User talk:Magnus Manske/Archive 1

Disambiguation link notification
Hi. In your recent article edits, you've added some links pointing to disambiguation pages. Such links are almost always unintended, since a disambiguation page is merely a list of "Did you mean..." article titles. Read the FAQ* Join us at the DPL WikiProject.


 * Thomas Gaspey (check to confirm | fix with Dab solver)
 * added links pointing to English, Gilbert à Beckett and Horace Smith


 * Crisp Gascoyne (check to confirm | fix with Dab solver)
 * added a link pointing to English


 * Harvey Pirie (check to confirm | fix with Dab solver)
 * added a link pointing to Scottish

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, DPL bot (talk) 10:56, 14 January 2012 (UTC)

stats.grok.se
A quick heads up, since I think some of your tools use the api of stats.grok.se: the json format has changed slightly to make it a bit more clear to which entries belong to which dates. If you access stats-classic.grok.se, you get the old format, but I'm not sure how long I'll keep running that service. henrik • talk  11:46, 17 January 2012 (UTC)

Urs Graf, Zeichnung von 1529
Hallo Magnus; du hast hier 2005 den Artikel über Urs Graf angelegt und darin geschrieben "In 1527, he vanished from the city, never to be found again, though a signed drawing from 1529 exists", was sich auch in der aktuellen Version des Artikels noch so erhalten hat. Der Artikel gibt daher in der Einleitung "died after 1529" an. Nun kann ich leider nirgends etwas über diese Zeichnung von 1529 finden, weisst du vielleicht noch, wo du die Information her hattest? Der Artikel im Historischen Lexikon der Schweiz (HLS), ebenfalls von 2005, gibt als Todesdatum "vor 13.10.1528" an, üblicherweise sollte das HLS auf dem aktuellen Stand der Forschung sein. Jedenfalls haben wir so aktuell eine Diskrepanz zwischen der englischen und der deutschen Wikipedia, die ich gerne bereinigen würde. Gestumblindi (talk) 03:08, 20 January 2012 (UTC)
 * Ich möchte ja nichts überstürzen, aber mangels Antwort würde ich dann also wohl demnächst die Angaben zum Todesjahr dem HLS anpassen und die Aussage zum "signed drawing from 1529", da unbelegt, aus dem Artikel entfernen. Gestumblindi (talk) 02:51, 31 January 2012 (UTC)
 * Keine Ahnung, wo ich das vor 7 Jahren gefunden habe. --Magnus Manske (talk) 09:19, 31 January 2012 (UTC)
 * Nun, danke, dann hab ich's mal rausgenommen und das ungefähre Todesdatum aus dem HLS genommen. Deinen Kommentar "what a character" kann man jedenfalls unterschreiben ;-) Gestumblindi (talk) 01:07, 5 February 2012 (UTC)

Table syntax
Magnus, How do we fix the column widths in a wikitable? (Thought you'd be the right person to ask because you invented the syntax!) Deryck C. 17:30, 25 January 2012 (UTC)


 * Not sure about the case at hand, but generally:

|width="20px"|blah||width="40px"|more blah If this doesn't work, please point me to the case you're working on. --Magnus Manske (talk) 13:11, 26 January 2012 (UTC)


 * The page we're working on is Wikimania 2013 bids/Hong Kong. We've got round the problem some other way (I think), and I'll drop you a line if that doesn't solve the problem next time we need it. Thanks! Deryck C. 17:01, 31 January 2012 (UTC)

Add OAuth+OpenID to TUSC
Hello Magnus. Do you have any plans to add OAth+OpenID support to TUSC? I would love to let users log in with their Google (or other) accounts, and be associated with a Wikimedia account for tool usage. If you'd be interested in doing this, I'll create a JIRA request ticket. — Pathoschild 07:20, 04 February 2012 (UTC)
 * Personally, I'd rather go with Mozillas browserID, but that's negotiable. The main issue here is that, with TUSC being linked to a Wiki(m|p)edia account, it checks if that account is currently blocked. If I allow any other kind of login, I need to provide my own blocking mechanism, otherwise people could use all tools that rely on TUSC to flood Wikipedia/Commons/etc. But who can block? Now I need an admin system. And interfaces for both. So, rather not ;-) But I'll float the idea on the toolserver mailing list, maybe something comes up. --Magnus Manske (talk) 11:29, 9 February 2012 (UTC)

Gadget/proposals
Hallo Magnus. Ich habe dein Tool als Gadget vorgeschlagen. Leider scheint es Probleme zu geben, die ich nicht lösen kann. Magst du mal vorbeischauen? --Leyo 10:41, 9 February 2012 (UTC)

Speedy deletion nomination of Great Escarpment


A tag has been placed on Great Escarpment requesting that it be speedily deleted from Wikipedia. This has been done under section A1 of the criteria for speedy deletion, because it is a very short article providing little or no context to the reader. Please see Wikipedia:Stub for our minimum information standards for short articles. Also please note that articles must be on notable subjects and should provide references to reliable sources that verify their content.

If you think that this notice was placed here in error, contest the deletion by clicking on the button labelled "Click here to contest this speedy deletion". Doing so will take you to the talk page where you will find a pre-formatted place for you to explain why you believe the page should not be deleted. You can also visit the page's talk page directly to give your reasons, but be aware that once tagged for speedy deletion, if the page meets the criterion, it may be deleted without delay. Please do not remove the speedy deletion tag yourself, but don't hesitate to add information to the page that would render it more in conformance with Wikipedia's policies and guidelines. If the page is deleted, you can contact one of these administrators to request that the administrator userfy the page or email a copy to you. Thetechexpert (talk) 00:45, 13 February 2012 (UTC)

Proposed deletion of FUBAR


The article FUBAR has been proposed for deletion&#32; because of the following concern:
 * The article is not encyclopedic per WP:DICDEF.

While all contributions to Wikipedia are appreciated, content or articles may be deleted for any of several reasons.

You may prevent the proposed deletion by removing the notice, but please explain why in your edit summary or on the article's talk page.

Please consider improving the article to address the issues raised. Removing will stop the proposed deletion process, 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. --> Gggh  talk/contribs 20:04, 17 February 2012 (UTC)

Nomination of FUBAR for deletion
A discussion is taking place as to whether the article FUBAR is suitable for inclusion in Wikipedia according to Wikipedia's policies and guidelines or whether it should be deleted.

The article will be discussed at Articles for deletion/FUBAR until a consensus is reached, and anyone is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on good quality evidence, and our policies and guidelines.

Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion template from the top of the article. --> Gggh  talk/contribs 09:45, 18 February 2012 (UTC)

Nomination of James Begg for deletion
A discussion is taking place as to whether the article James Begg is suitable for inclusion in Wikipedia according to Wikipedia's policies and guidelines or whether it should be deleted.

The article will be discussed at Articles for deletion/James Begg until a consensus is reached, and anyone is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on good quality evidence, and our policies and guidelines.

Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion template from the top of the article. -- Alan Liefting (talk - contribs) 05:39, 21 February 2012 (UTC)

File:Eric Voegelin.jpg
Hi, I noticed that the above file doesn't resemble those of Eric Voegelin in non-free sources such as this. Could it be a case of mistaken identity?Autarch (talk) 21:10, 21 February 2012 (UTC)
 * Probably. It's nominated for deletion now anyway, so I've removed it from the article. --Magnus Manske (talk) 21:54, 21 February 2012 (UTC)
 * Thanks!Autarch (talk) 03:23, 22 February 2012 (UTC)

MtC drive
Ebe 123  → report 14:48, 28 February 2012 (UTC) Ebe 123  → report 23:13, 29 February 2012 (UTC)
 * https://jira.toolserver.org/browse/MAGNUS-300
 * https://jira.toolserver.org/browse/CHTWO-43

Documentation for gadget authors
I saw you had done some work on gadgets. We're trying to start a library for gadget authors to use. Please check it out and post any questions or comments there — ☠ MarkAHershberger ☢ (talk) ☣ 19:06, 14 March 2012 (UTC)

CatScan V2
Hallo Markus Magnus Manske, zunächst entschuldige das ich Dich nicht im JIRA-System anschreibe. Dort komme ich irgendwie nicht zurecht. CatScan arbeitet bei der Suche im deutschen Sprachraum nicht mehr richtig. Die Antwort ist :MYSQL error : The MySQL server is running with the --read-only option so it cannot execute this statement [INSERT IGNORE INTO cat2 ( catname ) VALUES ( "Wikipedia:Defekter_Dateilink" )]. Mein Betriebssystem - Win XP, Browser - Opera oder FireFox. Kannst Du das Tool wieder in Gang bringen? Danke --Knochen ﱢﻝﱢ‎ 15:54, 17 March 2012 (UTC) — Preceding unsigned comment added by Knochen (talk • contribs)
 * Das Tool funktioniert, nur der Toolserver kränkelt etwas. Soweit ich weiss, läuft die Reparatur, dauert aber eine Weile. Und, wie der Seitentitel verrät, mein Name ist Magnus... --Magnus Manske (talk) 22:39, 18 March 2012 (UTC)
 * Hallo, entschuldige bitte das ich einen falschen Namen verwendete. Danke für die Auskunft zu dem Tool. Wenn die Reparatur bereits läuft wird es ja nur eine Frage der Zeit sein bis es wieder funktioniert. --Knochen ﱢﻝﱢ‎ 03:27, 19 March 2012 (UTC)

Benutzerseite
Hallo Magnus,

könntest du bitte meine Benutzerseite löschen? Danke, — Derschueler (talk) - German Wikipedia 06:58, 27 March 2012 (UTC)

Nomination of Mainstreaming for deletion
A discussion is taking place as to whether the article Mainstreaming is suitable for inclusion in Wikipedia according to Wikipedia's policies and guidelines or whether it should be deleted.

The article will be discussed at Articles for deletion/Mainstreaming until a consensus is reached, and anyone is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on high-quality evidence and our policies and guidelines.

Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion template from the top of the article.  Sandstein  22:33, 5 April 2012 (UTC)

Unicode in CatScan
Hello. Refering to https://jira.toolserver.org/browse/MAGNUS-203 I've been trying to use CatScan for categories on hi-wp, but haven't been able to get it to work with either unicode or percent encoding or Hexadecimal NCRs or javascript escapes.

Two example categories whose subset shouldn't be empty are hi:श्रेणी:इलाहाबाद_जिला_के_गाँव and hi:श्रेणी:उत्तरप्रदेश_के_गाँव. I've tried using catscan with other categories on hi-wp too, but it doesn't seem to work.

Hope this is easily fixable. Thanks.--Siddhartha Ghai (talk) 14:44, 20 April 2012 (UTC)

Suggest articles from interwikis seems to be broken
Hi Magnus, I believe that you are the one that is hosting this old tool of Joanjoc's. It doesn't seem to work anymore, though (e.g. plug in "Ecuador" as the category for es.wiki articles missing from en.wiki, and nothing comes up even though es:Usuarios de Internet en Ecuador doesn't exist in en.wiki). Any ideas about what broke this tool and how it can be fixed? Calliopejen1 (talk) 19:33, 25 April 2012 (UTC)

Whether our articles should link to corresponding ones at Encyclopedia of Life
Please see WP:COIN. Thanks, EdJohnston (talk) 00:46, 28 April 2012 (UTC)

Meetup/Cambridge/15
Any chance you could confirm one way or the other?©Geni 15:28, 19 May 2012 (UTC)

Category:Gray's Anatomy images
Hello! Could I trouble you, please, to take a look at Category:Gray's Anatomy images and its contents, much of which you created in 2004? There is an ongoing discussion about the category, but is it not clear what was the intended function of the member pages or if they are still needed. Your comments could prove most helpful. Thank you, -- Black Falcon (talk) 23:52, 26 May 2012 (UTC)

BaGLAMa
Hi Magnus,

Sorry if this is not the right place! Could you please add http://commons.wikimedia.org/wiki/Category:Victoria_and_Albert_Museum, http://commons.wikimedia.org/wiki/Category:Freer_Gallery_of_Art and http://commons.wikimedia.org/wiki/Category:Bodleian_Library to the reports? Many thanks, Johnbod (talk) 14:40, 29 May 2012 (UTC)
 * Done. Will run the May set some time next week. --Magnus Manske (talk) 21:55, 29 May 2012 (UTC)


 * Hi Magnus, could you please also add https://commons.wikimedia.org/wiki/Category:Museu_d%27Art_Jaume_Morera this category to the report? We have a Wikipedian in Residence in this museum. Thanks in advance!--Kippelboy (talk) 14:13, 5 August 2013 (UTC)

Integrating xISBN and xOCLCnum into Special:BookSources
Hi Magnus, I am the Wikipedian in Residence at OCLC and Liam Wyatt suggested I talk to you about my ISBN Idea. It's to use our service xISBN (which returns a list of related ISBNs on input of single ISBN) in Special:BookSources. I've started a main discussion at Wikipedia_talk:Book_sources. Thanks for your input. Maximiliankleinoclc (talk) 17:45, 18 June 2012 (UTC)

Need a tool quantify longevity of edits
Hey Magnus. Looks like you have developed quite a good toolbox. I would like to suggest a new one that can track the longevity of my edits as a means of doing self-appraisals, or to make comparisons with other editors. So could you develop a tool that looks at an editors article namespace edits and generate some sort of metric for their longevity? When I add content to WP I want to make sure it stays there (unless it needs updating). Is this feasible? Maybe it could be like the duplication detector tool I have seen used to check for copy vios. -- Alan Liefting (talk - contribs) 23:51, 25 June 2012 (UTC)

FYI
You have been mentioned at Wikipedia talk:WikiProject Opera... you may or may not want to comment.Moxy (talk) 23:01, 12 July 2012 (UTC)

Speedy deletion nomination of Tsunami


Hello, and welcome to Wikipedia. Although everyone is welcome to contribute to Wikipedia, introducing inappropriate pages, such as Tsunami, is not in accordance with our policies. If you would like to experiment, please use the sandbox.

If you think that the page was nominated in error, contest the nomination by clicking on the button labelled "Click here to contest this speedy deletion" in the speedy deletion tag. Doing so will take you to the talk page where you can explain why you believe the page should not be deleted. You can also visit the page's talk page directly to give your reasons, but be aware that once a page is tagged for speedy deletion, it may be removed without delay. Please do not remove the speedy deletion tag yourself, but do not hesitate to add information that is consistent with Wikipedia's policies and guidelines. Wgracen (talk) 09:46, 19 July 2012 (UTC)

Can you help?
Hi, can you help me? This image is needed for the Dana Delany article. However it is the wrong license? Is there anyway you can persuade the user to change the license, or upload another image of her? Much appreciated, thanks :) Put a talkback on my page when you reply, thank you very much! — M.Mario  (T/C) 17:36, 3 August 2012 (UTC)
 * Seriously now? --Magnus Manske (talk) 15:33, 8 August 2012 (UTC)
 * Sorry, I can see you dont have time - dont worry. — M.Mario  (T/C) 16:00, 8 August 2012 (UTC)

CatScan problem?
On any query through CatScan, I get the error: MYSQL error : The MySQL server is running with the --read-only option so it cannot execute this statement [INSERT IGNORE INTO cat2 ( catname ) VALUES ( "" )] It started approximately yesterday noon. Is it a known issue? Thanks, Ynhockey (Talk) 10:29, 16 August 2012 (UTC)
 * OK I see it works now :) whether you did anything or not, thank you for the great tool! —Ynhockey (Talk) 12:32, 16 August 2012 (UTC)
 * Toolserver trouble; didn't break it, didn't fix it either ;-) --Magnus Manske (talk) 12:45, 16 August 2012 (UTC)
 * Thanks. I suppose there's some problem with the Commons DB now, but in case it's something you can fix/influence, just letting you know:

MYSQL error : Access denied for user 'magnus'@'%.toolserver.org' to database 'commonswiki_p' [CREATE TEMPORARY TABLE cat1 ( catgroup INTEGER, catname VARCHAR(255) , depth INTEGER , PRIMARY KEY ca ( catgroup , catname ) , INDEX cb ( catgroup , depth ) ) STORAGE MEMORY]
 * FYI, your tool is being used as the basis for a couple important tools for Wiki Loves Monuments. Cheers, Ynhockey (Talk) 08:54, 18 September 2012 (UTC)
 * No pressure then :-) But, works for me. --Magnus Manske (talk) 09:00, 18 September 2012 (UTC)
 * UPDATE: Toolserver database foobar this morning, might not work for some wikis. Upstream problem. --Magnus Manske (talk) 16:15, 18 September 2012 (UTC)

Small present

 * Thank you :-) --Magnus Manske (talk) 07:39, 23 August 2012 (UTC)

Invitation to join the Ten Year Society
Dear Magnus,

I'd like to extend a cordial invitation to you to join the Ten Year Society, an informal group for editors who've been participating in the Wikipedia project for ten years or more.

Best regards, &mdash; Hex    (❝ ?!  ❞)   11:12, 1 October 2012 (UTC).

Use of historical images
Hi Magnus, Good to meet you again last Sunday. As we discussed image copyright, I wonder if you could tell me what guidelines I can use to decide whether a historical image can be uploaded? I understand it's 70 years after the creator's death, but he/she is usually unknown. Is it fair to assume that because the oldest living person is 122 years old and noone has been verified to live longer than that, everything before 1820 (2012 - 122 - 70) is definitely fine? If the creator is unknown, what other criteria can be used? I'm specifically thinking of uploading some paintings from http://www.penangmuseum.gov.my/museum/en/museum/historical-paintings. Thanks, cm&#610;&#671;ee&#9742;&#9993; 14:04, 5 November 2012 (UTC)
 * Hi cmglee, this sets a general limit of 70 years after the death of the author, and this a Malaysia-specific limit of 50 years after the death of the author. I would assume the latter applies. IANAL, YMMV. --Magnus Manske (talk) 11:45, 6 November 2012 (UTC)
 * Thanks, Magnus. I've used a conservative 1819 as a cut-off date and uploaded the images to http://commons.wikimedia.org/wiki/Category:Historical_images_of_Penang . cm&#610;&#671;ee&#9742;&#9993; 22:01, 8 November 2012 (UTC)

"Suggest articles from interwiki links" failing
I've had some problems using your clonned version of the Joanjoc tools lately, is it a server problem or a problem with the code? And could you make it work with Wikisource? Thanks! --Micru (talk) 20:08, 12 November 2012 (UTC)
 * Toolserver database reshuffling, should sort itself out soon. --Magnus Manske (talk) 20:19, 12 November 2012 (UTC)

Hi Magnus! Sugart is failing again now, gives an SQL error which I don't know what it means. Is this because of wikidata? Will there be a new tool that does the same thing, or is this only a temporary glitch? Of all the tools I have ever seen this is one of the most crucial because no other tool I know of will produce the same list. Til Eulenspiegel /talk/ 17:25, 26 March 2013 (UTC)
 * Joanjoc tools used deprecated database server names, which are now turned off. Fixed, I hope. --Magnus Manske (talk) 23:23, 26 March 2013 (UTC)

A barnstar for you!
Thanks! --Magnus Manske (talk) 14:47, 18 November 2012 (UTC)

Catscan 2.0
Hi Magnus, I've recently encountered errors with CatScan trying to find members of a category which have a particular template: MYSQL error : Table 'enwiki_p.globalimagelinks' doesn't exist [SELECT /* SLOW_OK LIMIT:2000 NM */ /* CATSCAN2 */ pid,gil_wiki,gil_page_namespace_id,gil_page_namespace,gil_page_title FROM temp2,enwiki_p.globalimagelinks WHERE pns=6 AND gil_to=ptitle] Is toolserver shuffling things around? I know I should learn how to report things properly with JIRA... sorry Dsp13 (talk) 01:56, 23 November 2012 (UTC)
 * Toolserver issue. --Magnus Manske (talk) 18:34, 23 November 2012 (UTC)

Krzysztof Penderecki
Hi Magnus, I reverted the infobox on the article, no slight on you, but its been discussed to death re composers, and the feelings seems to be that they are not needed or optimum for these articles. Hope your not offended. Ceoil (talk) 02:24, 23 November 2012 (UTC)

World geograph
Thought this might interest you. Please inform the others who were involved in our discussion a while back of this proposal. I think even if a merger with geograph is not possible we should fund this. Blofeld Dr. ( talk ) 21:51, 27 November 2012 (UTC)

Authority control tool
It's wonderful! I thought I should give you heads up that I've mentioned it at Wikipedia talk:Authority control. I'm going to file another JIRA ticket for the problem in positioning I mention there - the question is where the template should go in relation to the persondata template on en.wikipedia, and people may have strong preferences. (Until that's resolved, I won't be using it on wikipedia, but I'm loving it on wikisource!) Dsp13 (talk) 21:47, 17 January 2013 (UTC)

Wikivoyage - CatScan 2.0?
Hello Magnus. I don't know how to put an issue in the JIRA System, so i try here, sorry. The Wikivoyage is new as a WikiMedia Project - and i think they would be happy to use some of your tools. Perhaps it is only little coding required (just one more projectname and adress)? The most important would be CatScan 2.0. It is required to be able to use the categories rational. What do you think? --NatiSythen (talk) 07:01, 1 February 2013 (UTC)
 * CatScan now works with wikivoyage. I don't have a centralised list of projects, so I'll have to enable tools one-by-one. --Magnus Manske (talk) 09:16, 1 February 2013 (UTC)
 * Hey, thx a lot, that was a quick reaction!
 * Do we have to wait for any background work until it works fine? There is a big different between CatScan 2.0 and the real Kategorie. --NatiSythen (talk) 11:14, 1 February 2013 (UTC)
 * There seems to be a serious problem with the toolserver database mirror - the pages aren't actually there (though the category connections are!). I'll chase it up. --Magnus Manske (talk) 11:36, 1 February 2013 (UTC)
 * Do you get any new facts? --NatiSythen (talk) 20:29, 2 February 2013 (UTC)
 * Nope. --Magnus Manske (talk) 20:41, 2 February 2013 (UTC)

Flickr2Commons Attribution
For flickr2commons, could you an option for attribution within the CC template such as. Some of the photo-streams require a certain attribution to be present. Great tool btw!Smallman12q (talk) 15:13, 1 February 2013 (UTC)

Article
On another note, I just realized there isn't an article about you here! :( I've done some digging and I think I have sufficient sources to put something together. Any opposition from your side on fixing the redirect at Magnus Manske ?--Obi-Wan Kenobi (talk) 23:06, 28 May 2013 (UTC)
 * Not opposed, however, someone tried to write one on the German Wikipedia, and I was deemed not noteworthy ;-) --Magnus Manske (talk) 12:39, 29 May 2013 (UTC)
 * Hmm... but your bio is still there, right? I think I can find enough sources - your name is mentioned in a number of books and articles (for good reason). I could also try an auto-translate of the german article which seems a good start. I just think (if you're ok) that you should have a bio here - you've certainly earned notability in my book! :) --Obi-Wan Kenobi (talk) 13:17, 29 May 2013 (UTC)
 * Ok, done: User:Obiwankenobi/sandbox/Magnus - take a look and let me know what you think, esp if there are errors. I found a number of sources where you are referenced by name. --Obi-Wan Kenobi (talk) 22:15, 4 June 2013 (UTC)
 * You've been upgraded to Magnus Manske now. Take a look!--Obi-Wan Kenobi (talk) 19:56, 21 June 2013 (UTC)
 * Yay! Thanks, looks great!! --Magnus Manske (talk) 12:52, 22 June 2013 (UTC)
 * Already translated into French. Perhaps you should ping your friends at German-wikipedia, ask them if the new sources I put together would suffice to have you in mainspace there.--Obi-Wan Kenobi (talk) 03:29, 23 June 2013 (UTC)
 * I've added some material from German sources after Obi's call for help here. Please have a quick look, and if there are any errors, please let me know. Best, Andreas JN 466 05:39, 25 June 2013 (UTC)

Re: category_intersection.php
Hello Magnus, yes, I suppose it would be fine. As usual, redirects are always useful, even if working only partially. --Nemo 17:04, 2 June 2013 (UTC)

Question about WikiMiniAtlas and Scribunto
Hi Magnus,

Someone sent me your name about help for the Scribunto extension and wikiminiatlas. I am in Canada and I want to add geographic coordinates to my wiki page so that it shows geo-coordinates like the top right corner of this wikipedia page: http://en.wikipedia.org/wiki/One_Wall_Centre but I want my locations to be specific locations from my own wiki page not from Wikipedia, is this possible? My wiki site is not connected to wikipedia and I am using mediawiki 1.20.X.

I added the wikiminiatlas code from (http://meta.wikimedia.org/wiki/MediaWiki:Wikiminiatlas.js) and pasted it into my MediaWiki:Common.js page. I then enabled js by using $wgUseSiteJs = true; and then tried this coordinate on one of our wiki page, { { #coordinates:40.744924|-74.006943 } } but I don't see a globe.

Do I need to enable Scribunto extension along with the wikiminiatlas .js file before this will work? Can you provide the steps about how to do this?? I did a lot of online searches and asked people on mw support and also mediawiki irc chat but I'm not getting any good answers and I hope you can help. Thanks for your help.

Peter in Canada — Preceding unsigned comment added by Backto1992 (talk • contribs) 06:13, 20 June 2013 (UTC)
 * Hi, I'm not really familiar with either WikiMiniAtlas or Scribunto, but if you want to work, it appears you need to install the GeoData extension on your wiki. Hope that helps. --Magnus Manske (talk) 09:24, 20 June 2013 (UTC)

Deep out-of-sight tool
Hello. I've noticed this tool doesn't work on pl.wikipedia (for example: ). Are there any problems with server ? Thanks, Sir Lothar (talk) 11:35, 25 June 2013 (UTC)
 * Thanks, fixed now. --Magnus Manske (talk) 11:59, 25 June 2013 (UTC)

A barnstar for you!

 * Thanks, made my day! --Magnus Manske (talk) 14:58, 28 July 2013 (UTC)

Invitation


As there is a Wikipedia article about you, you are cordially invited to contribute a short audio recoding of your spoken voice, so that our readers may know what you sound like and how you pronounce your name. Details of how to do so, and examples, are at Voice intro project. You can ask for help or clarification on the project talk page, or my talk page. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 22:12, 4 August 2013 (UTC)

Add colon before File (and Category) links in wikitext output from the Category Intersection tool
It would be nice if a colon was added to links to File and Category entries in the wikitext output of the Category Intersection tool, as otherwise the page gets the images displayed (or is included in the category) which is not usually what is intended. Thanks for your many years of service of Wikimedia, and particularly for these tools. JesseW, the juggling janitor 08:12, 5 August 2013 (UTC)
 * Could you link to an example of this issue please? --Magnus Manske (talk) 10:28, 5 August 2013 (UTC)

GLAMorous - MySQL client ran out of memory
I tried to use GLAMorous on a user with a lot of images (>200.000). This does not work and I got: There was an error running the query [MySQL client ran out of memory]. It is not possible to check big collections like this? --91.221.58.5 (talk) 10:11, 7 August 2013 (UTC)

OSM
Hi. I started a list at mw:Talk:Wikimedia Foundation Design/Patterns and components, which mentions the OpenStreetMap gadget. I see at Gadget/proposals/Archive_3 that it was potentially going to be added here, but it doesn't seem to be available? Possibly you could point me towards further information about its usage here and elsewhere? (I haven't investigated at all, yet, but possibly we could (or already have?) consider merging the WikiMiniAtlas and this gadget?)

Your thoughts or additions at the first link, would also be most welcome. :) –Quiddity (talk) 20:14, 10 August 2013 (UTC)

Notice
Hello. There is currently a discussion at Administrators' noticeboard/Incidents regarding an issue with which you may have been involved. The thread is Suspicious/problematic behavior of Moheen Reeyad, TilottamaTitlee, and Leelabratee. Thank you. Jackmcbarn (talk) 16:10, 14 August 2013 (UTC)

DNB ratios upgrade
The categories of templates we were talking about this afternoon were Category:Attribution templates and Category:Article message templates (all subcategories to be included). Charles Matthews (talk) 20:02, 17 August 2013 (UTC)

CamelCase and UseModWiki
Hey Magnus

Some editors have been asking me some WP history questions, and I don't trust my own memory. Can you remember when we actually got rid of CamelCase? I recall you were making mods to UseModWiki during the latter part of 2001, but my memory is too fuddled to recall if CamelCase was gone before we made the PhaseII switch in January 2002. Can you remember?

Hope you are well, cheers Manning (talk) 05:30, 19 August 2013 (UTC)
 * Hi Manning, I know PhaseII had an option for CamelCase, but IIRC I turned it off on behalf of Larry Sanger, as I did with automatic subpage linking in main namespace. For the switch to work, CamelCase would have to have been removed before January 2011. Indeed this seems to be the case. My best guess is that we used CamelCase for about half a year in 2001 before we abandoned it. Mailing lists might contain the exact decision date. --Magnus Manske (talk) 09:17, 19 August 2013 (UTC)


 * Yay, my memory isn't as bad as I feared. I started in early Sept 2001 and we still had CC when I arrived, but I was certain that we got rid of it before PhaseII was turned on. Many thanks - Manning (talk) 09:26, 19 August 2013 (UTC)

FIST problem
Hi Magnus,

I would normally heed your notice asking people to post bug reports about your tools to JIRA, but from the 13 open reports for FIST there, the latest was modified over a year ago, so I hope you don't mind me posting this here. I am trying to do a search using the tool but I get these errors: Warning: mysqli::mysqli: (HY000/2005): Unknown MySQL server host 'be-x-oldwiki.labsdb' (0) in /data/project/magnustools/public_html/php/common.php on line 85 Fatal error: Call to a member function query on a non-object in /data/project/fist/public_html/fist.php on line 930 Can you take a look? --Waldir talk 22:06, 26 August 2013 (UTC)

Treeviews tool and adding Blake Archive to BaGlama
Hey Magnus. I have been working on creating metrics for WP:Blake. I have been trying to use Treeviews to track whats going on in Category:William Blake, but everytime I try to run it, it doesn't create a report: it just keeps loading and loading without results. Is this a common problem? Also, would you be able to add Commons:Category:Media from the William Blake Archive to BaGlama? Sadads (talk) 05:14, 4 September 2013 (UTC)
 * Nevermind on the Treeviews, realized I was using the toolserver one instead of http://tools.wmflabs.org/glamtools/treeviews/, Sadads (talk) 05:22, 4 September 2013 (UTC)

CatScan V2.0 bug?
I'm sorry if I'm posting this in a wrong place, JIRA apparently doesn't cover CatScan...

In CatScan V2.0 if I enter e.g.:
 * Depth: 5
 * Has any of these templates: WikiProject Croatia, Use talk pages instead: Check
 * Max age: 500, Only pages created during the above time window (overrides "lastest revision"): Check

I get:
 * MYSQL error : Column 'page_id' in where clause is ambiguous [INSERT INTO /* SLOW_OK LIMIT:2000 NM */ /* CATSCAN2 */ temp ( pid,ptitle,pns,ptouched,prev,plen,pgroup) SELECT /* SLOW_OK LIMIT:2000 NM */ /* CATSCAN2 */ DISTINCT p0.page_id,p0.page_title,p0.page_namespace,p0.page_touched,p0.page_latest,p0.page_len,0 FROM enwiki_p.page p0,enwiki_p.page p1,enwiki_p.templatelinks ,enwiki_p.revision WHERE tl_title="WikiProject_Croatia" AND tl_namespace=10 AND p0.page_title=p1.page_title AND p0.page_namespace+1=p1.page_namespace AND p1.page_id=tl_from AND p0.page_namespace IN (0) AND rev_timestamp >= "20130815154601" AND rev_page = page_id AND rev_parent_id = 0]

Should be easy to fix, hopefully... GregorB (talk) 11:55, 5 September 2013 (UTC)
 * Try it now. --Magnus Manske (talk) 12:53, 5 September 2013 (UTC)
 * Works fine, thanks! I see now that Depth was not needed since categories were not set, so this was a matter of incompatible parameters, rather than an actual bug. GregorB (talk) 09:31, 6 September 2013 (UTC)

Category intersections
Hi Magnus. I think Andreas may have mentioned a proposal I have made here: Wikipedia_talk:Category_intersection. Could you just take a look at the proposal and share your thoughts? I'm not thinking of doing this for all facets - just for a certain section of the tree - around biographies - the point would be to get rid of gender/sexuality/etc sub-cats, and use category intersection instead. Your tool is at the heart of this proposal, so I would really value your feedback, advice, and help if possible. Cheers, --Obi-Wan Kenobi (talk) 00:13, 4 May 2013 (UTC)
 * Hi, as I explained to Andreas, I very much like the idea of "flattening" the category hierarchy, but there are technical issues. One is that tag intersections perform not well in databases. Decompiling "African-American women poets" into tags "African-American", "women", and "poets", means that the query might (worst-case scenario) have to look at all women, check if they are African-American, and check if those remaining are poets. For this, the database will have to temporarily create a list of all women, then remove those not matching the other criteria. For every single query. That can eat compute resources at an alarming rate. Worse, as long as we have any kind of category hierarchy, there will be a need for category tree intersection. Trees are structures that databases are traditionally bad at, and it is why my CatScan can take several minutes for a query, or run out of memory (these temporary lists I mentioned).
 * Good news is, I think I have solved this for WikiData, and Wikimedia Germany actually paid someone to develop code to solve it for Wikipedia categories. Unfortunately, the latter code is not used anywhere, and WikiData is not quite comprehensive enough yet to take over. FWIW, I will port both CatScan and my WikiData "intersection" tool over to wmflabs as soon as possible. --Magnus Manske (talk) 12:22, 5 May 2013 (UTC)
 * Hi Magnus. That is great news, and thanks for the feedback. A few follow-ups, if you don't mind:
 * 1) Would it help trim the search space if we didn't disaggregate "nationality"? e.g. +  +  instead of  +  + ? This would be a less-drastic change to the categories in general (it would preserve most of them). I see your point about faceting every article fully as a potentially desireable end-state, but I was thinking more of a band-aid for now, just on the ethnic/gendered categories, because that is what causes the most clutter and conflict.
 * 2) I was thinking of creating a template which calls out to your tool, so editors don't have to muck around with URLs. If you move to labs, is it fair to assume we can just point the template to a new URL?
 * 3) Would you have any time or inclination to make a few UI tweaks? - I was thinking we could pass a parameter in the URL that makes you give us a much simpler result, formatted a bit more nicely and with most of the parameters/entry fields hidden - the goal would be to make the transition from a wiki-category page to the tool results feel a bit more smooth. I'd be happy to gin up some HTML to show you what I mean if that would help.
 * 4) If we add 'wikidata' facets to an article, is it possible to do intersection yet based on that, rather than based on categories? Or could one do an intersection of wikidata facets along with categories? I'm not sure if that's the german code you mentioned.
 * 5) Would it help optimize if we told you in advance, in the url, which was the largest category? our template for example could always pass in the "Nationality + gender" as the first facet, as that would be the largest - if you know that in advance, perhaps your tool can be faster/smarter? Just wondering...
 * Thanks again - for this tool, and for all the great work you've done in general here. --Obi-Wan Kenobi (talk) 14:36, 5 May 2013 (UTC)


 * Well, I actually have a simpler tool for that, but now I keep getting "gateway error" from the toolserver... Anyway, I ran a little test, a subsection of "American_educators" and "African-American_people", and that took ~3 seconds on the database level, coming up with Mary_Jane_Patterson, Priscilla_Ford, and Clarence_M._Pendleton,_Jr.. So if the categories are "flat" (no sub-categories to search!), and if wmflabs ever gets around to giving us a database mirror, it might be realistic. But remember, all I'm offering is the tech, nothing more... ;-) so once it's up and running, you need to convince people that this is the way forward! --Magnus Manske (talk) 17:32, 5 May 2013 (UTC)
 * Ok - I think the end result I was thinking of is, the highest level national/gendered and ethnic and religious categories would be mostly flat - e.g., , , etc. However, that seems nice in theory, but even those facets are likely to be subdivided - e.g. which would be sub-catted to  and so on. But in any case, those particular trees would *not* be that deep. On the other hand, the content categories - say the people by occupation - could get a lot deeper. For example, today under , we have at least 5 sub categories deep, and I'm not proposing to change that at this point - so we'd still need the ability to specify levels of recursion, like you have on the current tool.--Obi-Wan Kenobi (talk) 18:21, 5 May 2013 (UTC)

Hi Magnus. I think, besides porting the tool over to labs and (if possible) making it a bit faster, one of the things that will make this much more palatable to the wiki-audience would be a little bit of UI changes. I've been thinking of the following: Ok, so that's basically my dream feature list... I hope you don't mind me asking - would that be possible to do any of those? I've made a revised prototype here: - note how you can easily now look at arbitrary intersections. I think this could potentially work at scale, esp for the ethnic/gender categories, which multiply like mushrooms...
 * 1) I pass in a parameter in the URL called simplify or something, so you know to display the UI differently than standard.
 * 2) The results are on top, instead of the bottom (today users have to scroll) - so the whole data-entry box goes to the bottom
 * 3) Instead of being presented as a flat list with lots of additional data for each item, instead the results are presented in a screen, 200 at a time in 3 columns, with just the article names - the idea is to use the same font/size/layout as wikipedia - so it doesn't feel like you're going to a completely different place.
 * 4) Even better, if possible, would be, given the presence of an additional parameter, to sort the list by last name instead of first name (or if you want to get really crazy, based on the sort-key in the article itself)
 * 5) Not sure if this is possible, but could you present the first 200 articles once you have them in sorted order, and then asynchronously page out the rest on the server, so the user doesn't have to wait for a large # of articles to load?

cheers and thanks again, --Obi-Wan Kenobi (talk) 23:30, 6 May 2013 (UTC)
 * Sorry, one more random question - are the results based on live data, or based on a cache/copy/etc of the wikipedia database? Thanks! --Obi-Wan Kenobi (talk) 23:39, 6 May 2013 (UTC)


 * Hi Obiwankenobi, the results are "almost live" - the toolserver database mirrors lag behind, sometimes fractions of a second, sometimes hours or days (this will be better on Labs). I was actually thinking about a new, more specialised tool. CatScan2 is very generic, and thus takes >6 seconds pure query time for the "Singaporean women poets" intersection. I would imagine something more optimised, that "only" delivers data (an API), and the results are shown right on Wikipedia via JavaScript, so users don't have to leave the site. --Magnus Manske (talk) 17:49, 7 May 2013 (UTC)
 * Great! That would be even better from a UI perspective - I didn't know it was possible... (the reason I framed this approach as a hack is I was trying to find something that could be done even if wikipedia itself doesn't have to change)
 * If you can get inline results, there are really two main uses cases:
 * Category enumeration - if you look at, I added a link to the top to fully enumerate the category (using catscan), since people are complaining that they want a list of all novelists, so now they can really get one. So maybe you could write something specialized which can do just that, very fast - so someone could go to and say "show me all members of this cat as well as all members recursively" - (we'd probably need to put a max recursion depth on each page that has this however), then they wait a few seconds, and they have their results they can page through as before.
 * Category intersection - as discussed above. The simple solution would be to build a template, which editors can specific which categories they want intersected, and how many levels of recursion to do for each category. A more sophisticated version would allow users to check boxes/use drop-downs to select an arbitrary intersection at run-time. But that could be version 2.0, I think version 1.0 with a template that gets customized by editors as a header placed on top of each category that wants intersects (sort of like I hardcoded at Singporean poets) would be a great start.
 * A Question - I've reached out to several wikimedia people for help from their devs on this (say with writing bots, or templates, etc), but they've told me to hit the mailing list, or to apply for an individual grant. I'm not sure if it would help you, if we could apply for a grant and then you could use those funds to do some of this customization? If so, I'm happy to help draft the grant application. I'm not sure how much time/energy/effort it would take to put together what you're proposing above.--Obi-Wan Kenobi (talk) 20:04, 7 May 2013 (UTC)
 * Friendly ping... any thoughts on the above? cheers. --Obi-Wan Kenobi (talk) 16:09, 15 May 2013 (UTC)
 * Hi, I was (and still am) busy in RL, and next on my list "program off-work" is some wikidata work. That said, a few short answers. Funding would not help me, as I do have a job, and do these things for free anyway, but thanks for the offer. As said before, the real killer is the "depth" search; without that, the intersection would be "not too bad". I'll try something in the next couplea days, but the toolserver is in worse shape than ever, Labs does not offer database access yet, and using the API in the browser would flood the servers with queries and bring the browser to a grinding halt. If we can get this to work, Labs is our best mid-term bet. --Magnus Manske (talk) 15:13, 16 May 2013 (UTC)
 * Ok - of course I understand, we are all basically volunteers here :) When you say "depth" search, do you mean recursing over a tree like, which itself has many sub-cats? Is it a question of depth of recursion, or total number of articles, in terms of performance? Also, do you agree that if we tried this, we should perhaps force a national-level categorization of people (i.e. intersect "American men" + "American novelists" instead of "American people" + "Men" + "Novelists")? The tree is already set up that way (by nationality), so it would be less disruptive to keep it that way - I just fear "Men" would be a pretty brutal category (though even would be massive)
 * works fine for German Wikipedia. It would be a lot cleaner in the conceptual sense, but won't make a big difference for the tech.
 * I agree, it would be conceptually cleaner - and if you say it wouldn't make a big diff for performance, then we might as well go all the way. This would mean removing a *lot* more categories however - but I suppose in the long run that's a good thing.--Obi-Wan Kenobi (talk) 14:25, 21 May 2013 (UTC)
 * As for labs, I assume this means it's a bit out of our hands for now? Do we have to wait for labs to bring database access online - then you can port over and we'll get better performance over there? Thanks again and sorry if I ask stupid questions, I'm an ex-(recovering)-programmer but don't know anything about how wiki actually works under the hood.--Obi-Wan Kenobi (talk) 15:28, 16 May 2013 (UTC)
 * That about sums it up. They are working on it, and we might see first results this month. I expect to see all databases available on Labs in June, but that's more of a personal guess than based on deep insight. --Magnus Manske (talk) 11:50, 21 May 2013 (UTC)
 * Ok, please keep me posted, and let me know what more I can do from my side to facilitate and move this forward.--Obi-Wan Kenobi (talk) 14:25, 21 May 2013 (UTC)

new version of category intersections
Victory! American writers &times; American men, tree intersection to depth 9, 8.5 seconds, on WMFLabs! --Magnus Manske (talk) 09:14, 28 May 2013 (UTC)
 * awesome- thanks. That's a very deep recurse. I just ran same thing but it took 24 seconds - but still decent. Will labs always be faster than tool server? Ok I will change the prototype to point to the new URL. Let me know what you think about the other FRs above... --Obi-Wan Kenobi (talk) 11:39, 28 May 2013 (UTC)
 * Hah! Now try this! --Magnus Manske (talk) 13:35, 28 May 2013 (UTC)
 * whoah that's fast! Does it recurse? It seems to - but how deep? Does your recursion detect loops? What needs to be done to embed that into Wikipedia interface?
 * Recurses full depth; avoids loops; will try JavaScript for Wikipedia interface tonight :-) --Magnus Manske (talk) 14:00, 28 May 2013 (UTC)
 * Damn - it is so fast. What did you do differently? it's night and day. Seriously, this is bad-ass. --Obi-Wan Kenobi (talk) 14:29, 28 May 2013 (UTC)
 * The toolserver would have died under the query I am using here. So I had to hack around it. Now I can do what I always wanted to do. Now I am the master, Obi-Wan. Oh wait... --Magnus Manske (talk) 14:44, 28 May 2013 (UTC)
 * Only a master of evil, Darth.... You can't win, Darth. If you strike me down, I shall become more powerful than you could possibly imagine. ... But seriously, this is now fast enough to make it all worthwhile. A few UI tweaks and I think we could roll this out broadly. I assume labs is more stable than toolserver and will have better uptime? --Obi-Wan Kenobi (talk) 14:56, 28 May 2013 (UTC)
 * Good work, Magnus. The difference in speed is absolutely amazing. Andreas JN 466 01:28, 29 May 2013 (UTC)
 * JavaScript is live. Code here, then try the demo here. --Magnus Manske (talk) 14:15, 29 May 2013 (UTC)
 * Ok, got it. Looks great. I tried a triple intersect but it hung, so I stopped it - does that work? I think we will need to think of a double solution (1) something for those who haven't edited their javascript, random IP browsers, the general public (2) something faster, in-wiki, for those who have. Or, is it possible to have an in-wiki solution for everyone? Also, what are the chances of sorting based on the default-sort key? Otherwise it's snappy and sweet, I love it. I think you confirmed above, we don't need to give a recursion depth, it will just keep recursing? What happens if someone puts a top-level category, like intersecting Life with Society or something - will it die? kill the server? hang? churn for a while? I'm just concerned what strange things editors might try, and how the system will react. But it's awesome, I want to help roll this out, at least for the people categories and get rid of all of the gender/ethnic divides. --Obi-Wan Kenobi (talk) 14:48, 29 May 2013 (UTC)
 * Ok, I did a quadruple and it worked. not sure why the other one I tried hung. Anyway, this is great - how can I help? What are the next steps? --Obi-Wan Kenobi (talk) 15:00, 29 May 2013 (UTC)
 * Any number of categories should work; if you find a combination that doesn't work, tell me what it is. There could be safeguards (e.g. fixed max recursion depth at 15 or so) to prevent long queries. Otherwise, it appears that "the tech" is working for once. Of course, if we want to switch to this system, it needs to work for everyone out-of-the-box, which means it has to go into the default JavaScript. This is a battle I fought before, sometimes successfully, often in vain. I'll leave the lobbying for this in your capable hands! --Magnus Manske (talk) 21:29, 29 May 2013 (UTC)
 * Ah ok - but once you put something into the default javascript, will all users get it right away, or do they need to refresh their browser, etc? And who exactly do we need to convince? I've been tweaking the js, will send you some suggested changes in a bit.--Obi-Wan Kenobi (talk) 21:58, 29 May 2013 (UTC)

Ok, I've tweaked the js, you can get the new version here User:Obiwankenobi/intercat.js. Changes made:
 * I fixed so it links to categories correctly
 * Bit of a hack but it deals with namespaces now (it works for mainspace, userspace, and wikispace). I'm sure there is a more elegant way to deal with this.
 * Changed to use bullet points, 200, 3 columns (just like wikipedia standard)
 * Updated the header somewhat, and added an overview of total records returned, to match wikipedia.
 * I also tweaked so that when there is only one category, it lists it as an enumeration instead of an intersection.

My JS skills are pretty poor so feel free to rewrite whatever I did in a better way - but first try it out - add my JS (remove yours), and then try out the page here: Category:Wikipedia_categorization_intersect_test so you can see what it looks like on a real category page. I added some illustrative intersects for American novelists.

Some FRs for Magnus:
 * I'm grabbing the pagecount variable from the json but when you page 200-at-a-time it seems you don't fill that in with the total number of records - can you update the API to put the total number of records in pagecount even if you only page 200 at at time?
 * Could you make it so when you "open" one intersect on a page, it closes all the others - I couldn't figure out how to do this.
 * The way I handled namespaces is a bit of a hack. How do you suggest we improve? For example, could you change the API so we pass in a text namespace instead of a number? That way, the user could specify the namespace like, then we could pass "User" in, and use the same string as a prefix for article titles directly - that would clean up the JS and prevent need for logic to deal with converting.
 * We should see if we can set up the JS so that users can easily click/select different characteristics to build a string, e.g. gender, ethnicity, sexuality, etc - instead of us having to enumerate all of them, which as you can see is crazy, and I only listed 3 ethnicities. Any other thoughts on how to do the display portion?

What really makes this magic and transformative is the speed. Getting people to wait 10 seconds was never going to work, so whatever magic you did, nice job, it is so speedy I can't believe it. --Obi-Wan Kenobi (talk) 00:21, 30 May 2013 (UTC)
 * Hi Magnus - any thoughts on the above? I think perhaps the biggest one missing now is sorting of results via undefined, and grouping-by-letter - and eventually having templates to skip to a new letter like we allow now. If you like, I can try to find someone else to help with refining the javascript to render the grouping-by-letter, but we still need the server side to do the sorting I think.--Obi-Wan Kenobi (talk) 19:59, 21 June 2013 (UTC)


 * Obi, have you thought about writing an op-ed for the Signpost in preparation for this, explaining the difficulties with the present categorisation system, how unstable and opaque it is, the confusion over diffusing and non-diffusing categories, and how most casual users simply get it wrong (and always will)? You could present a rationale to the community why intersections do a more efficient and flexible job, enabling users to search for whatever intersection they want, while preventing mis-categorisation, ghettoisation by gender, race, etc., arguments over "tagging" biographies etc. I'm sure you could present the case well that way. Andreas JN 466 15:51, 30 May 2013 (UTC)
 * Sure, it's a good idea, we could write it together. Just perhaps not quite yet. I'd like to be sure this is going to work. The main hurdle now seems to be ensuring that the javascript can be added to the default. Otherwise we need a different solution. Did you look at the test here? you need to add the javascript file User:Obiwankenobi/intercat.js to your commons.js file to see how it works. We need some help in figuring out what the ultimate UI for this should look like. I'm a bit worried about the possibilities - for example, do we list every possible intersection of ethnicity + gender + sexuality? we *could* but is that a good idea? I took a first shot at it on the test category page, but feel free to modify and add your own versions.--Obi-Wan Kenobi (talk) 16:14, 30 May 2013 (UTC)

Category intersection - final tweaks
Hi Magnus. I see you've been continuing to improve the tools - that is awesome. I just wanted to centralize my high-priority feature requests here, so you can let me know which ones you might have time to do. I can also look into getting some other volunteer assistance, for example on the javascript side - let me know if that's worth doing.
 * 1) Sorting of results via the Default sort values. I think this one is crucial, and needs to be done server side.
 * 2) When you page 200-at-a-time it seems you don't fill in the total number of records - it just says "200". Can you update the API to put the total number of records in pagecount, even if you only page 200 at at time?
 * 3) I'd love to be able to group the results by letter as well (in the same way default sort does) to match the current category pages. Is seems the solution may be a combination of how the API returns things and then how it's rendered in the JS, but I'm not sure on the best way to do this. Should I seek assistance from a WMF JS developer, to replicate what they do on category pages?
 * 4) Sub-categories. I've just realized that we only return articles - what happens if there are categories that are also part of an intersection? (this may happen with eponymous categories, for example). Any thoughts on what to do here welcome.

Hope you're having a lovely summer. If you're ever in NYC let me know, I'd love to buy you a beer. :) --Obi-Wan Kenobi (talk) 19:38, 26 July 2013 (UTC)
 * Howdy... hope all is well. Any thoughts on the above? I think at least 1/2/3 are likely needed before we can go to primetime. --Obi-Wan Kenobi (talk) 23:00, 6 September 2013 (UTC)
 * OK, results are now sorted by sortkey. There is an optional grouping of results by first letter of the sortkey. You can get the total number of possible results as well; you can set the max=0 to just get the number, without actual results. See the tool for the options. I have taken the liberty of editing your JS to use "sparse mode", which only returns the namespace-prefixed page title to save time. I also changed it to load/display letter groupings. As four your #4, currently all queries are namespace-specific. You can use "ns=14" to get subcategories. It would be possible to allow multiple namespaces, but I don't really see the need for it at the moment; subcats can/should be loaded in parallel, separate from pages. --Magnus Manske (talk) 13:29, 7 September 2013 (UTC)
 * Ah ok excellent - thanks so much this looks good. I will play with it and get back to you. Thanks again! --Obi-Wan Kenobi (talk) 15:05, 9 September 2013 (UTC)