Wikipedia talk:WikiProject Chess/Archive 35
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 30 | ← | Archive 33 | Archive 34 | Archive 35 | Archive 36 | Archive 37 | → | Archive 39 |
Possibility of lowering rating threshold of 2400 for automatic infobox rating updates
Could we possibly lower the threshold of 2400 for automatic infobox rating updates? Archive page explains this threshold with the fact that the full rating list is heavy, currently 7.1 MB. As I don't really know how the bot works, I don't understand either why this is a problem. My main motivation for lowering this threshold is because of female players. Currently only 62 female players' pages are automatically being updated as compared to more than 2000 male players'.
The bot creator, to whom I am very grateful for such an amazing work, says that low ranked players are not of much interest, but they probably forget about female players. I am, e.g. a big fan of Khanim Balajayeva who is currently No. 63 (R. 2397) female player in the world, but her rating isn't updated automatically. So, could we lower the threshold for automatic infobox rating updates, taking into account the aforementioned nuances? -- Guardian of the Divine RabbiT (talk) 05:36, 7 November 2018 (UTC)
- The bot has been out of operation for over three years. The updates are partially automatic. Any changes would require someone to write and maintain a bot. (DrTrigon is seemingly inactive here and on de-wiki.) Hrodvarsson (talk) 01:56, 8 November 2018 (UTC)
- Hi @Hrodvarsson:, thanks for your reply. Are you sure that the bot is out of operation, because all of the chess players' pages that I follow are being constantly updated? What do you mean by "partially automatic"? -- Guardian of the Divine RabbiT (talk) 08:57, 8 November 2018 (UTC)
- The ratings are updated on Template:Elo rating. If you look at the history of that template you see the last update by DrTrigonBot was in June 2015. By partially automatic I mean that an editor on de-wiki updates the ratings each month, and his updates are replicated here manually. Hrodvarsson (talk) 23:49, 8 November 2018 (UTC)
- I see now, @Hrodvarsson:, thanks. As as compromise, we can lower ratings for female players to 2200 (there's a 'Sex' attribute in the data source). But I can't find the code that shrinks the data source to just only 2 attributes. Is it also done manually and then uploaded to Template:Elo rating? -- Guardian of the Divine RabbiT (talk) 01:38, 9 November 2018 (UTC)
- I am unaware of the script the de-wiki editor uses to update the ratings each month. Template:Elo rating is updated simply by copy and pasting the updates from de-wiki, and the rating updates here are currently entirely dependent on the de-wiki editor's updates. The ratings here went without update from June 2015 to April 2016, and from April 2016 to August 2016. Since then Chessrat and recently myself have regularly updated them from de-wiki. To return to the original question, any change would require someone here to write and maintain a bot, or for the de-wiki editor to lower the rating limit himself. Hrodvarsson (talk) 02:40, 11 November 2018 (UTC)
Featured quality source review RFC
Editors in this WikiProject may be interested in the featured quality source review RFC that has been ongoing. It would change the featured article candidate process (FAC) so that source reviews would need to occur prior to any other reviews for FAC. Your comments are appreciated. --IznoRepeat (talk) 21:47, 11 November 2018 (UTC)
We need a seperated article about multiplayer chess to include chess variants for more than four players. --Sharouser (talk) 16:09, 20 November 2018 (UTC)
Separate template for chess openings?
The template {{Chess}} seems a bit over-crowded. Do people support a separate {{Chess openings}} template that includes all the variations? (I've also asked on Template talk:Chess but that talk page probably isn't watched) power~enwiki (π, ν) 18:44, 10 December 2018 (UTC)
Wang Yue unbeaten streak in 2008
Could use a helping researching mind to resolve this apparent discrepcancy. CaradhrasAiguo (leave language) 06:56, 18 December 2018 (UTC)
Country in infobox
This came up in connection in Fabiano Caruana, and I posed the question in the talk page, but got only one answer, so I thought the prudent thing would be to bring the question here.
Does the "country" field in the infobox for chess player refer to the country that the guy is a citizen of, or the federation he is registered to (i.e. the country under whose flag he plays in FIDE events)?
For example, Caruana is a dual citizen of the U.S. and Italy, and has been since birth. His registered federation is currently the U.S.; it was Italy for a few years. Bruce leverett (talk) 05:26, 8 December 2018 (UTC)
- In Caruana's case I'd definitely say USA. Yes he has Italian citizenship but he was born in USA and his connection to Italy is through his parents. Nationality is not as black and white as it used to be so just take these on a case by case basis. MaxBrowne2 (talk) 05:56, 8 December 2018 (UTC)
- In the case of a sportsperson, in my opinion, what is most interesting for readers is sports nationality. In this case, the country they represent in international events. So I think what was presented in Caruana's article until a few weeks ago was correct: his sports nationality, indicating the years he held each. This was changed quite abruptly by a user, but, as I say, I think it should be taken back to the previous situation.--Gorpik (talk) 11:54, 19 December 2018 (UTC)
- My understanding has always been that it is for the country they represent as a chess player. In this case it should be the US.-- Pawnkingthree (talk) 14:31, 19 December 2018 (UTC)
- I see Gorpik's point: Caruana's infobox used to mention that his country was United States, but also used to note that he represented Italy from 2005 to 2015. Now it doesn't mention Italy.
- I checked Kateryna Lagno and Sergei Karjakin, and both of their infoboxes mention their current country and their former country, so I guess that Caruana's should do so also. I will fix it shortly. Bruce leverett (talk) 19:37, 19 December 2018 (UTC)
Indian, or other, origin of chess
I assume there has been some discussion, here and in other talk pages, about how to handle the question of where Chess originated. If there are some well-known rules we should be following, or consensus or something close to it, I'd like to know.
Right now, the hot topic (based on a recent edit to Chess by User:MaxBrowne2), is David Li's theory of the Chinese origin of chess. While this has grave flaws, perhaps we can't ignore it. And there are other theories that we may not be able to ignore.
I would assume that the discussion in Chess should have less detail than the discussion in History of Chess. Would this mean, for example, that doubtful theories such as those involving China, Afghanistan, etc., should be discussed in the latter, but not in the former?
Analogously, in any one article, one would expect that the lead section would have less detail than the inner sections; for example, in Chess, could one discuss alternate theories in the History section, but not in the lead section?
Also, should we refrain from singling out one particular alternate theory, but just mention that "there are alternative theories", citing reference to the relevant sources? Bruce leverett (talk) 02:02, 26 January 2019 (UTC)
- I think it would be better to remove this sentence from the chess article: "Chaturanga is also the likely ancestor of the Eastern strategy games xiangqi, janggi, and shogi; a minority opinion holds that chess originated in China." It isn't necessary to the article lead of chess and would be a better fit for the lead of chaturanga. The minority theories aren't important enough for the chess lead and I think deserve only brief mention in the body of the article. Quale (talk) 02:47, 26 January 2019 (UTC)
- No strong feelings here. If you think it should be removed from the lead, go ahead and remove it. MaxBrowne2 (talk) 07:03, 26 January 2019 (UTC)
Scholar's mate content dispute
Can I get a third opinion on this dispute at the Scholar's mate article about the value of this video to the article? Thanks. Banedon (talk) 23:31, 18 February 2019 (UTC)
Categorisation
Category:Chess grandmasters (1,141 entries) strikes me as a vast unwieldy unnavigable unsearchable mess. IMO it (and some of the other chessplayer categories) should be split by nationality (or, multiple nationalities where relevant). Compare Category:Contract bridge players by nationality for a model of how to arrange such big categories tidily.
This is something I could do, but I have other WP:WikiGnomeish projects which take up most of my time. Therefore, I post the idea here and walk away whistling cheerfully, to relax with another of agadmator's videos... Narky Blert (talk) 22:58, 28 February 2019 (UTC)
- Should the category be broken down by decade or year? Unlike other categories ("actresses"), it's very easy to assign a date to when somebody first achieved the title of grandmaster. power~enwiki (π, ν) 23:06, 28 February 2019 (UTC)
- How is the category hard to navigate and how is it messy? It is essentially a large alphabetized list, which seems perfectly neat and easy to navigate to me. List of chess grandmasters gives you another way to find articles in a table with sortable column headers. Quale (talk) 01:44, 1 March 2019 (UTC)
- Category:Chess players by nationality already exists, so readers who are interested to navigate players based on nationality can use those categories. You could create "[Nationality] chess grandmasters" sub-categories where relevant and depopulate the existing categories, but that seems like a lot of work for a negligible benefit. Sub-categories for the year/decade the title was awarded would be somewhat useful. Hrodvarsson (talk) 04:12, 1 March 2019 (UTC)
- Subcategories for the year/decade the title was awarded would be horrible. It's much easier to get that information from List of chess grandmasters. What use would you make of it anyway? Quale (talk) 05:28, 1 March 2019 (UTC)
- Sub-categories for each year would be overkill but some "Nationality GMs" templates are ordered by decade, which aids readability for countries with large numbers of GMs. If the problem is the GM category's large size, this would be the best way of separating it in my opinion as dividing GMs by nationality is largely redundant as we already have chess players by nationality. But again I am not sure how much benefit any of this sub-categorization would have and I will not be implementing it myself. Hrodvarsson (talk) 04:12, 4 March 2019 (UTC)
- First, I completely disagree that this aids in readability in templates. If you don't know the birth decade of the player you have to search the entire template anyway, so sorting by last name makes players easier to find. But beyond that, the "Nationality GMs" templates are stupid and should just go away—I really hate those kinds of navigation boxes. Russia has over 200 living GMs and quite a few more who are deceased. Admittedly Wikipedia doesn't have 250 articles on Russian GMs (yet anyway), but at some point it could and even now for several countries we have more articles on chess players than can usefully be included in a navigation box. The selective chess templates such as {{American Chess Masters}} also suck. (That template includes the immortal Constant Ferdinand Burille but omits that utterly forgettable fish Isaac Kashdan, and that probably isn't even its greatest sin. For example, unless I overlooked someone, exactly zero women are included. In 2019, that's unpardonable.) Lists of GMs by country is exactly the kind of information that belongs in sortable tables in articles such as Chess in China and Chess in Russia. Quale (talk) 06:00, 4 March 2019 (UTC)
- Nominate the templates for deletion in that case. American Chess Masters was discussed here last year. I don't think anyone is happy with it (even the capitalization is flawed), but no one cares enough to update, reorganize or otherwise do something about it. Hrodvarsson (talk) 03:51, 6 March 2019 (UTC)
- First, I completely disagree that this aids in readability in templates. If you don't know the birth decade of the player you have to search the entire template anyway, so sorting by last name makes players easier to find. But beyond that, the "Nationality GMs" templates are stupid and should just go away—I really hate those kinds of navigation boxes. Russia has over 200 living GMs and quite a few more who are deceased. Admittedly Wikipedia doesn't have 250 articles on Russian GMs (yet anyway), but at some point it could and even now for several countries we have more articles on chess players than can usefully be included in a navigation box. The selective chess templates such as {{American Chess Masters}} also suck. (That template includes the immortal Constant Ferdinand Burille but omits that utterly forgettable fish Isaac Kashdan, and that probably isn't even its greatest sin. For example, unless I overlooked someone, exactly zero women are included. In 2019, that's unpardonable.) Lists of GMs by country is exactly the kind of information that belongs in sortable tables in articles such as Chess in China and Chess in Russia. Quale (talk) 06:00, 4 March 2019 (UTC)
- Sub-categories for each year would be overkill but some "Nationality GMs" templates are ordered by decade, which aids readability for countries with large numbers of GMs. If the problem is the GM category's large size, this would be the best way of separating it in my opinion as dividing GMs by nationality is largely redundant as we already have chess players by nationality. But again I am not sure how much benefit any of this sub-categorization would have and I will not be implementing it myself. Hrodvarsson (talk) 04:12, 4 March 2019 (UTC)
- Subcategories for the year/decade the title was awarded would be horrible. It's much easier to get that information from List of chess grandmasters. What use would you make of it anyway? Quale (talk) 05:28, 1 March 2019 (UTC)
Proposed deletion of 365Chess.com page.
Hello everybody, I have recently written the 365Chess.com page (translated fron the Italian wikipedia page, which I made with data provided by the site's owner and manager. Soon after it was tagged for proposed deletion, with main reason that it lacked references and also that it is "not important". Today I added a reference (though I think it will be insufficient), but I don't agree on the "not important" reason. The site is well known and quite good in my opinion. It has a lot of well done and interesting features and it contains the largest database online (the site says 4 million games and they can be something less, but no other online database has even 3 million. Games can be linked separately, something that most other chess websites (except chessgames.com) won't do. In short, it would be wrong to delete this page. Can someone suggest how to avoid its deletion? Thank you, --Gab.pr (talk) 00:26, 2 April 2019 (UTC)
- I see that you have removed the Proposed Deletion tag, but I suspect that the article is doomed anyway. I am not sure that I can help you with this. I will try to give you some general advice, but I do not claim to be an expert.
- You can start with this link: WP:NOTPROMOTION. There are ways to promote your favorite websites, but Wikipedia is not one of them. You might, for example, put a link to 365Chess.com on your Facebook page, if you are on Facebook. You might send email to everyone in your address list. These are reasonable ways of letting people know about something that you like. But Wikipedia is not a good way of doing so.
- Also, you should look at this link: WP:NOTE. It says, among other things, "... if no reliable third-party sources can be found on a topic, then it should not have a separate article." None of the information in your article is from a third-party source; it is all from the website itself, or from the owner. That by itself was one of the reasons for the original Proposed Deletion tag. Proposed Deletion tags are supposed to be for "uncontroversial" deletions. The complete lack of third-party sources is enough by itself to make this uncontroversial.
- You might consider searching through the established chess websites, such as chessbase.com or chess.com, or even through Google, to find information about 365Chess.com that is not directly from the site itself or from the owner. If it is an important site, you should have little trouble finding such information. Of course, you can also search through the print media, such as chess magazines or books. Good luck. Bruce leverett (talk) 03:21, 2 April 2019 (UTC)
- Thank you for your interest. I removed the tag on suggestion of user EggOfReason, though he said that it could be deleted anyway. I quote his message: You may prevent the proposed deletion by removing the proposed deletion notice, but please explain why in your edit summary or on the article's talk page. Please consider improving the page to address the issues raised. Removing the tag stops the deletion process for a while, but other deletion processes exist. In particular, the speedy deletion process can result in deletion without discussion, and articles for deletion allows discussion to reach consensus for deletion.
- Exactly the same problem was encountered by the page Chessgames.com, though later the page was kept (I don't remember the relevant reasons). It may seem strange, but I could not find any references on the web other than blogs, which I know are not accepted as reliable sources. Nonetheless, I will try again. I have checked many sites with games databases, and my opinion is that Chessgames.com is the best for various reasons, though it contains much fewer games than 365Chess.com. Of all the others, I think that 365Chess is the next better one. It has plenty of useful and interesting features, as reported in the article. Some of them are not offered by Chessgames.com, such as the analysis of board positions with the site's engine (Stockfish). Also the search of tournament results is very good. Please note that I know something about chess, I have written over 900 pages on chess (and about 600 on other topics) on the Italian wikipedia, where I have the same username as on this wikipedia. As I already said, I will try to find reliable external sources on the web and elsewhere, but I hope that the "speedy deletion process" will not cancel the page before I can find and publish them. Greetings, --Gab.pr (talk) 16:56, 2 April 2019 (UTC)
- A strict interpretation of WP:GNG would probably result in the deletion of Chessgames.com too. Specialist websites in general are rarely referred to in Reliable Sources. MaxBrowne2 (talk) 23:31, 2 April 2019 (UTC)
A new newsletter directory is out!
A new Newsletter directory has been created to replace the old, out-of-date one. If your WikiProject and its taskforces have newsletters (even inactive ones), or if you know of a missing newsletter (including from sister projects like WikiSpecies), please include it in the directory! The template can be a bit tricky, so if you need help, just post the newsletter on the template's talk page and someone will add it for you.
- – Sent on behalf of Headbomb. 03:11, 11 April 2019 (UTC)
interactive game viewer (again)
hi all.
so some of you may remember previous attempts to enable "interactive chess viewer" on enwiki.
some previous discussions can be found on Wikipedia:WikiProject Chess/Interactive chess boards and its talk page (i think this contains links to even older discussions).
some updates:
- it seems that User:Fred Gandt abandoned (or at least, "postponed indefinitely") his proposal
- i did make some changes in my viewer, that answer some, though not all, of User:Krinkle's critique (summarized below)
- some more projects started using the viewer. specifically, Hebrew wikibooks, Russian wikipedia, and Russian wikinews
- viewer changes summary (* marks items criticized by Krinkle)
- * dependencies: reduced the dependencies to just "tabs"
- * script no longer loads images, and instead, images are loaded via CSS (using background-image). pretty much same list of images any use of {{Chess diagram}} loads, plus some images for the buttons.
- animation: animation is now done by css "transition" instead of jquery .animate() from the script (except "smooth scrolling", which still uses jquery animation).
- * display inconsistency: gone. until script finishes loading, there's a space on the page waiting for it
- viewer is now mobile friendly. some cooperation from "interface editors" (people with permission to edit in mediawiki namespace) is needed, so for now, only hewiki _actually_ enables the viewer on mobile, and it's possible that the mobile view is loees polished
i suggestion you look again at the viewer, and consider again using it on enwiki. here is a list of pages with the viewer active:
- hewiki: he:מיכאל בוטביניק (interwiki of Mikhail Botvinnik). can be viewed on mobile too.
- ruwiki: ru:Штейн, Леонид Захарович (interwiki of Leonid Stein. viewer at bottom of page)
- hebooks: he:b:שחמט/רישום מהלכים (this is a "guide" page. the viewer is used in some books which are still in draft stage)
- runewes: ru:n:Template: Pgnviewer (the template page, with demo. afaik, runews did not yet use it in actual article)
in light of the stagnation in Gandt's project, the improvement in my script, which, i believe, amends the bulk if the issues in Krinkle's critique, and the fact that more projects are starting to use it, i'd like to propose (again) that enwiki will consider adopting it too.
peace - קיפודנחש (aka kipod) (talk) 20:45, 18 April 2019 (UTC)
- The way the knights move sorta makes me seasick. How about changing to go as the crow flies? (Picking an particular "L"-pattern is subjective, & in a way misleading, too - it is how to learn the knight's move; not how to think or visualize it.) I'd also suggest to speed up the movements. (Clicking thru a game is slow-mo otherwise compared to other playthrus, e.g. Chessgames.com, 365chess.com etc.) --IHTS (talk) 04:11, 30 April 2019 (UTC)
- the knight's move is just a little whimsical thing i did on hewiki - on ruwiki, runews, hebooks and ukwiki the knight moves as the crow flies, and i might straighten it on hewiki too. this is 100% incidental, and can be "fixed" by removing the 2 lines which implement this whim from the CSS. by using templatestyles, enwiki can control all the aspects of the display, up to and including changing the images used for the pieces, or even using the unicode "chess piece" characters instead of images (i think this is what Fred Gandt's demo uses).
- the tempo is set by the "transition" attribute in CSS (same one that implements the knight's perverted dance), and can be tuned from "teleportation" (zero animation, piece jumps from current square to destination) to "as slow as you want" - again, by utilizing templatestyles, enwiki can have full control, and be free of my, or anyone else's whims. (it's also possible to get the CSS from hewiki, in parallel to the script itself, for projects that choose not to be bothered with maintaining templatestyles). you may want to look at ruwiki - though they run the script from hewiki, both animation and the "algebraic notation" tab behave differently, under the control of the template.
- the delay between moves can be set by the template, or left to the reader (when the template does not set delay, the script displays "slower" and "faster" buttons).
- but all this is secondary - before discussing the knight's movement or the game tempo, enwiki has to decide whether it wants to utilize this. if it does, then we can start discussing tweaks. peace - קיפודנחש (aka kipod) (talk) 18:25, 30 April 2019 (UTC)
Need help with diagram
The first diagram in Queen and pawn versus queen endgame#Queen and two pawns versus a queen is not displaying correctly. It should be White: Kh4, Qe4, Pg5, Ph6; Black: Kh8, Qd7. I can't get it right. Is it because of the new align to put two diagrams together? Bubba73 You talkin' to me? 23:35, 19 May 2019 (UTC)
- Fixed.[1] There must be two unnamed parameters before the diagram. PrimeHunter (talk) 00:03, 20 May 2019 (UTC)
- Thanks, it has been years since I've done diagrams, and I used to use a program I wrote to generate new ones. Bubba73 You talkin' to me? 00:16, 20 May 2019 (UTC)
- However, the captions were lost. Is the | on the right no longer needed? Is it optional? Bubba73 You talkin' to me? 00:19, 20 May 2019 (UTC)
- Another editor made a change that restored the captions, but it messed up the positions of some pieces again. I reverted it for now, until someone can fix it. Bubba73 You talkin' to me? 05:08, 20 May 2019 (UTC)
- I took care of it. --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:24, 20 May 2019 (UTC)
- The Lputian vs. Haroutjunian position is still incorrect. See the talk page. Bubba73 You talkin' to me? 05:29, 20 May 2019 (UTC)
- Yeah; I see what's going on now. Give me a few minutes to figure it out. <grab the really big hammer> --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:36, 20 May 2019 (UTC)
- All done. <cross fingers> --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:46, 20 May 2019 (UTC)
- It still wasn't right. My first fix inserted an unnamed parameter before the diagrams but forgot to compensate by removing an unnamed parameter after the diagram in order to position <footer> in the right unnamed parameter counted from the start. Now fixed with [2] (diff includes text edits I didn't make). PrimeHunter (talk) 10:05, 20 May 2019 (UTC)
- I had to revert that because it put erroneous piece positions back into both charts. (I'm not sure what this "<footer>" thing you're referring to is, or why it matters, as nothing appears amiss to the naked eye in mainspace at this point. --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 12:00, 20 May 2019 (UTC)
- <footer> is the term for an unnamed parameter in the documentation of Template:Chess diagram small. I set the positions which were requested by Bubba73 and were attempted in the source at the time. I don't know whether those are actually the right positions so I will not make further changes. PrimeHunter (talk) 20:23, 20 May 2019 (UTC)
- I plan to expand and clarify that section within a few days. Bubba73 You talkin' to me? 00:40, 21 May 2019 (UTC)
- IMO, it is better and easier to use the
fen=
parameter of the template, than to count dozens and dozens of pipes (i.e.,|
signs). compare{{Chess diagram small |tright |Lomonosov Tablebases | | | | | | | | | | | | | |pd|kd|qd | | | | | | | | |pl| | | | | | | | |kl| | | | |ql| | | | | | | | | |pl| | | | | | | | | | | | | | | |<center>White mates in 297.</center> }}
- IMO, it is better and easier to use the
- I plan to expand and clarify that section within a few days. Bubba73 You talkin' to me? 00:40, 21 May 2019 (UTC)
- <footer> is the term for an unnamed parameter in the documentation of Template:Chess diagram small. I set the positions which were requested by Bubba73 and were attempted in the source at the time. I don't know whether those are actually the right positions so I will not make further changes. PrimeHunter (talk) 20:23, 20 May 2019 (UTC)
- I had to revert that because it put erroneous piece positions back into both charts. (I'm not sure what this "<footer>" thing you're referring to is, or why it matters, as nothing appears amiss to the naked eye in mainspace at this point. --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 12:00, 20 May 2019 (UTC)
- It still wasn't right. My first fix inserted an unnamed parameter before the diagrams but forgot to compensate by removing an unnamed parameter after the diagram in order to position <footer> in the right unnamed parameter counted from the start. Now fixed with [2] (diff includes text edits I didn't make). PrimeHunter (talk) 10:05, 20 May 2019 (UTC)
- All done. <cross fingers> --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:46, 20 May 2019 (UTC)
- Yeah; I see what's going on now. Give me a few minutes to figure it out. <grab the really big hammer> --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:36, 20 May 2019 (UTC)
- The Lputian vs. Haroutjunian position is still incorrect. See the talk page. Bubba73 You talkin' to me? 05:29, 20 May 2019 (UTC)
- I took care of it. --2601:444:380:3A90:ED8D:CA52:2C40:4B5E (talk) 05:24, 20 May 2019 (UTC)
- Another editor made a change that restored the captions, but it messed up the positions of some pieces again. I reverted it for now, until someone can fix it. Bubba73 You talkin' to me? 05:08, 20 May 2019 (UTC)
a | b | c | d | e | f | g | h | ||
8 | ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() | 8 | |||||||
7 | 7 | ||||||||
6 | 6 | ||||||||
5 | 5 | ||||||||
4 | 4 | ||||||||
3 | 3 | ||||||||
2 | 2 | ||||||||
1 | 1 | ||||||||
a | b | c | d | e | f | g | h |
- with
{{Chess diagram | fen = 8/5pkq/8/P7/1K4Q1/8/P7/8 | align = tright | header = Lomonosov Tablebases | footer = <center>White mates in 297.</center> | size = 22 }}
- with
a | b | c | d | e | f | g | h | ||
8 | ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() | 8 | |||||||
7 | 7 | ||||||||
6 | 6 | ||||||||
5 | 5 | ||||||||
4 | 4 | ||||||||
3 | 3 | ||||||||
2 | 2 | ||||||||
1 | 1 | ||||||||
a | b | c | d | e | f | g | h |
- the first syntax is very brittle - every small mistake in counting pipes screws it up, as this section demonstrates. the 2nd syntax uses named params, it's clear what each does, it's not sensitive to order (i.e., moving the "header" param to be first creates no problem), no confusion.
- as a side, i don't like the "center" tags of the footer. if the footer should be centered, this should be done by the template itself, and affect all footers for all diagrams. otherwise, let the footer be where all other footers are. peace - קיפודנחש (aka kipod) (talk) 20:51, 21 May 2019 (UTC)
Opinions sought
See Talk:Glossary_of_chess - should "confirmation bias" be included in a glossary of chess terms? I could try a RFC but I prefer to keep it within wikiproject chess. I fear if I continue the discussion with User:Hollarbohem and User:Quale I will simply end up looking for arguments to back up my preconceived opinion. MaxBrowne2 (talk) 00:47, 24 May 2019 (UTC)
- I don't think so. It is not unique to chess, chess-like games, or even games in general. Bubba73 You talkin' to me? 00:52, 24 May 2019 (UTC)
- I agree. But I will take this to the Glossary talk page. Bruce leverett (talk) 02:27, 24 May 2019 (UTC)
anon editor removing referenced material and added un-referenced material
At King and pawn versus king endgame an anonomous editor removing referenced material and added un-referenced material. Bubba73 You talkin' to me? 23:20, 25 May 2019 (UTC)
Improving the table of GMs
I have just added FIDE IDs to List of chess grandmasters. I'd like to collaborate to make further improvements, see Talk:List of chess grandmasters#Improving the table of GMs for what I have in mind. The really interesting work is writing some simple programs to help automate discovery of new GMs from FIDE data to add to the table. (I wanted FIDE ID in the table to make it easier to match our table entries to FIDE records.) The article also needs help to verify and cite the table entries. Quale (talk) 08:11, 14 May 2019 (UTC)
- I don't think any distinction should be made between "honorary grandmasters" and "full grandmasters". The majority of the honorary gm's genuinely were of grandmaster strength at their peak but were overlooked by the system as it was at the time. MaxBrowne2 (talk) 22:40, 14 May 2019 (UTC)
- I thought FIDE got rid of the honorary system altogether and now regards them all as full GMs anyway.Pawnkingthree (talk) 22:52, 14 May 2019 (UTC)
- Perhaps there could be a note column where miscellaneous details such as "retrospective award" could be recorded. (I prefer "retrospective" rather than "honorary" which implies that they didn't truly earn the award). Also I don't agree with a "table of shame" for revoked GM titles, that could be included in the note column (or just omit them from the article). MaxBrowne2 (talk) 01:31, 15 May 2019 (UTC)
- Another note - I'm not sure that a "place of birth" column would be all that useful. Children of military and diplomatic personnel are often born in countries other than their actual nationality, one example I recently edited is Audra McDonald, and there are probably chess players born in similar circumstances. Nakamura also comes to mind - he was born in Japan but moved to the US when he was 2. He is ethnically Japanese but his nationality is US and Japan had no part in his chess development. MaxBrowne2 (talk) 01:42, 15 May 2019 (UTC)
- Interesting points. I had hoped to keep most of the discussion on Talk:List of chess grandmasters nearer the article (my mention here was supposed to be only a pointer to discuss on the article talk page), but since we're here,
- Re: honorary GMs – yes, as far as I know FIDE isn't awarding any more HGM titles. I don't know if FIDE considers the older honorary awards to still be honorary or if they are now considered exactly the same as the standard GM award. In some ways an honorary GM is a more exclusive award since there are so many fewer of them than regular GMs. If we have a Notes column we can put a mention there if we like.
- About the revoked titles – The fact that there are any revoked titles at all is a point of a fair amount of interest. Putting fraudulent GMs in the main table with a note as we did before was bad. We could state in the list article that two revoked titles are not included. I see the title revocations are not described in grandmaster (chess), something we should fix regardless of whether the revoked GMs have a section in the list article.
- You make good points about including birthplace, and I think I won't make any efforts in that direction right now. I agree with your observation that birthplace isn't always indicative of where a chess player developed her skill, although I think the exceptions are a small fraction of the whole and we do have a federation column to help clarify that. (This would be especially true if we consistently recorded the player's federation at the time of the title award rather than claiming (and failing) to use the player's current federation.) Birthplace would also make the table significantly wider (possibly bad on devices with smaller screens) and the article larger and slower load (bad for everyone, especially on slow networks). The final negative is that it would increase the maintenance burden. On the plus side place of birth is an interesting biographical detail, and along with DOB, DOD and place of death is considered basic biographical information in encyclopedias. For many GMs we will have a bio article that should provide this, but some number of GMs probably won't have articles. On a strictly numerical count without weighting the relative importance of the pluses and minuses, there are more minuses. Also adding the column would be a lot of work, and it can be done later if we decide we want it. Quale (talk) 03:33, 15 May 2019 (UTC)
- I thought FIDE got rid of the honorary system altogether and now regards them all as full GMs anyway.Pawnkingthree (talk) 22:52, 14 May 2019 (UTC)
- If you want you can copy this material to Talk:List of chess grandmasters. MaxBrowne2 (talk) 04:41, 15 May 2019 (UTC)
- I found 9 incorrect birthdays by looking at GMs with birth years that did not match the birth year in the FIDE rating lists. (Recent FIDE rating lists include only the birth year so the automated checking we can do is limited.) This suggests that there are probably quite a few more bad birthdays in the list, either due to typos we made or typos or mistakes in the online sources that we may have used such as chessgames.com. Quale (talk) 07:28, 16 May 2019 (UTC)
- For "older" GMs, the complete date of birth can still be found at OlimpBase, which has records of all rating lists from 1971 to 2001. Sophia91 (talk) 21:33, 16 May 2019 (UTC)
- Another online resource for old ratings is benoni.de, which includes the period from 2001, e.g. http://www.benoni.de/schach/elo/elohis.html?id=1700014&lang=en. It should be treated as unofficial, but it can be used to check the rating history of individual players and the data there can be confirmed at https://ratings.fide.com/download.phtml. MaxBrowne2 (talk) 01:04, 17 May 2019 (UTC)
- The FIDE rating lists are hugely helpful but frustratingly hard to use. As a site for the old FIDE rating lists I like http://www.mark-weeks.com/chess/ratings/. Olimpbase is good also, and does a much better job of documenting the origin of each file. Of course from Jan 2001 on official lists can be downloaded from FIDE at https://ratings.fide.com/download.phtml, but I imagine everyone here knows that. (Scripting using wget or cURL is recommended over downloading all the old FIDE lists by hand unless you have a lot more patience for clicky-clicky-clicky than I do.) Lists since August 2012 are available in XML which I find a little more convenient to load than fixed column text, but YMMV depending on your tools. Of course the file format is no guarantee of data quality, and neither is the year. I understand 40 year old data OCRed from magazines might be a little imperfect, but
standard_jan18frl_xml.xml
has a few entries where the players have no names. This is quite a trick. For an official file distributed in 2018 I expect a higher standard.- No FIDE ID before July 1998 so it's difficult to positively identify some players across lists through the years.
- Once FIDE introduced the ID they changed the numbering at least once or twice or more, so some players have multiple ids. Worse, some of the original ids were reassigned to other players. For a time FIDE also reassigned some IDs when players changed federations. I hope they don't still do that, it's utterly moronic. Of course FIDE used to reassign IDs for some married women who changed their names, and that's even more stupid.
- Player names spellings are all over the map. Some players appear twice on the same rating list, often women under married and unmarried names. Even better, they have different ratings for the two entries for the same player.
- Some lists drop inactive players, then some months later they may reappear on a subsequent list without warning. For a time FIDE dropped hundreds of players from federations that were behind in their dues or some damn thing. They just suddenly vanish from the list. The 1974 list is missing all or almost all women. I think probably FIDE did this, but it's possible that Chess Informant performed edits before publication.
- The fields on the early lists change frequently. The only fields on every list are name, rating and federation. Titles are missing from 23 years of lists from 1975 until July 1998, absolutely tremendous.
- Birth date or even just year doesn't appear at all on any list for the 25 years from 1975 until July 1999. We do get full DOB with month and day in every list in 1971–1974 and July 1999 to July 2005 except for Jan 2000 and Jan 2001 that have no birth info. From October 2005 on only birth year is given.
- The FIDE Chess Profile pages can be used to find some info not available in the rating lists such as the year the title was awarded. The data quality here was rather sketchy although it may have improved. In 2011 I noticed that the FIDE Chess Profile for this little known fish https://ratings.fide.com/card.phtml?event=5000017 had the wrong year for his GM title award. (He was World Champion at the time.) When I emailed the FIDE webmaster they corrected the error and sent me a pleasant reply asking me if there were any other errors in their data. I found that not at all reassuring since I have no idea what other errors they have made in their record keeping, I had hoped to rely on FIDE as a WP:RS. On the other hand, the Chess Profile provides a convenient link from which you can find a PDF of the title applications for some (but by no means all) players. This can provide info that is hard to find in other reliable sources. I think Sophia91 has used the title application documents. Quale (talk) 05:53, 17 May 2019 (UTC)
- I was thinking maybe include a column for year of award for International Master title, that way we could identify the exceptional cases like Tal and Christiansen who skipped over the IM title and were awarded the GM title directly. But maybe better to walk before we run. MaxBrowne2 (talk) 07:00, 17 May 2019 (UTC)
- I don't find that a very compelling reason to make an already wide table wider and add the maintenance burden of an extra column that is rather uninteresting for 1800 rows just to demonstrate some trivia for a tiny fraction of the players. But there are actually quite a few GMs who were never IMs, at least 27 are listed at grandmaster. I find it curious that you think that the IM title year would be of greater interest than the birthplace, but chacun à son goût. Quale (talk) 22:35, 17 May 2019 (UTC)
- this thing looks like something that should be done at wikidata level, no? so the extraction tool/script, whatever it is, written in python or javascript or whatever, should update wikidata, and then the table, as well as the info template in each GM article, could utilize standard tools to extract whichever field it wants from WD. it seems kinda waste to go through the exercise of extracting all these goodies from FIDE, and then only one page on one wiki can enjoy the results, no? maybe i'm missing something, but if i don't, please consider enriching WD and puling the data from there, instead of populating a single article on enwiki. peace - קיפודנחש (aka kipod) (talk) 21:40, 17 May 2019 (UTC)
- Yes, I think relying on wikidata is an excellent idea. I expect at some point that's what we will be doing, whether it is next week or next year. I looked at Wikidata briefly but it probably needs someone who understands it better to proceed. My main hurdle is lack of familiarity with wikidata. I have an idea of what I can do in wikipedia, but I don't have a good notion of what would be required to use wikidata. I don't know if it would be organized as a single item in wikidata (chess grandmasters) or distributed across items for each of the individual GMs. I also don't know how to set up the mechanical updates in wikidata. Other challenges include the data in the table that isn't easy or feasible to extract mechanically (convenient FIDE data give only birth year, date of death isn't in FIDE records), and it's unclear how to indicate sourcing (although we do a poor job of that now). But if someone else could spearhead such an effort I would support it wholeheartedly. I know how some of the information can be extracted from the FIDE rating lists, and a little more can be gotten by web scraping the FIDE Chess Profiles. Quale (talk) 23:01, 17 May 2019 (UTC)
- I was thinking maybe include a column for year of award for International Master title, that way we could identify the exceptional cases like Tal and Christiansen who skipped over the IM title and were awarded the GM title directly. But maybe better to walk before we run. MaxBrowne2 (talk) 07:00, 17 May 2019 (UTC)
- The FIDE rating lists are hugely helpful but frustratingly hard to use. As a site for the old FIDE rating lists I like http://www.mark-weeks.com/chess/ratings/. Olimpbase is good also, and does a much better job of documenting the origin of each file. Of course from Jan 2001 on official lists can be downloaded from FIDE at https://ratings.fide.com/download.phtml, but I imagine everyone here knows that. (Scripting using wget or cURL is recommended over downloading all the old FIDE lists by hand unless you have a lot more patience for clicky-clicky-clicky than I do.) Lists since August 2012 are available in XML which I find a little more convenient to load than fixed column text, but YMMV depending on your tools. Of course the file format is no guarantee of data quality, and neither is the year. I understand 40 year old data OCRed from magazines might be a little imperfect, but
- Another online resource for old ratings is benoni.de, which includes the period from 2001, e.g. http://www.benoni.de/schach/elo/elohis.html?id=1700014&lang=en. It should be treated as unofficial, but it can be used to check the rating history of individual players and the data there can be confirmed at https://ratings.fide.com/download.phtml. MaxBrowne2 (talk) 01:04, 17 May 2019 (UTC)
- For "older" GMs, the complete date of birth can still be found at OlimpBase, which has records of all rating lists from 1971 to 2001. Sophia91 (talk) 21:33, 16 May 2019 (UTC)
- @Quale, Sophia91, MaxBrowne2, and קיפודנחש: FYI: Above, someone mentioned Olimpbase and Benoni. These sources are fine, but they are not machine-readable. In 2018, someone made a large database with all FIDE players that have or had an Elo rating: Database. This database was the result of this scientific article that was published 2006. The creator seems to have undergone the tedious work of working out the problems of different IDs for the same player, different spellings etc. (in the scientific article he talks about these problems in detail). This database includes full names, titles, ratings, DOBs, sex, and federation. It is a Microsoft Access database and can be simply queried with SQL. I have already used it to import all Elo ratings between 1971 and 2018 to Wikidata (which led to nice Elo graphs like de:Viswanathan_Anand#Elo-Entwicklung). 147.142.162.238 (talk) 07:09, 4 June 2019 (UTC)
Discussion at Talk:History of chess#Greco-Roman origin of chess
You are invited to join the discussion at Talk:History of chess#Greco-Roman origin of chess. No Great Shaker (talk) 10:25, 7 July 2019 (UTC)
Unofficial world champion claims
I'm not crazy about language such as "La Bourdonnais was considered to be the unofficial World Chess Champion from 1821". Was considered by who? I don't think Wikipedia should be writing this. The concept of a world chess champion didn't really exist in 1821, so stating that he was considered to be the WC isn't good. Louis-Charles Mahé de La Bourdonnais is not the only article affected as there are similar claims introduced in other articles concerning the strongest players before Steinitz. The description in World Chess Championship#Unofficial champions (pre-1886) is more circumspect and historically accurate. Quale (talk) 03:50, 23 August 2019 (UTC)
- This article by Edward Winter might be of interest. Cobblet (talk) 04:20, 23 August 2019 (UTC)
- I've removed the references to "unofficial World Chess Champion" from the La Bourdonnais article as no source is given. I think it's particularly inappropriate in an infobox.-- Pawnkingthree (talk) 13:12, 28 August 2019 (UTC)
Article Evaluation function
Evaluation function probably isn't Stub class any more. It's Start class. Can I fix this myself, or does a member of this group need to fix it?Sbalfour (talk) 19:02, 1 October 2019 (UTC)
once more, with feeling: pgn viewer
FYI:
the proposal to activate pgn viewer in enwiki is discussed now (yet again) in WP:VPT ("Enable chess PGN viewer as default gadget for chess articles"). if you are interested, weigh in with your nay or yay.
peace.
- Can you provide a link to a sample on the ruwiki (so can observe e.g. non L-shaped knight moves)? Also where is it explained how implementation w/ impact existing WP articles w/ diagrams & game scores? Also, the code s/ intercept PGN "O-O" and display back "0-0" when accompanying game scores are displayed (else it misleads how algebraic notation is presented). --IHTS (talk) 18:25, 30 July 2019 (UTC)
- i recently added "update" to Wikipedia_talk:WikiProject_Chess/Interactive_chess_boards#Kipod's_PGN_viewer, with links to pages in ruwiki, hewiki and ukwiki. i don't understand your comment/question about castling. afaik, the viewer displays castling with O-O or O-O-O. i vaguely remember that if the original PGN says 0-0 or 0-0-0, the script is able to take it as castling - i think i encountered some such files. i don't know if it's legit. please clarify.
- also note the different annotation on ruwiki (piece icons instead of P/R/N/B/K/Q) and on hewiki. it can also support different numerals for row, e.g. ١٢٣٤٥٦٧٨ ("real" Arab numerals, corresponding with the more common "fake" Arab numbers we use - 12345678). peace - קיפודנחש (aka kipod) (talk) 19:38, 30 July 2019 (UTC)
- When moves in AN are displayed for the WP user to see, they s/ see "0-0" not "O-O". ("O-O" is a PGN internal-coding thing.) --IHTS (talk) 22:26, 30 July 2019 (UTC)
- not sure what you mean by "they s/ see". is "s/" short for should? please view the page on ruwiki and compare to the one on hewiki. notice that the "notation" differ: the ruwiki one uses pieces icons instead of (P)RNBKQ, while the hebrew notation uses צ-פ-ר-מ-מה for the pieces, and אבגדהוזח instead of abcdefgh for file. this is done not by using different script, but by passing parameters: the template creates a "div" element with
data-config
attribute, containing the configuration. this allows the template to control the behavior of the script, and one of the controllable "behaviors" is notation substitution, so if 0-0 is desired, np. - part of the configuration convention was developed in collaboration with some editors on ruwiki, specifically User:Раммон. i refreshed a little bit the english demo page on hewiki, and added some documentation - you may want to take a look.
- HTH. if you have
questionscomments, please remember that i'm not fluent in enwiki jargon, so try to be as clear and simple as possible. peace - קיפודנחש (aka kipod) (talk) 06:08, 31 July 2019 (UTC)- Yes (they should see "0-0" not "O-O"). (Sorry for confusion.) Ok methinks you're saying it's easily substitutable. --IHTS (talk) 07:53, 31 July 2019 (UTC)
- @Ihardlythinkso:: on 2nd thought, this is a bit fishy. (slides into geekspeak): the problem is that O-O is a substring of O-O-O. we configure the replacement using JS "object", and in principle, we can't guarantee the order of replacements, so if the O-O replacement will hit first, O-O-O will become 0-0-O (and of course, the O-O-O replacement will not catch it). in practice, all major browsers preserve the order when looping (see, e.g., [3], section "for loop order"), so if the config defines the O-O-O replacement 'before it defines the O-O, everything should work as desired, albeit relying on "undocumented" behavior. i could fix the script such that we won't depend on this little perversion, but i really really don't want to... peace - קיפודנחש (aka kipod) (talk) 16:33, 31 July 2019 (UTC)
- Almost understand you.
(My coding background is assembly lang on IBM mainservers.) Code + imagination + resources can accomplish anything. --IHTS (talk) 17:28, 31 July 2019 (UTC)
- Almost understand you.
- @Ihardlythinkso:: on 2nd thought, this is a bit fishy. (slides into geekspeak): the problem is that O-O is a substring of O-O-O. we configure the replacement using JS "object", and in principle, we can't guarantee the order of replacements, so if the O-O replacement will hit first, O-O-O will become 0-0-O (and of course, the O-O-O replacement will not catch it). in practice, all major browsers preserve the order when looping (see, e.g., [3], section "for loop order"), so if the config defines the O-O-O replacement 'before it defines the O-O, everything should work as desired, albeit relying on "undocumented" behavior. i could fix the script such that we won't depend on this little perversion, but i really really don't want to... peace - קיפודנחש (aka kipod) (talk) 16:33, 31 July 2019 (UTC)
- Yes (they should see "0-0" not "O-O"). (Sorry for confusion.) Ok methinks you're saying it's easily substitutable. --IHTS (talk) 07:53, 31 July 2019 (UTC)
- not sure what you mean by "they s/ see". is "s/" short for should? please view the page on ruwiki and compare to the one on hewiki. notice that the "notation" differ: the ruwiki one uses pieces icons instead of (P)RNBKQ, while the hebrew notation uses צ-פ-ר-מ-מה for the pieces, and אבגדהוזח instead of abcdefgh for file. this is done not by using different script, but by passing parameters: the template creates a "div" element with
- When moves in AN are displayed for the WP user to see, they s/ see "0-0" not "O-O". ("O-O" is a PGN internal-coding thing.) --IHTS (talk) 22:26, 30 July 2019 (UTC)
I got this script worked in the Russian Wikipedia, Russian Wikinews and Ukrainian Wikipedia. Also I failed to do this in the French Wikipedia. Nobody claimed that something didn't work ("O-O" or "O-O-O" or something else). Раммон (talk) 06:37, 31 July 2019 (UTC)
UPDATE: the proposal was archived without implementation from Wikipedia:Interface administrators' noticeboard. not versed in enwiki policy and politics, but basically, as far as i understand, even though this proposal was supported, practically unanimously by the community, no interface admin chose to implement it, and it aged out of the noticeboard. peace - קיפודנחש (aka kipod) (talk) 16:47, 4 October 2019 (UTC)
- That one is too technical for me. It looks like xaosflux is worried about a slippery slope and would prefer an extension. Does that mean the discussion is over? I'm not sure how these things work. A few pings: @Xaosflux, Wugapodes, and Galobtter: — Rhododendrites talk \\ 17:34, 4 October 2019 (UTC)
- archive link — Rhododendrites talk \\ 17:35, 4 October 2019 (UTC)
- The two simplest options are to use kipod's js as a gadget as proposed at the village pump and discussed on IAN or use the existing mw:Extension:PgnJS which works with MW 1.33.0 but unknown how it plays with the rest of enWikipedia. Wug·a·po·des 17:49, 4 October 2019 (UTC)
- TBH, i do not believe the existing extension is an option. this extension was not approved by WMF developers, and rightly so, and i think it's highly unlikely it will ever be, in its current state. since it's pretty much abandoned, its "current state" is not likely to change either, so i think this direction is a dead end. peace - קיפודנחש (aka kipod) (talk) 17:59, 4 October 2019 (UTC)
- I mean, agree, but it's worth mentioning that an extension exists because that's been a concern in previous discussions (and I think briefly at IAN). It doesn't seem anyone's actually against the gadget option, just that some aren't in favor and the discussion stalled (even xaosflux said
I'm not opposed to the gadget route
in response to Writ Keeper). I think the principle disagreement is really whether common.js should be modified to load this gadget specifically, or whether it should be modified to load generic gadgets of which this pgn viewer is one. Wug·a·po·des 18:08, 4 October 2019 (UTC)- There appeared to be agreement with having a generic loading gadget that would be a dependency of the load-on-demand gadgets, so the generic loader would get loaded to handle the on-demand gadgets. Would anyone with the required skills be interested in working on it? isaacl (talk) 19:40, 4 October 2019 (UTC)
- as far as i understand, "skills" and "interested" are less relevant than "has permissions" (unless you meant political skills). this is what the noticeboard is for - it organizes the work of the people with the required permissions (interface administrators). this proposal was pushed out of the noticeboard and archived, after being stale for a long time. true, there is essentially unanimous support, but it's not enough. i doubt anyone intends to do much about it. peace - קיפודנחש (aka kipod) (talk) 05:51, 5 October 2019 (UTC)
- Someone with the right permissions would have to put the changes in place; the spadework of creating the framework as proposed could be done by someone else. I appreciate, though, that given the relatively small size of the change, it might be more efficient for an interface admin to do the implementation (since they would have to perform testing, no matter who implements it). So yeah, unless an interface admin is interested in picking up at least the testing aspect, or someone interested in doing the work becomes an interface admin, progress is stalled. isaacl (talk) 06:19, 5 October 2019 (UTC)
- i have declared in the archived discussion that i'm willing to do any or all of the "spadework", to see this implemented. FWIW, i do have "interface admin" permissions on hewiki, so at least _someone_ thinks i'm capable enough. however, there's not much opportunity for spadework until someone with the required permissions says they'll do it, and asks for help. קיפודנחש (aka kipod) (talk) 17:42, 8 October 2019 (UTC)
- Someone with the right permissions would have to put the changes in place; the spadework of creating the framework as proposed could be done by someone else. I appreciate, though, that given the relatively small size of the change, it might be more efficient for an interface admin to do the implementation (since they would have to perform testing, no matter who implements it). So yeah, unless an interface admin is interested in picking up at least the testing aspect, or someone interested in doing the work becomes an interface admin, progress is stalled. isaacl (talk) 06:19, 5 October 2019 (UTC)
- as far as i understand, "skills" and "interested" are less relevant than "has permissions" (unless you meant political skills). this is what the noticeboard is for - it organizes the work of the people with the required permissions (interface administrators). this proposal was pushed out of the noticeboard and archived, after being stale for a long time. true, there is essentially unanimous support, but it's not enough. i doubt anyone intends to do much about it. peace - קיפודנחש (aka kipod) (talk) 05:51, 5 October 2019 (UTC)
- There appeared to be agreement with having a generic loading gadget that would be a dependency of the load-on-demand gadgets, so the generic loader would get loaded to handle the on-demand gadgets. Would anyone with the required skills be interested in working on it? isaacl (talk) 19:40, 4 October 2019 (UTC)
- I mean, agree, but it's worth mentioning that an extension exists because that's been a concern in previous discussions (and I think briefly at IAN). It doesn't seem anyone's actually against the gadget option, just that some aren't in favor and the discussion stalled (even xaosflux said
- TBH, i do not believe the existing extension is an option. this extension was not approved by WMF developers, and rightly so, and i think it's highly unlikely it will ever be, in its current state. since it's pretty much abandoned, its "current state" is not likely to change either, so i think this direction is a dead end. peace - קיפודנחש (aka kipod) (talk) 17:59, 4 October 2019 (UTC)
- The two simplest options are to use kipod's js as a gadget as proposed at the village pump and discussed on IAN or use the existing mw:Extension:PgnJS which works with MW 1.33.0 but unknown how it plays with the rest of enWikipedia. Wug·a·po·des 17:49, 4 October 2019 (UTC)
Category rename
Where is the discussion of the renaming of category Women's World Chess Champions to Women's World Chess Championship Winners? I don't know how to find these things. The first I knew about it was when some articles on my watchlist got their category renamed.
Is the category World Chess Champions going to be, or has it been, renamed to World Chess Championship Winners?
What about other sports? Bruce leverett (talk) 15:40, 8 October 2019 (UTC)
- There was a discussion here, at Wikipedia:Categories for discussion/Speedy. It seems the original proposal was to change the category to "Women's world chess champions." How it ended up as "Women's World Chess Championship winners" I'm unclear on. It seems to have been suggested by User:Fayenatic london, so I'm pinging them - categories aren't something I'm too knowledgeable about.-- Pawnkingthree (talk) 16:00, 8 October 2019 (UTC)
- Thanks for the ping. The category is, as you say, currently at Category:Women's World Chess Championship winners. There was no further comment after the permalink in the discussion that you've already provided above, so I processed it two days later. In retrospect I should have waited longer. If I have misjudged this, then I will revert the move.
- My reasoning in the discussion was that in some other cases of "XYZ World Championships", the individual champions are categorised as "XYZ World Championship winners". These women are winners of the Women's World Chess Championship, per the parent category Category:Women's World Chess Championships, hence the new name.
- However, I now see that "Women's World Chess Champions" (the original category name) is used at least as a heading in the lead article.
- Would it be best if I revert the move and reopen the discussion on the Speedy page? – Fayenatic London 20:55, 8 October 2019 (UTC)
- Looking through Category:Sports world champions, I do not see a consistent pattern. For instance there is Category:FIFA World Cup winners, but there is Category:World chess champions, and Category:World champions in snooker, and other styles as well. Presumably there would be some advantage if they were more uniform, but I don't know what that might break, e.g. British vs. American usage.
- I think that reopening the discussion must be correct, though I don't know what that will lead to. I am uncomfortable with the present situation, in which the WC category and the WWC category are sharply inconsistent with each other, but I can see why the discussion was opened in the first place. Bruce leverett (talk) 21:53, 8 October 2019 (UTC)
- OK, I have moved the category back to Category:Women's World Chess Champions, and reopened the discussion at the bottom of WP:CFDS#Current requests. Please comment there. – Fayenatic London 20:58, 9 October 2019 (UTC)
- The renaming has now been processed as originally proposed. I think the final comment was [4]. – Fayenatic London 20:22, 14 October 2019 (UTC)
Question about the utility of section anchors
I have a question about section anchors. What is the purpose of edits like this one to add a section anchors? I realize this is old news and we have these all over our chess opening articles now, but I don't understand why. Yes, I get that it's inconvenient to link to [[Indian Defence#Nimzo-Indian Defence: 3.Nc3 Bb4]] and I can see the appeal of mechanically creating nicer anchor shortcuts and gnoming edits can be kind of relaxing. But there are issues:
- Clutters the wikitext for all subsequent editors
- Anchor template expansion is an admittedly microscopic drag on resources, but if could be avoided if it isn't needed
- Shows up in the edit summary line every time someone does a section edit. This is fugly and just takes space from the already limited length of the edit summary
- Why would anyone ever want to link [[Indian Defence#Nimzo-Indian Defence: 3.Nc3 Bb4]] anyway? Just link to the main Nimzo-Indian Defence article.
- Once an anchor like this is created we can't realistically tell if it's in use so it must be maintained forever.
Quale (talk) 23:29, 19 August 2019 (UTC)
- Re "
takes space from the already limited length of the edit summary
", I did some test edits, & maybe inconsistent w/ WP:ES#The 500-character limit (not sure), I get 388 numeric chars for an editsum regardless the length of the secname plus length of anchors ([5] [6]), whereas I got variable (430 to 490 chars [7] [8]) when editsum was alpha only (maybe xplained by the different secnames/anchors involved). --IHTS (talk) 04:58, 23 August 2019 (UTC) - Re your Q "
What is the purpose to add a section anchors?
", yes, the sole purpose is/was convenience (e.g. who wants to link to [[Ruy Lopez#Modern Arkhangelsk Defence: 4.Ba4 Nf6 5.0-0 b5 6.Bb3 Bc5]] when [[Ruy Lopez#Modern Arkhangelsk Defence]] is available). --IHTS (talk) 05:04, 23 August 2019 (UTC) - Re "
we can't realistically tell if it's in use
", doesn't that apply equally to section names? (If so, just as flexible/inflexible to change as secnames thus no "forever" maintenance.) --IHTS (talk) 05:12, 23 August 2019 (UTC) - Re "
clutters the wikitext
", lots of article-invisible things clutter wikitext in WP articles, e.g. citations. In every case when I added secname anchor, I put it to the right of secname, not the left as others have done, so the secname remains detectable in wikitext reading left to right. Also the secname w/ anchor still occupies only one horizontal line in wikitext (I doubt there is even one example that occupies more than one line). --IHTS (talk) 05:19, 23 August 2019 (UTC) - Re "
Shows up in the edit summary line every time someone does a section edit. This is fugly.
" Yes, it is ugly. (I think the ultimate solution to that, however, would be a s/w change that excludes displaying {{anchor|...}} pointlessly, whenever found in the processing of an edit. Rather than bending to poor s/w design. There must be a venue for submitting s/w change requests, I've never done so before but w/ be glad to explore it, submit one, to get the tech dept's feedback & return here to share.) --IHTS (talk) 05:33, 23 August 2019 (UTC) - Re "
Why would anyone ever want to link [[Indian Defence#Nimzo-Indian Defence: 3.Nc3 Bb4]]
". Someone might, & I didn't want them to be pissed off having to specify the extra ": 3.Nc3 Bb4" when other secnames in the same article carry the natural anchor shortcuts. --IHTS (talk) 05:42, 23 August 2019 (UTC)
- Re "
- Those were my rationales when creating the achors. Of course am willing to do (or undo) whatever WP:CHESS specifies re different. --IHTS (talk) 05:46, 23 August 2019 (UTC)
- Wait! (There's HTML option at template:anchor ... allow me to read & test ... ) --IHTS (talk) 03:33, 24 August 2019 (UTC)
- I'm a content guy, I'm not all that concerned with markup issues. They tend to resolve themselves in the end. MaxBrowne2 (talk) 03:53, 24 August 2019 (UTC)
- Yes, it works slick, replacing {{anchor|...}} with {{subst:anchor|...}}. Then only the true secname appears in the edit window. Just as described at template:anchor:
A drawback of this approach is that having a template in the section header causes problems with the edit summary window each time that a section edit is done for this section. ... Within section titles, it may be preferable to simply use direct HTML, which may be achieved by substitution like this: == {{subst:anchor|Foo bar}} Foo == which is saved into the article as: == <span id="Foo bar"></span> Foo == This provides the stable, linkable anchor, but without the edit problem.
- So unless objection, methinks I'll start replacing all the openings articles secname anchors I've previously created, w/ the HTML option. (Also in some games articles as shortcuts to games, e.g. in World Chess Championship 1972, ====Game 6: Fischer–Spassky, 1–0 (QGD Tartakower) {{anchor|Game 6|game 6}}====. --IHTS (talk) 04:19, 24 August 2019 (UTC)
- (The only drawback seems to be that when {{subst:anchor|...}} is expanded, if there are more than one anchors specified, each one generates its individual <span id="..."></span>, so in the wiki code that'll take up more line space than before; however, multiple anchor names is pretty much the exception not the "rule".) --IHTS (talk) 04:29, 24 August 2019 (UTC)
- In the interrim, have chg'd to the HTML for articles Indian Defence, Vienna Game, & King's Gambit. --IHTS (talk) 04:56, 24 August 2019 (UTC)
- Thanks for taking the time to explain another viewpoint. IHTS, are you familiar with YAGNI? Cluttering wikitext with pointless edits to add anchors that are never going to be used is idiotic. When someone needs to link to a section they can add an convenience anchor if needed. Otherwise, YAGNI. Quale (talk) 02:41, 26 August 2019 (UTC) And certainly many things clutter wikitext, but those things usually have some value. Unlike these section anchors which have no value, or in fact negative value as they make things worse and not better. Quale (talk) 02:47, 26 August 2019 (UTC)
- So you want all those I've added removed (openings articles & games articles), and my adding them was idiotic. Got it! --IHTS (talk) 03:26, 26 August 2019 (UTC)
- What *I* think is "idiotic" is having section names consisting of both natural names plus colon plus notated moves, making them fussy to link to because of the meticulous specification required to wlink to them. (And that situation is unique to chess articles, no other project category has such combined secnames.) Eat me! --IHTS (talk) 03:30, 26 August 2019 (UTC)
- I don't concentrate on the Opening articles, but I took a look at this, and I see the dilemma.
- Normally when you create a section head, you get an anchor "for free". So for instance, you can refer to Bobby Fischer#Forfeiture_of_title. Nobody has to stop and think whether the anchor is ever going to get used or not; it's just there in case anyone needs it.
- But in these opening articles, the section titles are somewhat error-prone to use as anchors. In typing Indian Defence#Blumenfeld_Gambit:_3.Nf3_c5_4.d5_b5, quite aside from knowing whether it's "Gambit" or "Counter Gambit", you have to remember the colon, and whether or not each move number is followed by a period (and whether or not that period is followed by a space). I can see that a case can be made for adding simplified anchors, in spite of the annoyance of having all the anchor scaffolding laying around. Bruce leverett (talk) 14:20, 26 August 2019 (UTC)
- Exactly. (For a thought experiment, imagine *all* openings articles secnames that have the natural name, followed by colon, followed by notated moves, if they had the natural name as anchor. In that world, editors w/ know, or soon learn, they never have to specify the meticulous long secname perfectly in order to wlink. All they need to specify is the natural name. They w/ be conditioned to not even have to look into the wikicode to see what is the anchor [it w/ always include the natural name]. Not only is editing easier in that world, it saves unnecessary wikicode clutter [unnecessarily long & detailed secnames] in all articles that link, or ever will link.) So what is the cost? The anchor template does not even add a single additional line to the wikicode. (So how is it a viable argument that it "clutters" the wikicode?!) The edit box can be misleading since the editsum length isn't restricted to the rectangular box displayed, it extends to up to 500 char elements, large enough for practically any editsum. The processing time to interpret the anchor template is most undoubtedly a nit. --IHTS (talk) 15:42, 26 August 2019 (UTC)
- The idea of adding anchors in this setting (secname has natural name plus colon plus notated moves) "only when needed", is appreciated; however, am not exactly sure the definition of "needed". (Technically, anchors are *never* needed. They are a convenience. So when are they "needed"? Presumably, if the linking article needs to link numerous times to the same secname. Ok. But, will an editor in that case, *add* the anchor? Probably not. [Since it involves more "work" to edit a different article, add the anchor wikicode. They'll probably just duplicate their long secname wlinks numerous times w/ copy/paste.] So I think the "when needed" argued by Quale w/ never exist in real terms, or if it did & the user went thru the step of adding the anchor to a second article, it w/ be the rare exception. So ... the result is wlinks in numerous linking articles w/ unnecessarily long secnames, cluttering multiple articles.) --IHTS (talk) 15:58, 26 August 2019 (UTC)
- Thx for the above post, Bruce. (The kind of "drill-down" I've added is what I consider part of "discussion", but it's clear Quale wasn't in any mood for any discussion. This Talk is supposed to be for good-faith discussion? Or for complaint & unwarranted insult that immediately ends discussion cold?) --IHTS (talk) 16:09, 26 August 2019 (UTC)
- Bruce, I'm finding it hard to believe that your wikitext editing skills are that lame. "In typing Indian Defence#Blumenfeld_Gambit:_3.Nf3_c5_4.d5_b5, quite aside from knowing whether it's "Gambit" or "Counter Gambit", you have to remember the colon, and whether or not each move number is followed by a period (and whether or not that period is followed by a space)." First, you don't (see below), and second, yeah, so??? Unless you are advocating that the section anchors should be pre-created with "Gambit", "Countergambit", "Counter Gambit", "Defense", "Defence", "Attack", "Reversed", and every other possible variation that applies in the particular context, you still have to check that the anchor exists and exactly how it is spelled. Because in fact you can't ever count on the anchor you want existing unless you add it yourself, so you will have to check, and since the anchor is not apparent in the displayed article, you have to check the section wikitext itself. While you're there with the section opened for edit, just add whatever damn anchor you want when you are actually going to use it. I'm afraid don't see how this is a sensible argument for creating unused section articles since it completely ignores that the reality of what you need to do to use a pre-created anchor requires work on the same order of what you are claiming to save. Claiming to save in the hypothetical case that one of these pre-created anchors might actually get used, but since we know the vast majority will never get used it's actually a lot of extra work to create unused section anchors. That extra work wouldn't matter if unused section anchors were harmless, but they aren't.
- I've got a much easier suggestion that doesn't require remembering anything at all or examining the wikitext for pre-existing anchors or adding a new anchor – just copy and paste the section title into the link exactly as it appears in the displayed article. (Yes, Indian Defence#Blumenfeld Gambit: 3.Nf3 c5 4.d5 b5 works with the section title pasted as it appears in the displayed article verbatim, no need to translate the spaces into underscores which is generally discouraged anyway.) No worries about "Defence" vs. "Defense" and the numbers and punctuation will work just fine pasted into the link. Characters that would actually mess up the link in a way that wouldn't allow a simple copy and paste such as | or ] are not used in these section titles. (In those very rare cases section anchors can be truly helpful, but again only if the anchors are used. I don't know why it's so hard to understand that adding unused crap to the wikitext for entirely speculative reasons is not a good idea even if it would be hypothetically useful in the rare case where, you know, it was actually used.)
- Why do you guys make this so goddamn hard? You act like creation of a link to these sections is an act akin to neurosurgery when all it requires is copy and paste, a simple computer skill mastered by millions of 8-year-olds. Another option is to change the section titles to remove the move notation wherever that is reasonable. I think that would improve Indian Defence, for example. So the rationale for adding many dozens of these shit unused anchors is because "someone might" want to use them and "I didn't want them to be pissed off" by needing to paste in the section titles (oh, the horror!!) in the case where they hypothetically might want to use the anchor. What about the editors pissed because we're dragging all those unused crap section anchors around forever? I guess the hypothetical editor who is pissed because he expected to his anchor sitting unused but created already is more important than the certainty of forcing useless shit into the the edit summary of every section edit. I really can't express in words how stupid that rationalization is, and how idiotic I find it to prioritize completely unproven supposed potential future benefits over obvious current costs. So far no one has identified a single time where these anchors have been beneficial compared to just adding anchors only when they are actually used, and the cost paid every day is readily apparent to anyone following Recent Changes: Index of chess articles. Quale (talk) 07:10, 28 August 2019 (UTC)
- Thanks for taking the time to explain another viewpoint. IHTS, are you familiar with YAGNI? Cluttering wikitext with pointless edits to add anchors that are never going to be used is idiotic. When someone needs to link to a section they can add an convenience anchor if needed. Otherwise, YAGNI. Quale (talk) 02:41, 26 August 2019 (UTC) And certainly many things clutter wikitext, but those things usually have some value. Unlike these section anchors which have no value, or in fact negative value as they make things worse and not better. Quale (talk) 02:47, 26 August 2019 (UTC)
Specify again what are "obvious current costs
". (Your contention anchor templates "clutter wikicode" is idiotic and makes no sense - the addition of an anchor template doesn't even add one line to the wikicode. That the anchor code shows up in the editbox doesn't meaningfully subtract from avail length for editsums, so that was an idiotic argument too and also makes no sense. Apparently the cost you are most having a cow about is having the anchor code show up in the editbox and in article edit histories?) Specify succinctly rather than having an idiotic fit on the Talk page. --IHTS (talk) 08:43, 28 August 2019 (UTC)
Also, explain "needed" in "they can add an convenience anchor if needed
". (To suggest an anchor can be "needed", without letting know how or why, while simultaneously shouting down openings articles natural-name secname anchors, is idiotic and makes no sense.) --IHTS (talk) 09:39, 28 August 2019 (UTC)
BTW, & FYI, the "reality" is, most all openings articles w/ secnames consisting of natural name plus colon plus notated moves, have natural-name anchors today. (Can think of no major exceptions.) And that has been the "reality" for some time (maybe essentially years). --IHTS (talk) 09:50, 28 August 2019 (UTC)
Well, yes, I remembered copy & paste, after or perhaps while writing my comment. That's the sensible way to cite sections, but how often have I actually used that method?
I am also familiar with YAGNI though I didn't know the acronym. Programmers have a tendency to write stuff they might need someday. Perhaps even the more experienced they are, the more they acquire that bad habit.
I would not recommend that we get too heated up over this issue. Although I am frequently shocked at how much of the Web is copied from Wikipedia, thank goodness they do not copy our source. Bruce leverett (talk) 12:06, 28 August 2019 (UTC)
To Quale's "Unless you are advocating that the section anchors should be pre-created with "Gambit", "Countergambit", "Counter Gambit", "Defense", "Defence", "Attack", "Reversed", and every other possible variation that applies in the particular context
", I have resolved or eliminated that issue, by changing the secname (and article name) to the OCC (1996) name of "Blumenfeld Countergambit". (Therefore, the fuss re that is now a red herring non-issue. Openings variations names are either OCC names, or names from RSs that override the OCC name, depending on sources. But there is *one name* in any event, with synonym names possibly, but those synonym names can or don't have to be additional anchors or REDIRECTs - whatever.) So to make that particular openings variation name an issue is ... idiotic, and makes no sense. --IHTS (talk) 12:41, 28 August 2019 (UTC)
FYI, today went thru *every* openings article containing anchor(s) & reduced shortcut specifications such that no anchor contains more than one shortcut name (the section title natural name portion). (Namely, in Bird's Opening, English Opening, Indian Defence, Sicilian Defence, Dragon Variation, Sicilian Defence, Scheveningen Variation, Pirc Defence, Caro–Kann Defence, Sicilian Defence, French Defence, Danish Gambit, Vienna Game, King's Gambit, Philidor Defence, Petrov's Defence, Giuoco Piano, Two Knights Defense, Ruy Lopez, Blackmar–Diemer Gambit, Slav Defense, Queen's Gambit Accepted, Semi-Slav Defense, Grünfeld Defence, Nimzo-Indian Defence, and King's Indian Defence, Sämisch Variation.) --IHTS (talk) 09:10, 29 August 2019 (UTC)
Another way to do (for no extended text in edit box when editing the sec) is to place anchor above & out of the sec title wikicode altogether, as someone did at article Fast chess by placing {{anchor|Rapid|Rapid chess|Rapid play}} above secname ===Rapid (FIDE) or quick (USCF)===. --IHTS (talk) 03:28, 15 October 2019 (UTC)
Rapid (FIDE) or quick (USCF)
Online chess databases as sources
The use of online chess databases as sources is troublesome. This is not a new observation as concerns have been discussed on this chess project talk page in the past. I think previous discussions were primarily concerned with use of online games databases to compile lifetime records in chessplayer bios. Here my concern is use of chess games and openings databases to provide win/loss statistics for chess opening articles. I see five issues with online chess databases when used as sources for wikipedia. Online chess databases are:
- Incomplete/Incomprehensive
- Inaccurate
- Indiscriminate/Irrelevant
- Insufficient/Insignificant
- Inconstant/Irretrievable/Irreproducible/Irretentive
Some chess editors have disputed concerns 1 and 2 asserting that the online databases are complete and accurate. I disagree, especially for older games, but these are not the key issues with use in chess opening articles.
Concern 3 is that the selection of games is indiscriminate and games from all sources may be included. When used in Wikipedia results are not characterized by player rating, time control, etc. Often chess writers will restrict their attention to games by strong players (say IMs or GMs) and may often prefer games played at classical time controls to blitz games, but that isn't done when presenting win/loss numbers in Wikipedia.
Concern 4 is that the evidence from the the databases is insufficient for the claims made. Durkin Opening was recently edited to add the claim "The Durkin Opening has the highest win percentage of all first moves for white in both the Chessgames.com and Chess365.com opening databases." This was a good faith edit and in fact it is true (although we would write "White" rather than "white"), but it lacks important detail relating the the significance of the claim. Chessgames.com has 25 games starting with 1.Na3 and chess365.com has 32 games. This is far too small a sample to draw any conclusion about chances to win or lose. A single additional game added to the database could change the results by 3 or 4 percent. I don't know the precise minimum number of games necessary for the database results to be relevant, but I am skeptical about the significance of any such statistics based on less than a thousand games.
Concern 5 may be hard for some editors to understand or accept, but for me it is the absolutely fatal flaw for most uses of online databases to give win/loss percentages. The database results are not verifiable. The databases are not static so in a year or month or possibly even a week or a day there will be no way to verify that the claim in the article was ever true. What is really needed is for the database search to be recorded in another reliable source and Wikipedia can report that. Ironically the older databases on CD or DVD do better here. Although they will never be up to date, a claim made about the games database distributed on the Fritz 8 CDs could be verified in a way that is simply not possible for an online database. Quale (talk) 20:52, 13 October 2019 (UTC)
- Agree on all points, and let me also mention that a many games that appear in databases are submitted by the winner. This is probably most of the explanation for the good percentage for the Durkin. Who is going to submit a win against the Durkin? But if someone wins with the Durkin, they are going to want to show it off.
- Regarding Incomplete/Incomprehensive, I have to admit that I was pleased to find that Chess365 had all the games from the 1985 Midwest Masters, presumably because there was a print tournament book for that tournament, which somebody got hold of and transcribed. But for some other tournaments, their coverage of me is spotty, including games in which I had a notable opponent.
- It sounds like we have a consensus of two, but I suppose I would like to hear from some others before I tear through Wikipedia chess articles ripping out references to win-loss percentages. Bruce leverett (talk) 21:46, 13 October 2019 (UTC)
- Online databases absolutely should not be used as sources for the success rate of openings, for all the reasons Quale gives above. Pawnkingthree (talk) 23:57, 13 October 2019 (UTC)
- I made the edit. I guess it was intended more as "trivia" than as encyclopedic objective truth, and I see no problem with removing the information. It is, however, an interesting fact that this opening according to some sources, however fallible, has a high win percentage - it could merit to be mentioned if it could be referenced better. The thing is that chess is a pragmatic game - it does not matter whether an opening is theoretically or "encyclopedically" flawed as long as it works in practice, and current chess opening articles do not seem to reflect this perspective at all. (Note the tendency of the current world champion to play unconventional openings. The point is precisely that the opponent's knowledge is Incomplete/Incomprehensive, Inaccurate, Indiscriminate/Irrelevant, Insufficient/Insignificant, Inconstant/Irretrievable/Irreproducible/Irretentive, which means that playing "objective" chess is not necessarily the strategy of every player.) I do agree that the mentioned databases are a somewhat flawed source of information, at the same time opening databases like this do convey interesting information. Narssarssuaq (talk) 16:02, 14 October 2019 (UTC)
- Agree with Quale on all points. With respect to the point Narssarssuaq is trying to make, there are numerous books (e.g., any book on offbeat openings will discuss this) one could cite regarding pragmatic chess opening choices, without having to rely on cherry-picked statistics. Indeed, talking about Carlsen's opening repertoire would illustrate the point much better than quoting 1.Na3 "statistics". Mark Twain comes to mind. Cobblet (talk) 16:26, 14 October 2019 (UTC)
- Agreed, except that I did not use "cherry-picked" statistics, as these are the best available free opening databases. Like it or not, our best available sources suggest that the Durkin opening is the chess opening with the highest win percentage. Narssarssuaq (talk) 14:47, 15 October 2019 (UTC)
- A provocative thing to say. So I went to chesslab.com, and requested Durkin games in which one player or the other was rated 2400 or above, since 1991. I got 11 games, and White scored 50%. I think you could do better than 50% with almost any respectable opening. Of course, I would have gotten different numbers if I had modified my parameters; for example, when I went for 2500 or above, I only got 7 games, and White was less than break-even.
- This exercise, however silly, reminded me of Quale's points. I don't know the circumstances of most of the games -- tournament? simul? casual? I don't know how reliable any of the information is, or where chesslab.com got it. If I ran the same queries next month, I might get different results, if more Durkin games had been submitted. Is chesslab better than chessgames.com or chess365.com? I have no idea.
- To be fair, I should mention that when I use chesslab.com to get on the ground level of an opening that I want to play or even just study, I do check the win and loss counts. If the counts aren't "normal", that can be a useful warning flag -- something to keep in mind while I play over the games. But that's a long way from being suitable for use in Wikipedia. Bruce leverett (talk) 19:17, 15 October 2019 (UTC)
- I completely agree, but again: Maybe you are looking for an objectivity that doesn't really exist in chess. Perhaps chess is not a science but merely a game of cunning. Narssarssuaq (talk) 21:34, 15 October 2019 (UTC)
- You looked for facts to support your theory, but when confronted with other facts that contradicted it, you denied that objectivity was possible. Perhaps chess would appear more like a science if you tried instead to look for a theory to support the facts, but it appears that such an approach does not interest you. To each their own, but on Wikipedia, NPOV is the rule. Cobblet (talk) 03:57, 16 October 2019 (UTC)
- I am trying to explain the subtle point that practical chess is decision-making under uncertainty, as opposed to decision-making with perfect knowledge. This might have some bearing on how chess openings optimally should be represented on Wikipedia (it would put into question the idea of "theoretically sound" opening moves as a gold standard for how to play openings). At the same time, such an exercise could be excessively ambitious and open the door to an endless mess of accidental and contextual information, and I thus do not see any reason to fundamentally disagree with what you write. Narssarssuaq (talk) 21:44, 16 October 2019 (UTC)
- I agree completely with these statements, although I might phrase it in a different way: it is not so much that players like Carlsen, Rapport, Jobava, and others have questioned the idea of theoretical soundness, but rather that they have shown that there are many more openings that are theoretically sound, and moreover will pose opponents problems in practice, than what had previously been assumed. There are reliable sources that acknowledge this trend (Axel Smith's book on 1.Nf3/2.e3 comes most readily to mind) and those are the sources we should be using to explain this advance in our understanding of chess. Cobblet (talk) 03:17, 17 October 2019 (UTC)
- I am trying to explain the subtle point that practical chess is decision-making under uncertainty, as opposed to decision-making with perfect knowledge. This might have some bearing on how chess openings optimally should be represented on Wikipedia (it would put into question the idea of "theoretically sound" opening moves as a gold standard for how to play openings). At the same time, such an exercise could be excessively ambitious and open the door to an endless mess of accidental and contextual information, and I thus do not see any reason to fundamentally disagree with what you write. Narssarssuaq (talk) 21:44, 16 October 2019 (UTC)
- You looked for facts to support your theory, but when confronted with other facts that contradicted it, you denied that objectivity was possible. Perhaps chess would appear more like a science if you tried instead to look for a theory to support the facts, but it appears that such an approach does not interest you. To each their own, but on Wikipedia, NPOV is the rule. Cobblet (talk) 03:57, 16 October 2019 (UTC)
- I completely agree, but again: Maybe you are looking for an objectivity that doesn't really exist in chess. Perhaps chess is not a science but merely a game of cunning. Narssarssuaq (talk) 21:34, 15 October 2019 (UTC)
- Agreed, except that I did not use "cherry-picked" statistics, as these are the best available free opening databases. Like it or not, our best available sources suggest that the Durkin opening is the chess opening with the highest win percentage. Narssarssuaq (talk) 14:47, 15 October 2019 (UTC)
- Without a standard context across which all openings can be compared, the relative win percentages are meaningless, though. isaacl (talk) 21:51, 15 October 2019 (UTC)
- Agreed - but is such a standard context available? Narssarssuaq (talk) 21:46, 16 October 2019 (UTC)
- No, which is why we should not cite such percentages. Cobblet (talk) 03:17, 17 October 2019 (UTC)
- As has been well explained by the first post, no. And given that different players will favour different openings against different opponents, I don't see how a common context can be established. Since you agree the relative win percentages are meaningless, I'm not sure why you want to cite them. isaacl (talk) 18:13, 17 October 2019 (UTC)
- The standard context for comparison has already been explained by several people in this discussion. A standard context is provided not by Wikipedia or by the database, but rather by a WP:RS reliable source that interprets the database. Wikipedia can then cite the reliable source that interprets the database rather than citing the database itself. Such sources do exist, for example Larry Kaufman cites some database results in The Kaufmann Repertoire for Black and White. John Watson cites database results in his books as well. Quale (talk) 02:47, 18 October 2019 (UTC)
- I was commenting specifically on
our best available sources suggest that the Durkin opening is the chess opening with the highest win percentage
, which compares the win percentages of different openings. But since the openings are used in all sorts of different circumstances, the win percentages can't be compared. I fully agree that reliable sources, as always, are the appropriate basis for content in a Wikipedia article. isaacl (talk) 03:35, 18 October 2019 (UTC)
- I was commenting specifically on
- The standard context for comparison has already been explained by several people in this discussion. A standard context is provided not by Wikipedia or by the database, but rather by a WP:RS reliable source that interprets the database. Wikipedia can then cite the reliable source that interprets the database rather than citing the database itself. Such sources do exist, for example Larry Kaufman cites some database results in The Kaufmann Repertoire for Black and White. John Watson cites database results in his books as well. Quale (talk) 02:47, 18 October 2019 (UTC)
- Agreed - but is such a standard context available? Narssarssuaq (talk) 21:46, 16 October 2019 (UTC)
- Without a standard context across which all openings can be compared, the relative win percentages are meaningless, though. isaacl (talk) 21:51, 15 October 2019 (UTC)
- My view is that chessgames.com is a convenient way to get a URL to a specific chess game, but not a reliable source as such. MaxBrowne2 (talk) 05:15, 15 October 2019 (UTC)
- +1. any attempt to draw conclusion from statistical (or other) analysis of large number of games, is clearly Original research, and should not be used. let researches go over the data, publish their finding, and then cite those publications. data mining is tricky business, with some of the pitfalls highlighted above. online databases may be legit sources, but we should not engage in mining those databases for overall claims and conclusions. קיפודנחש (aka kipod) (talk) 23:21, 16 October 2019 (UTC)
- At the same time, though, the win perecentage of a first move is a very simple statistic and thus not really the result of data mining. If we had a comprehensive database with a certain authority that we could cite, which we apparently do not, it would likely not appear as strange to cite the win percentage of each first move. Narssarssuaq (talk) 00:16, 17 October 2019 (UTC)
- Statistics are never simple. Methods of data selection are acutely prone to bias and nobody has ever bothered in the context of generating meaningful chess opening statistics to devise mathematically rigorous methods of controlling for bias. Cobblet (talk) 03:17, 17 October 2019 (UTC)
- If this information was something we should be striving to include, then It would be simple to find these stats being cited by notable chess writers, so Wikipedia editors wouldn't have to do any original(?) research. But it's not simple to find, because, the stats are actually not that useful. --SubSeven (talk) 06:41, 17 October 2019 (UTC)
- At the same time, though, the win perecentage of a first move is a very simple statistic and thus not really the result of data mining. If we had a comprehensive database with a certain authority that we could cite, which we apparently do not, it would likely not appear as strange to cite the win percentage of each first move. Narssarssuaq (talk) 00:16, 17 October 2019 (UTC)
Request for information on WP1.0 web tool
Hello and greetings from the maintainers of the WP 1.0 Bot! As you may or may not know, we are currently involved in an overhaul of the bot, in order to make it more modern and maintainable. As part of this process, we will be rewriting the web tool that is part of the project. You might have noticed this tool if you click through the links on the project assessment summary tables.
We'd like to collect information on how the current tool is used by....you! How do you yourself and the other maintainers of your project use the web tool? Which of its features do you need? How frequently do you use these features? And what features is the tool missing that would be useful to you? We have collected all of these questions at this Google form where you can leave your response. Walkerma (talk) 04:24, 27 October 2019 (UTC)
Requested move for casing
A requested move regarding upper and lower casing of multiple chess pages is open at Talk:Three-Man Chess#Requested move 28 October 2019. Randy Kryn (talk) 10:34, 28 October 2019 (UTC)
Recent growth
The project has nearly 5400 articles now (see WP:CHESS for a count of articles marked with {{WikiProject Chess}} on the talk page). I didn't notice when the 5000 article milestone was reached. The project had 4000 articles in 2013. Earlier milestones are recorded in WP:CHESS#Milestones. Thanks to every chess editor, and especially those responsible for the recent surge in new chess biographies. Quale (talk) 05:20, 4 November 2019 (UTC)
Requested moves
Several dozen requested moves at Talk:David Pritchard (chess player). Bruce leverett (talk) 02:59, 18 November 2019 (UTC)
Carissa Yip
A "Did You Know" nomination at Template:Did you know nominations/Carissa Yip could use attention from a chess editor. Anyone interested, please weigh in. —David Eppstein (talk) 19:00, 8 December 2019 (UTC)