the major errors if these articles were kept first. Jaguar   (talk)  19:00, 28 May 2012 (UTC)
 * Jaguar, please make the plan a bit clearer... What would you like to add (interlangs? chinese name? deeplinked source? also inhabitants?) and how do you plan to do it? Would it help to userfy it all to you? L.tak (talk) 19:16, 28 May 2012 (UTC)


 * It is easier to create new stubs without the problems than to fix them, so there is no point in keeping the stubs around. I am willing to press the delete button on a couple of them, but I won't use a bot, and I won't do all, and not today. —Kusma (t·c) 21:14, 28 May 2012 (UTC)
 * Hi, Kusma. I don't know how you come to your conclusion. But to me it is easier to improve existing stubs than to create new ones. I will work on them and will not spend time on arguing here. If you press the delete button, please at least don't delete those that I have improved. Thank you. --Pengyanan (talk) 04:26, 29 May 2012 (UTC)
 * I am not planning to delete any stubs that contain a real reference or information that is not in the parent list article. And I do check history before I delete, so rest assured, I will try not to let your effort go to waste. —Kusma (t·c) 06:55, 29 May 2012 (UTC)
 * Thanks a lot. --Pengyanan (talk) 06:57, 29 May 2012 (UTC)
 * Although I supported the deletion, I did state that my preference is that the articles be fixed somehow. Can the articles be pulled from main space en masse? Articles that have been deleted in the past can be put into user space, but several thousand stubs is pretty hefty. Blackmane (talk) 09:16, 29 May 2012 (UTC)

I would endorse the deletion of List of township-level divisions of Heilongjiang entries though, as they'd require multiple AWB edits to sort. They all have multiple problems, wrong link to township list, stae something is a district when it is actually a prefecture etc. They're a mess, appear to be a lot worse than than the others.♦ Dr. Blofeld  13:41, 29 May 2012 (UTC)
 * Indeed. I just fixed one that you had already touched, it was still a bit of a mess. (Harbin is the prefecture, not the district etc.) Is there an easily accessible list of these? —Kusma (t·c) 18:19, 29 May 2012 (UTC)
 * Er, disregard what I said (non-red links in List of township-level divisions of Heilongjiang). Would you mind me deleting those that still have errors (even if you have touched them)? —Kusma (t·c) 18:24, 29 May 2012 (UTC)

If you want, yes but not ones I've significantly expanded like Abei Township. I only noticed those errors later and indeed a lingering Location in Fujian hidden caption in the infobox..♦ Dr. Blofeld  20:22, 29 May 2012 (UTC)
 * Deleted a couple. —Kusma (t·c) 07:58, 31 May 2012 (UTC)

Oakley77 is creating a lot of unsourced sub stubs. I expanded one here.♦ Dr. Blofeld  22:02, 29 May 2012 (UTC)

Deleted
I deleted everything that was listed on User:Kanguole/ChinaStubs/NoNativeName; I see nothing in the later discussion that is nearly substantial enough to overcome the overwhelming community consensus to delete. These deletions are, of course, without prejudice to recreation of error-free stubs properly supported by sources. If someone fixed the errors in an article but neglected to remove the article from the list, leave a note here and I'll restore it. T. Canens (talk) 17:15, 29 May 2012 (UTC)
 * Thank you. —Kusma (t·c) 18:20, 29 May 2012 (UTC)
 * Please restore Beilin Road Subdistrict, Beilin Road, Dashiqiao Subdistrict, Duling Avenue Subdistrict, Duling Avenue, Yongshun, Beijing, which I have fixed. --Pengyanan (talk) 18:43, 29 May 2012 (UTC)
 * Restored. —Kusma (t·c) 18:48, 29 May 2012 (UTC)
 * Thanks. --Pengyanan (talk) 18:54, 29 May 2012 (UTC)
 * Ok and how do we get them back? Destroying 8000 articles on one's own cognizance seems a little extreme. Rich Farmbrough, 23:40, 29 May 2012 (UTC).


 * If you read the discussions above, you will see that no articles were destroyed, and in fact no information was removed from Wikipedia (all of the information that was in the substubs is still accessible through their Whatlinkshere), while a lot of disinformation has been removed. It would be great if somebody could start writing articles on these Chinese townships. —Kusma (t·c) 01:22, 30 May 2012 (UTC)
 * China is very hard to research. We set a program up at Jag's talk page, and I did a little preliminary work on one province.  Getting China properly covered will take cooperation, and while this has all been very even tempered and collegial, it is clear that people are pulling in different directions. Rich Farmbrough, 01:45, 30 May 2012 (UTC).


 * Please listify the deleted articles. Rich Farmbrough, 23:42, 29 May 2012 (UTC).


 * The list of deleted articles are at User:Kanguole/ChinaStubs/NoNativeName. They may probably be restored with a bot (like Liangent-bot or whatever at Chinese Wiki) that obviously has a few more functions for handling Chinese information. GotR Talk 00:10, 30 May 2012 (UTC)
 * That would take an admin-bot. I have tried to research one article but with the loss of the given information it is extremely hard. http://www.chinadaily.com.cn/life/2009-08/11/content_8554053.htm for example may or may not be relevant. Copying the content of the articles to a list before deleting them would make unwinding this ball of string possible. Rich Farmbrough, 01:35, 30 May 2012 (UTC).


 * There was no content in the "articles", and you seem to be imagining this loss of information. What was there can be trivially obtained again (maybe without errors this time) from lists like List of township-level divisions of Heilongjiang. For example, here is the content of Xingshugang Township: "Xingshugang Township  is a township-level division situated in the Daqing province of Heilongjiang, China." Actually, Daqing is the prefecture, not the province, the county (Honggang District) is missing, as is the Chinese name (杏树岗乡). All of this information is present in List of township-level divisions of Heilongjiang, without misleading stuff like mentions of Fujian province in the infobox and references section. —Kusma (t·c) 01:48, 30 May 2012 (UTC)
 * It would have been easier to fix them up, but no matter I shall get back to my turtles. Rich Farmbrough, 02:17, 30 May 2012 (UTC).


 * Honestly I thought this discussion was done a week or two ago, and was surprised to come back here (after I found Jaguar creating more of the same articles) and see that somehow the overwhelming early consensus had been watered down. It seemed quite clear to me that the issues here are beyond mere stylistic changes that AWB can fix... but rather a fundamental problem of using a dataset that either one doesn't understand fully, or one can't trust for accuracy or completeness, or one can't interpret... etc. It's exactly why mass creations like this are so dangerous, and why we should expect anyone who does them to have a thorough understanding of the data set, but also the information, etc. I don't have any confidence that any of that is present in these articles.


 * I am so baffled by this blind desire to create more pages when, as in this case, their basic accuracy and usefulness is in question. I'm all about creating frames for others to build on, but this is like putting up "for sale" signs and moving on. Shadowjams (talk) 07:20, 30 May 2012 (UTC)

Well I just have no words to say on what everyone has done. All I can say is that you've just wasted hours of my work, all for nothing. I can't take this no more.

Rich Farmbrough was right on keeping those articles, it would have been easier to fix them and it would have been better for them to stay but the trigger happy admins have made the wrong decisions. I can't believe all this unprovoked abuse I had taken shit from people above from this ANI discussion, and I could not do anything because I would have got blocked. I created 12 extra articles so I could prove to people that I could do it correctly, but I have failed.

I want to thank everyone who has supported me; I know who you are :). But if somebody wants to block me now, they can do it. I'm not going to go on this website ever again. Jaguar   (talk)  14:14, 30 May 2012 (UTC)
 * You've done other good work in the past. If you'd like to stick around and work on articles which meet WP:N and so on, that would be very welcome. Your choice. bobrayner (talk) 14:54, 30 May 2012 (UTC)


 * It is very sad that these articles were deleted in haste especially after some editors and I had expressed our willing to fix them. I am very sorry that administrators cannot wait for at least several weeks or even several days. Then hundreds of articles could have been saved. So Sad. --Pengyanan (talk) 08:15, 31 May 2012 (UTC)
 * If you want to write decent articles on these topics, which meet wikipedia standards, you are still free to do that; if you do, it would be very impressive. bobrayner (talk) 08:22, 31 May 2012 (UTC)
 * I personally prefer editing existing articles, not creating new articles. I don't know how many editors are also like me. Those stubs created by Jaguar provided starting points, even though not so decent. Anyone who understands Chinese and randomly encounters these stubs can easily find their Chinese Wikipedian counterpart articles and add to them Chinese characters and inter-language links. Those editors, I assume, are more than those who like to create decent articles in Wikipedia. Unfortunately all of these articles have been deleted. I don't know how long we have to wait before other editors create again those thousands of articles. Again, sad, very sad. --Pengyanan (talk) 09:02, 31 May 2012 (UTC)
 * In that case, there are many thousands of other similarly-deficient geographical stubs for you to work on. These 8000 were the tip of the iceberg. bobrayner (talk) 09:06, 31 May 2012 (UTC)
 * Those thousands of other stubs may not be of my interest. I am sad because these article were deleted just less than one day after I showed my willingness to help. I know it's useless for me to say anything here now. I just want to express my sadness. --Pengyanan (talk) 09:30, 31 May 2012 (UTC)
 * Those that are left still need a lot of work. To pick just the most obvious stuff: 71 of them lack interwiki links; in the infoboxes 70 lack native names, 145 lack tone-marked pinyin and 90 lack the district name; only 11 of them have coordinates; only 17 are sourced to government data; 21 of them are still sourced to Google translate; 12 of them have map captions naming the wrong province.  Kanguole 11:48, 31 May 2012 (UTC)
 * As the pages that were deleted were not articles, not even proper stubs, I do not quite see your problem. The main thing that happened is that Jaguar dropped almost 200 places in List of Wikipedians by article count, but that seems justified to me. Creation of proper stubs about Chinese townships by somebody who has the competence to do so would of course be much appreciated. —Kusma (t·c) 09:08, 31 May 2012 (UTC)
 * My only problem is that those stubs were deleted just less than one day after I showed my willingness to help. I don't want to bother anyone to restore those articles. I just want to express my sadness. I hope that administrators will be more patient in the future to allow other editors to help. OK, I will be quiet and not argue here anymore.--Pengyanan (talk) 09:30, 31 May 2012 (UTC)
 * This conversation's been going on for over a week now. Dedicate your efforts towards the many articles still left. Shadowjams (talk) 06:00, 1 June 2012 (UTC)
 * I do not need your advice (or order, sounds to me, since there is not even a "please" in your last sentence) on how to spend my time. Some editors would rather dedicate their efforts towards arguments here or press the delete button than fix, or wait for other editors to fix, articles there (three days after the list at User:Kanguole/ChinaStubs/NoNativeName was created and before the mass deletions were enforced, 80 articles were fixed and removed from the list). This conversation and the result of it only have chilling effect on me. I will keep my edits on a minimum level. Period. --Pengyanan (talk) 10:31, 1 June 2012 (UTC)
 * They were removed from the list of stubs lacking the most basic information and facing imminent deletion, and some of them were expanded substantially, but most of the remainder at User talk:92.6.200.56/MCGeoList are still greatly lacking, and cannot be said to have been "fixed". Kanguole 10:50, 1 June 2012 (UTC)
 * So what? OK, not fixed. But whatever word you choose, those stubs have been improved to a level that they do not need to be deleted. I make myself clear again that this hasty deletion makes me so sad that I will keep my edits on a minimum level. Bye. --Pengyanan (talk) 11:24, 1 June 2012 (UTC)
 * Stop whinging, drop the bloody stick, and move on. You've said you will desist from commenting here — so do so. —Strange Passerby (talk • cont) 11:32, 1 June 2012 (UTC)
 * No, I did not say that l would desist from commenting here. I said that I would not argue, I said period and bye, but not no comment. I am not arguing (to save any stub). I will keep making myself clear until no one misunderstands me or gives me wrong advice. If you do not, then you can move on. Bye. --Pengyanan (talk) 11:56, 1 June 2012 (UTC)
 * This reminds me teenagers in love on the telephone "you hang up first", "no you hang up", "no I want you to hang up"," no I'm waiting for you to hang up" blah blah blah. There are deletionists on wiipedia, there are inclusionists too, let's all just get along together. He has expressed his sadness, there is no need to keep at him about it. Just let him have the last word.


 * Pengyanan, don't worry about the savage deletionists, they are all naughty, and will face the cosmic defragmentation in the afterlife. They don't know any better, we should feel more pity for them (although I'm not very good at it, I never made a turn red to go with the Turn green and if they keep at it like this I never will :P ) Penyulap  ☏  06:35, 2 Jun 2012 (UTC)
 * During travails on the above I stumbled across this quote from an insider at Baidu Baike, which is apparently a collaborative Web-based encyclopedia with more articles than the en Wikipedia: "there is, in fact, no reason for China to use Wikipedia ... It's very natural for China to make its own products"; that seems to me to entirely miss the point - if Baidu≠China wishes to be a closed world all on its own, fine; the only article I've looked at is this on Lhünzhub County; the Baidu article may me somewhat fuller, but what it lacks is this down the side, which I think is a step in the right direction:

བོད་ཡིག Deutsch Euskara Français 한국어 Italiano Nederlands 日本語 ‪Norsk (bokmål)‬ Português Tiếng Việt 中文, Maculosae tegmine lyncis (talk) 11:11, 1 June 2012 (UTC)