User talk:Whatamidoing (WMF)

Old discussions: June 2013 to May 2014, June 2014 to May 2015, June 2015 to May 2016

List of HTTP bots on dewiki

Hi! Can please post a list like this for the German Wikipedia? -> here. Thx --Euku 09:15, 21 June 2016 (UTC)

This list includes bots and script-users for all wikis. Let me see if I can find the most current version... Whatamidoing (WMF) (talk) 18:10, 21 June 2016 (UTC)

User:Euku, the most recent list was posted yesterday at phab:T136674#2394147. w:de:User:EmausBot and w:de:User:Merlbot may be the most important for the German Wikipedia. Whatamidoing (WMF) (talk) 18:16, 21 June 2016 (UTC)

Publish

Hi Whatamidoing, I brought Save vs Publish back to VPT - I know you are involved in this and thought you might be able to share some of the foundation research that could be useful to the discussion. Thank you, — xaosflux ^Talk 02:35, 2 July 2016 (UTC)

Thanks for the note; I'll take a look. Unfortunately, most of the research isn't published, or the information is very brief. (On the other hand, what did we expect? It doesn't take much space to write "Half the users in this study, without any prompting by the interviewer, said that the 'Save page' button was confusing, and some volunteered the suggestion that it ought to be labeled 'Publish page' instead"? It's not like the subject is complex enough to need a dissertation.) Whatamidoing (WMF) (talk) 02:12, 3 July 2016 (UTC)

Thank you so much for your attention !

Hello, I reported a error in the visual editor but it turned out to be some momentary thing because in today's attempt the "upload image" section worked fine. I'm sorry if this isn't the right place to message you, I'm kind of Noob when it comes to Wikipedia's workflow :) — Preceding unsigned comment added by Hydrocat o (talk • contribs) 02:47, 14 July 2016 (UTC)

Hey Hydrocat o, thanks for your message and for caring to share with us! Best, --Elitre (WMF) (talk) 05:57, 14 July 2016 (UTC)

I'm happy to hear from you. This is a great way to reach me. Thanks for the update. If you see that glitch again, feel free to let us know.

Happy editing, Whatamidoing (WMF) (talk) 21:04, 16 July 2016 (UTC)

WP:AN

This message is being sent to inform you that there is currently a discussion at Wikipedia:Administrators' noticeboard regarding an issue with which you may have been involved. Thank you. Section is: Wikipedia:Administrators'_noticeboard#Content_translator_tool_creating_nonsense_pages. — xaosflux ^Talk 00:27, 27 July 2016 (UTC)

Visual editor only plays with cite book, slamming WP:CITE into oblivion

• My new goal in life is to get template developers to create three citoid flavors of cite book: cite book APA, cite book MLA and cite book Chicago templates... All three formats should be available as templates that interact with other aspects of Wikipedia in every way identical to cite book (but merely display differently) ... "But wait!" you're thinking, "that's controlled by volunteers, not by WMF!" No, in practice that is incorrect. Only cite book plays well with Visual editor, and visual editor is the default standard now. New editors get Visual editor, and bundled with VE they get cite book, and all our talk about WP:CITEVAR is solely and only empty talk.
• WMF (not the WP:OWNERS of cite book) is de facto enforcing a citation style, like it or not... thus WMF should be behind an effort to make all three of the citation formats used most widely in every corner of the English-speaking world other than Wikipedia available as templates that interact with other aspects of Wikipedia in every way identical to cite book (but merely display differently). Tks  Lingzhi ♦ (talk) 23:03, 17 August 2016 (UTC)
  • Hello, Lingzhi. Local users can change which templates are included in Citoid's list. You just need to update MediaWiki:Visualeditor-cite-tool-definition.json and MediaWiki:Citoid-template-type-map.json to include whichever templates are preferred locally. Also, as User:Maunus indicated, there's nothing in VisualEditor that forces anyone to use either the automated citation-filling system, or the templates that happen to be listed as most common, or even any templates at all. People seem to prefer automatic citations overall, but they don't have to use it.
  • On a practical front, you might try making friends with Trappist the monk, who AFAICT knows more about the English Wikipedia's main citation templates than anyone else. Maunus' idea of having a |style-guide=mla is probably the best way to go about it, if editors choose to create such a thing. (It's unclear to me whether they would choose that, especially given the /FAQ at the top of WT:CITE, which says "It is unusual for Wikipedia articles to strictly adhere to a formally published academic style.") Whatamidoing (WMF) (talk) 09:17, 18 August 2016 (UTC)
• I think it makes sense to have a single template for the sake of learnability - but it should be both possible and useful to have different display options within that single cite template. I don't think it is quite right that wikipedia articles adhere to a formally published academic style in so far as bibliography formatting goes - indeed as Lingzhi states the cite-book template pretty much enforces wikipedias "own" citation style (CS1) for bibliography items (which seems to mix APA and Chicago). I think having an option to follow a different styleguide would be excellent - but I also know that a lot of editors believe that WP:CITEVAR should be abolished and CS1 be made wikipedia's official citation format, so I think it would be an uphill struggle.·maunus · snunɐɯ· 09:27, 18 August 2016 (UTC)

  @Lingzhi, @·maunus: The door was opened to allow editors to specify identified styles when |mode= was introduced to cs1. Thus far no one to my knowledge has ever asked that cs1 render strict mla, apa, cms, bluebook, or any other of the defined styles. Here is not the place for such discussions so editors are encouraged to raise the issue at WT:CS1

  —Trappist the monk (talk) 11:25, 19 August 2016 (UTC)

  • @Trappist the monk: It's not a matter of VisualEditor forcing anyone to do anything; it's not a matter of forcing anyone to do anything (no one coould get away with doing thta, could they?) but rather of other options being left prohibitively difficult in terms of learning curves and lacking in terms of functionality. The end of the story is that WP has one template option. That would be "one" meaning "one". That would be CS1. As for VE, I have it turned off of course (as I believe Maunus has as well). I am not familiar with how it interacts with articles (and have no desire to become so). However, I seem to recall reading threads/discussions where editors added references using Visual Editor that ended up appearing in a different format than all the others, and they didn't see an obvious way to repair the issue, so they ended up making things worse. This apparently was due to VE's close interaction with CS1 templates and lack thereof with other options, including manual edits... If I take all this "One Format to Rule Them All" talk into any forum, there will be a !vote. But consensus seems to me to be a viable decision-making option only when choices are mutually exclusive. If 60% of all editors do not want new format options, but 40% do, do we then call "No Consensus for Change" and the 40% can go whistle in the park somewhere? How big must a minority be before they can be heard? Or even better, in the case where all we are considering are harmless additional optional formats, why even bother to have an iron rule of consensus, unless someone can clearly establish that the word "harmless" is not true? WP:CONSENSUS works very well for deciding what to do with the text of an article, where debates often center around mutually exclusive alternative directions, but in this case...  Lingzhi ♦ (talk) 11:52, 18 August 2016 (UTC)

Methinks you are talking to the wrong person. This is my first post ever on Editor Whatamidoing (WMF)'s talk page. I have had nothing to do with visual editor either as a user or a developer. I have not colluded with wmf to get them to use cs1|2 in preference to any other citation style. I do not understand why you are railing at me about consensus.

—Trappist the monk (talk) 12:41, 18 August 2016 (UTC)

Oops. I pinged you only to conclude you, because WhatamIdoing mentioned you as the cite book expert (see discussion above about formatting options). I wasn't railing at anyone (sorry if it appeared that way!). Sorry to have upset you needlessly (and accidentally).  Lingzhi ♦ (talk) 12:49, 18 August 2016 (UTC)

I think it might help if you had some idea how it works. You could find this out by switching to it and clicking buttons, but let me save you the trouble. Here's what you'll find in the "Cite" button:

See that thing in the "Manual" tab that's labeled "Basic form"? If you don't want a template, then you click on that. It gives you this (except bigger):

It's an empty box with the usual formatting options. You type your citation into it, just as if you were typing on the main page. Or type some explanatory text. Or hieroglyphics (seriously). It works just like a non-ref text area. If someone doesn't want to use a citation template, then they can still do that in the visual editor. Whatamidoing (WMF) (talk) 15:17, 18 August 2016 (UTC)

• Thanks. Perhaps I misunderstood what was said by others about VE. That still leaves all of Wikipedia with one and only one citation format, but I doubt that is anywhere on your "to do" list. Thanks for posting this reply.  Lingzhi ♦ (talk) 15:23, 18 August 2016 (UTC)

  Sorry, Ling, I'm not following this. Why does it leave us with only one format? What would happen if someone tried to use the format you think is now hard to use, in VE? Mike Christie (talk - contribs - library) 11:28, 19 August 2016 (UTC)

  I think you (WAID and Mike Christie) are missing Lingzhi's point here. Yes, one can manually override the citation formats, but put yourself in the shoes of a new editor who tries to add a reference and is faced with two buttons, one labeled "Automatic" and one labeled "Manual". Do you check the "Automatic" button, which gives you a single box marked "URL, DOI or PMID" and automatically creates a {{cite book}} etc from that, do you check the "Manual" button which gives you options marked "Website","Book", "News" and "Journal", each of which automatically creates a {{cite book}} etc from data you enter, or do you check "Manual", then select "Basic form", then select "General references", then enter the data from scratch in a markup language you probably don't understand? Given that presumably most new editors don't even realize that Wikipedia doesn't have a uniform citation style, this pretty much inevitably leads to a situation where new editors who are trying to do things by-the-book will add cite book/web/journal templates regardless of the existing style of the article, and be surprised and upset when they see other people change the citations (which as far as they're concerned, have been entered in "the way Wikipedia wants") and get chided for breaching WP:CITEVAR.

  In VE, unless you're following a route that leads to the output of a {{cite book}}, referencing is totally unintuitive unless you're already intimately familiar with wikitext (in which case you're probably not using VE; I assume even VE's staunchest defenders will concede that it's made very little headway into persuading the existing editor corps to adopt it). Realistically the "General references" button will rarely if ever be used, so the devs have unintentionally imposed a uniform citation style for new editors by default. This is not necessarily a bad thing, as it's probably preferable to the existing setup of good-faith newbies dumping raw URLs into articles, but we shouldn't try to pretend it's not happening, and (assuming the WMF aren't planning to radically change VE) at some point there will need to be a bitterly-fought discussion over whether existing articles need to have their referencing styles retrofitted to be compatible with VE's approach. ‑ Iridescent 18:56, 19 August 2016 (UTC)

  Thanks; that makes complete sense and explains what Ling was talking about. (Just FYI, I'm a multi-year editor who now edits solely in VE unless something forces me into wikitext; I agree I'm a fairly rare, though not unique, case.) As it happens I use short form citations so I use the Manual/Basic option almost exclusively, though I do use automatic for web citations. What's the mechanism behind the automatic conversion -- is that VE, or, as I would expect, a separate engine (Citoid)? Mike Christie (talk - contribs - library) 19:35, 19 August 2016 (UTC)

  You should probably click through to the visual editor and try it. It doesn't use wikitext markup. You don't have to add any code. If you want the text in italics, then you select it and click the item in the dropdown menu for character formatting. If you can type a bibliographic citation in Microsoft Word or Google Docs, then you will be able to figure out how to type a freeform citation in the visual editor's ref box. Whatamidoing (WMF) (talk) 21:06, 19 August 2016 (UTC)

  (Why do WMF-ers without exception assume that anyone criticizing them must be doing so because they don't understand whatever widget the WMF is trying to push, rather than that they've tried it and found it wanting?) Of course you don't have to add code in VE, just as you can create a perfectly valid citation with just a couple of <ref> tags in Wikitext, but if you're trying to add a remotely complicated citation in anything other than CS1 format in VE without using Wikitext you're going to be doing an awful lot of selecting and clicking. Open up VE and time how long it takes you to replicate this reference in Bluebook format without lapsing into Wikitext, for instance. ‑ Iridescent 21:22, 19 August 2016 (UTC)

  I just tried it; it took me about 90 seconds. I would have thought that was acceptable. I have to say it would have been a bit faster if the template documentation had been filled in; maybe not much over a minute. I am pretty sure that's faster than I would be able to create it in wikitext, though for someone who knows all the params it might be faster in wikitext. Mike Christie (talk - contribs - library) 21:45, 19 August 2016 (UTC)

  It just took me about 65 seconds to retype the citation from scratch in a text editor, copying by sight, and cutting and pasting the long text blocks such as the URL, which is what I did for the VE entry of the ref. That's with the existing citation in front of me so I don't have to figure out which params to use. I'd say VE is definitely faster for me for a template I don't have memorized, and I'd be surprised if it were significantly slower for anyone, given what I just tried (and assuming the template documentation is filled in). Mike Christie (talk - contribs - library) 22:23, 19 August 2016 (UTC)

  About 50 seconds in VisualEditor, pasting everything. Half of that time was spent adding the smallcaps. Plain text citations are faster than filling in citation templates, if you know what you want to type there. If the automatic system works for your URL, then that's quicker, of course, but the next-quickest option is frequently just adding the text by hand, sans templates. Whatamidoing (WMF) (talk) 09:34, 20 August 2016 (UTC)

(←) By saying refs within VE are faster you are arguing in favor of my point. I am not arguing aginst it at all (I dislike it, but that is purely a harmless personal pref). I am saying that it makes cite book templates the de fact standard across wikipedia, because we do not have |style-guide=mla, |style-guide=apa, and |style-guide=chicago. VE is the default editing environment forced upon first-time users; VE makes cite book easier to use; VE has no REAL citation formats (just a B.S. Wikipedia-only format that the template editors force us to employ, via lack of other options).  Lingzhi ♦ (talk) 00:17, 20 August 2016 (UTC)

I wasn't disagreeing with you; now Iridescent has clarified it I see your point. I was just responding to the comment about the slowness of reproducing a Bluebook citation. I would like to know the answer to the question I asked above, though; what is it that generates that citation from the automatic tab in VE -- is that Citoid? If it's parsing the data into some internal representation I can't believe it would be that hard to set a switch to output it in whatever format we want to specify. Mike Christie (talk - contribs - library) 02:04, 20 August 2016 (UTC)

Thanks for the clarification... as for your question, you'll have to ask WAID or Trappist or another relevant domain expert.  Lingzhi ♦ (talk) 09:28, 20 August 2016 (UTC)

Yes, it's the citoid service, and yes, it's just copying and pasting bits into the matching spot for a template. Both the template and the "matching spot" are defined locally, so any local admin can change them at will. The WMF devs do not write local templates or user scripts; the volunteer devs own those areas. Editors can also make a |mode=mla for {{Cite book}} whenever they want. (Only, as a practical matter, I'm going to guess that Lingzhi doesn't know how to edit those templates, and he seems to have accidentally annoyed one of the people who could do it for him. This is unfortunate, but hopefully not permanent.)

I'm objecting to the use of language about "forcing" people to use certain tools, especially when it's based upon these factual errors. So above we have the erroneous claim that anyone is forced to use cite book, despite the existence of freeform citations. Now we have the erroneous claim that the visual editor is "the default editing environment forced upon first-time users". This is not true; new editors at the English Wikipedia start in the wikitext editor, and at any given time, 80 to 90% of them are using the wikitext editor. If they want to use the visual editor, they have to already know that it exists, know how to switch, and manually choose to do so. And even if they choose to use it, then they're not "forced" to use {{cite book}}. In fact, very, very, very few of them use that template. Or even cite books at all, as discouraging as that sounds. I just checked the last 100 edits by newly registered editors in the visual editor, and exactly zero (0) of them used {{cite book}}. A few used {{cite news}} and {{cite web}}, a few used bare URLs, and most didn't add any citations at all. If you want {{cite book}} to be used less, then your problem is definitely not over-use of that template by new editors. Whatamidoing (WMF) (talk) 09:41, 20 August 2016 (UTC)

• It's not just cite book, it's cite everything (journals, news...). And I don't want people to use them less, I want people to have other options. It's the formatting that is (by and large) forced upon people. If you have more thana small number of references, templates make internal consistency much easier (please witness the steaming pile of wrongness that was the refs section at Jane Austen at the time it was taken to FAC). If we had MLA as a template option, we would have had the option to use MLA extensively while leveraging the consistency and (importantly!) the COinS of templates. But we are denied that option. It does not exist.  Lingzhi ♦ (talk) 09:48, 20 August 2016 (UTC)
  • The WMF devs will not write citation templates for you. The template editors right here at the English Wikipedia can do that for you. In fact, they went to some trouble a while back to set up the infrastructure that makes it possible to do exactly what you want done. I wish you luck with your task of convincing them to do what you want. Whatamidoing (WMF) (talk) 09:52, 20 August 2016 (UTC)
    • Perhaps my memory is wrong, but I was speaking under the (confabulated?) memory that when I created this new acct, I was immediately plopped into VE, and had to explicitly select options to get my way out of it. That would make VE the default option. Clearly, however, I am very much in the wrong, either for being here, or in what I've said, or both. Sorry to have bothered you.  Lingzhi ♦ (talk) 04:28, 21 August 2016 (UTC)
      • According to Special:CentralAuth/Lingzhi, you created this account in December 2012. The VisualEditor was originally released to registered editors in July 2013 (and then withdrawn). You probably did get a dialog box at some point during 2016, because there have been a few config changes and new preference options. Whatamidoing (WMF) (talk) 15:44, 22 August 2016 (UTC)

Highbeam translator is live

[1] Just needs to be pulled into Citoid now  czar 19:00, 2 September 2016 (UTC)

Thank you! I'll see if I can help nudge this process along. Whatamidoing (WMF) (talk) 15:03, 3 September 2016 (UTC)

I put the request into Phab anyway. HTH --Elitre (WMF) (talk) 14:35, 12 September 2016 (UTC)

Czar, done now. --Elitre (WMF) (talk) 15:06, 13 September 2016 (UTC)

A barnstar for you!

	The Technical Barnstar
	Thank you for helping me at VPT VarunFEB2003 07:29, 14 September 2016 (UTC)

Talkback! You've got messages!

Hello, Whatamidoing (WMF). You have new messages at Wikipedia:Village_pump_(technical).
Message added 12:14, 1 October 2016 (UTC). You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

VarunFEB2003 12:14, 1 October 2016 (UTC)

Talkback: you've got messages!

Hello, Whatamidoing (WMF). You have new messages at VarunFEB2003's talk page.
Message added VarunFEB2003 13:24, 3 October 2016 (UTC). You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

Q&A for math

Hi Whatamidoing,

unfortunately, I cannot get the context of your question https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&oldid=prev&diff=741471645

I was trying to collect information on how make best use of new math format now available from production at https://www.mediawiki.org/wiki/Extension:Math#Viewing_math

Is there something that should be added there?

If something is broken or not well documented a phabricator ticket would be awesome.

thanks a lot --Physikerwelt (talk) 14:07, 3 October 2016 (UTC)

Unfortunately, I don't know what the problem is. You might have to talk to Sphilbrick directly to get more information. (He might also be interested in your page on mediawiki.org.) I hope that, if it were a significant problem, that other people would be telling us about it. So perhaps it was not a significant problem. Whatamidoing (WMF) (talk) 05:33, 4 October 2016 (UTC)

(WMF) in usernames

Thanks for the details; I had forgotten that standardisation was comparatively new, since the only person with a staff account I remember meeting, User:Rob Schnautz (WMF), registered his account in 2012. You remember the difficulty over superprotect and staff accounts, I'm sure, but I figured that use of (WMF) in account names was largely standardised before then, that the staffer involved in the incident (I forget now who it was) had been a permitted exception to the rule, or something like that.

Have you ever seen a userbox for current staff and a userbox for former staff? Presumably we could attach a category to each one, and using the category (whether or not with the userbox) would make it significantly easier to find staff. Nyttend (talk) 12:08, 11 October 2016 (UTC)

You may be looking for Category:Wikimedia Foundation staff, whose counterpart at m:Category:Wikimedia Foundation staff will be more complete.

NB that many devs who began their wikilife in the community avoid using their staff accounts on wiki, as a show of solidarity and to discourage anyone from thinking "Oh, he must be right about this technical question, since he's on staff". So even if you have a complete list of all staff accounts, there will still be a bit of memorization involved. The very recent trend (i.e., started with me in 2013) of creating staff account names to match existing volunteer account names tends to reduce some of that (well, it tends to transfer the burden to staff who are trying to figure out how to send an e-mail message to invite me to a meeting  ;-), but for finding former staff accounts, it's anyone's guess what the username will be. Whatamidoing (WMF) (talk) 18:34, 11 October 2016 (UTC)

Using VE with a new account

Hi What. I've got a couple of new editors wanting to edit my sandbox with VE but it's not showing up as a tab for them. I created a new account and am finding the same thing. Can you help? Anthonyhcole (talk · contribs · email) 04:21, 3 November 2016 (UTC)

A handful of our fellow editors decided a few months ago that it was necessary to hide the visual editor from new people. The single edit tab system means that you have to pick one or the other to be "first", and they decided that the wikitext editor must be first. Editors (even new ones) can switch to the visual editor after first opening the wikitext editor by clicking on the pencil icon in the upper right corner (on userspace, mainspace, and a few other types of pages). Based on the last item in the screenshot for prefs ("Remember my last"), it should remember which editing system was used most recently. To make that more persistent, then change that setting to either all-visual, all-wikitext, or both tabs. Whatamidoing (WMF) (talk) 04:32, 3 November 2016 (UTC)

Sheesh. :o) OK. Got it. Thanks. If I go through this again, I'll create their accounts for them and tweak their settings.

How are things going on VE, anyway? Is there a timeline/summary page somewhere that shows its evolving functionality? Last time I looked (six months ago) it seemed pretty good for newbies (and old-timers — especially tables & citation). --Anthonyhcole (talk · contribs · email) 05:19, 3 November 2016 (UTC)

m:VisualEditor/Newsletter/2016/October lists the updates for the last three or four months. You can subscribe at Wikipedia:VisualEditor/Newsletter if you want to receive the newsletter (probably three to five times per year).

The next major feature is the mw:2017 wikitext editor. This will use the same toolbar as the visual mode, so you'll have access to the citoid service and other tools (all of which will put the wikitext code in the window). Whatamidoing (WMF) (talk) 16:25, 3 November 2016 (UTC)

Thank you. I'll take a look. --Anthonyhcole (talk · contribs · email) 05:57, 4 November 2016 (UTC)

Whatamidoing, I think it's unfair for you to say editors wanted to "hide the visual editor from new people". That isn't true. You were the one who came to us selling single-edit-tab as an improvement. We accepted it as the WMF trying to make a good-faith improvement for making both editors easily available, or at least we didn't oppose it. We merely said that if one editor will "primary" and the other "secondary", then obviously the primary editor should be the one that is overwhelmingly used.

If you are going to suggest that anyone had an intent to "hide" an editor, then you need to assert that motive onto someone(s) at WMF.

If you believe single edit tab made things worse for new editors, you could open a Village Pump proposal on whether it should be rolled back. I recall several editors saying they thought two edit links is the simpler and clearer solution, but not strongly enough to actively oppose the WMF's belief that it was making an improvement. Alsee (talk) 11:26, 27 November 2016 (UTC)

Alsee, don't bother trying Whatamidoing to admit that perhaps WMF or Whatamidoing was wrong in something like this. They are rather consistently unfair in these kind of things, but as it is in the defense of his employer, they get away with it. Many people are aware that trying to discuss this kind of thing with them is a waste of time. If you need a community liaison, use one of the others, most of them are better (although at least one is rather terrible as well). Apart from that, let them play with their little toy in their dusty corner, the money on it is wasted anyway so if the 5% are happy with it, then let them be. Fram (talk) 08:15, 28 November 2016 (UTC)

Wiki / encyclopedia / knowledge technology

Am I mistaken, or is this an interest of yours?

What's on your wish list? The Transhumanist 19:50, 10 November 2016 (UTC)

P.S.: I answered your question over at WP:VPT. -TT

I'm a classic Wikipedia editor.[2]

My wish list begins with world peace and ends with understanding why the lemon bread I made yesterday doesn't taste like I wanted it to. It's a l-o-n-g list.  ;-) Whatamidoing (WMF) (talk) 21:35, 10 November 2016 (UTC)

Well, I think wikis have brought world peace one step closer. And there's probably a wiki out there that explains the lemon bread thing. But what about wiki / encyclopedia / knowledge tech features that don't exist yet? Any desires? The Transhumanist 08:03, 11 November 2016 (UTC)

Definitely phab:T89970 and phab:T95739 and phab:T33597; I can see personal benefits to each of these.

In terms of solving bigger problems, I'd like to see something like phab:T31793 happen. How about you? Whatamidoing (WMF) (talk) 16:57, 11 November 2016 (UTC)

Hmmm. I wish for...

1. Off-line installer of Wikipedia datadump into MediaWiki (that actually works)
2. Easy-to-follow instructions on installing multiple wikis into MediaWiki (to have both Wikipedia and a personal wiki installed on one's machine)
3. Automatic taxonomy construction bots to build outlines, indices, glossaries, lists, categories, and portals
4. Auto-tab-loader of search results or selected links
5. Built-in activated-by-hot-key mind mapper for displaying key concepts/links
6. Built-in multi-screen tiling window manager
7. Multi-tile slideshow feature, for auto-browsing articles
8. Wikipedia displayed in edit-mode by default, WYSIWYG-style. The reader can simply start typing anywhere on the page, and when the user goes to leave the page, it asks if he wants his changes published in Wikipedia.
9. Content fork harvester bots to copy new information from {{Main}} sections and integrate it into the main articles corresponding to those sections
10. Content fork harvester translator bots to copy new information from articles in each language WP and integrate it into the corresponding articles in all the other language WPs, in those languages
11. Bots capable of automatic summarization of the World Wide Web by subject to produce, expand, and maintain Wikipedia articles, especially on missing topics
12. Intelligent personal assistant as an interface to the encyclopedia. I.e., "Jarvis".

Well, that makes an even dozen, but I'd settle for the first one. I greatly desire a local MediaWiki-hosted mirror of Wikipedia. The Transhumanist 07:38, 13 November 2016 (UTC)

P.S.: the link in #5 displays a mind map of your talk page. -TT

Have you thought about how much space that would need? Text doesn't take up much space, but we have a lot of it... Whatamidoing (WMF) (talk) 17:32, 26 November 2016 (UTC)

Looking for editors

I'm willing to volunteer.

I use Google hangouts (though I've never yet used it to send a screenshot). I use Chrome, and have Firefox and Safari installed. I use the old editor, not Visual Editor. Maproom (talk) 14:19, 24 November 2016 (UTC)

Hello, Whatamidoing! I also want to help with the test. I live in Uruguay (UTC-3), and usually edit on the latest Firefox. --NaBUru38 (talk) 13:14, 25 November 2016 (UTC)

Suggestions for the Visual Editor Citation Tool

Hi, sorry to disturb. Should suggestions for improvements to the Visual Editor Citation tool go on the VE feedback page or somewhere else? Red Fiona (talk) 10:05, 26 November 2016 (UTC)

You can post anything related to VisualEditor to WP:VEF. If that's not the ideal place, then we'll figure out where it goes later.  :-) Whatamidoing (WMF) (talk) 17:30, 26 November 2016 (UTC)

My Right

"বিশ্ব জোড়া পাঠশালা মোর সবার আমি ছাত্র " I don't want to get more wealth,I just want to stop Golden People death. I don't want to take rich path, I just want your Faith. SoniaAlam 17:59, 28 November 2016 (UTC) — Preceding unsigned comment added by SoniaAlam (talk • contribs)

Linter

@Jonesey95: Got an email from ssastry that mw:Extension:Linter extension will go beta in the next few months. Not sure of CheckWiki's role, but AWB and WPCleaner should be tied into it at some point. Bgwhite (talk) 06:15, 2 December 2016 (UTC)

I can only partially understand the page, but it looks like progress. I only really understand things when I see them in action. – Jonesey95 (talk) 19:25, 2 December 2016 (UTC)

Anything that begins with "Download software to your computer" is not going to be widely used. I think that the on-wiki/preferences-based tool will be used by more people.

What's your initial feeling about the new CX template tool? Is it a net improvement yet? Whatamidoing (WMF) (talk) 08:38, 5 December 2016 (UTC)

Testing on en.voy

I can help What do I need to do and when? You can use my talk, use {{Ping}} on your talk, or just email me directly. —Justin (koavf)❤T☮C☺M☯ 22:58, 5 December 2016 (UTC)

Wonderful! Thanks. I'll have Daisy get in touch with you right away. Whatamidoing (WMF) (talk) 23:55, 5 December 2016 (UTC)

Merry, merry!

From the icy Canajian north; to you and yours! FWiW Bzuk (talk) 23:09, 26 December 2016 (UTC)

Ciao

Just testing something. Happy New Year, BTW. Elitre (WMF) (talk) 17:21, 3 January 2017 (UTC)

A cookie for you!

Thanks for your help with finding the issue with Special:ContentTranslation. Cheers! RezonansowyakaRezy (talk | contribs) 00:12, 13 January 2017 (UTC)

Great length

Tracked in Phabricator
Task T92432

Tracked in Phabricator
Task T52568

This isn't great length, but there's only so long I can rant before you'll get annoyed at me.

The bug [and it definitely is a bug, not a feature] in question would not be a problem if Wikipedia were write-once-edit-never. Semantic values increase human usability and reduce confusion; they're why we have domain names instead of just IP addresses. And I am pro-human.

Denying VE users even the option to use ref names is a non-optimal design choice that makes things much more difficult for everyone else. DS (talk) 20:13, 19 January 2017 (UTC)

I love being able to agree with a good rant. I, too, could go on at great length (the product manager might say that I have already done so, more than once), but we already agree, and there's no need to preach to you.

The team basically accepts the need for this, but it doesn't seem to be among the next ~100 things that they are going to do. So "someday", but probably not soon. Whatamidoing (WMF) (talk) 20:36, 19 January 2017 (UTC)

Edit outages April 2017

I just read in the Education mailing list about the possible outages in April 2017 on all Wikimedia platforms. We will be organising two edit-a-thons on April 22 and April 29 in Brussels (10 a.m. till 5 p.m. local time). If these timeslots could be avoided, that would be great. If not, it would be good to know in advance when exactly it will happen. Beireke1 (talk) 14:18, 13 February 2017 (UTC)

Beireke1, those dates should be safe. They are unlikely to do this work on a Saturday. I'll make a note about it, though, and I'll try to let you know ASAP if they pick those days. Whatamidoing (WMF) (talk) 18:53, 13 February 2017 (UTC)

The planned dates are Wednesday, 19 April and Wednesday, 3 May 2017. Whatamidoing (WMF) (talk) 20:01, 13 March 2017 (UTC)

VE oddity

Do you think this deserves a phab ticket? It seems to be a parsoid error, judging by the comments in response; I saw you skipped it and was wondering if it's unfixable in some way, or perhaps already recorded. It's an odd bug because it's very difficult to even notice it's happening -- it just looks like ordinary text when it renders. Mike Christie (talk - contribs - library) 22:22, 3 March 2017 (UTC)

Hi Mike,

Thanks for this note. This one is on my list for the meeting that starts in 61 minutes. Assuming that James is there, I hope to get an answer about whether this one should be considered "new bug" or "sfn-like templates are still not officially supported". Whatamidoing (WMF) (talk) 15:59, 6 March 2017 (UTC)

Hi Mike,

The answer is that {{sfn}} isn't supported, and that {{efn}} "isn't supportable". James F sends his apologies for the poor user experience. Whatamidoing (WMF) (talk) 23:07, 6 March 2017 (UTC)

Well, it was worth asking. The bug is quite annoying in appearance, but I suspect that instances of it in practice will be rare. Thanks for letting me know. Mike Christie (talk - contribs - library) 23:13, 6 March 2017 (UTC)

I hope that it will continue to be a rare experience. I agree that it's confusing and annoying. (Imagine if that paragraph were edited by a newbie.) Whatamidoing (WMF) (talk) 00:29, 7 March 2017 (UTC)

Confusing phabricator task

At the 2017 Wikitext editor RfC Wikipedia:Village pump (proposals)#Proposal to submit blockers on replacing our wikitext editor you said "There is no plan to remove the old wikitext editor" (section header) and "there is no plan to remove any of the wikitext editors from any WMF site." At Phabricator (which I hav no intention to use considering who are the most active and enforcing editors there), I now see [3]: "Provide the VE wikitext editing mode as the mobile wikitext editor". Is that "The mobile wikitext editor" as in "the only one"? It sure reads that way. The associated Gerrit patch[4] says "Fully replace triggers to the old editor to launch NWE instead. " which again very strongly gives the impression that here at least, the intention is to remove the standard wikitext editor from mobile, and only provide the 2017 WE (and VE) instead. Which would be rather different from the message you sent at the RfC... Fram (talk) 11:55, 30 March 2017 (UTC)

@Alsee and Bethnaught: as you might be interested in this or want to share your views. Fram (talk) 12:01, 30 March 2017 (UTC)

@BethNaught: this should work better :-) Fram (talk) 12:02, 30 March 2017 (UTC)

My statement was true at the time that it was made in January. The Phab task that you link about changing the mobile editor (again; the current one is the second wikitext editor there) was written in March.

It is still true that there is no plan to remove the software that most Wikipedia editors call "the current wikitext editor" (the 2010 WikiEditor).

Another change since January is that a plan is being developed to remove the 2006 wikitext editor: it's got a long-term bitrot problem, and since only ~50 editors at this wiki have used it during the last month (and since many of those editors [see the discussion at VPT last week] would probably be better served by the 2003 wikitext editor), it's unlikely to be worth the resources necessary to fix it. Whatamidoing (WMF) (talk) 18:13, 30 March 2017 (UTC)

Can you perhaps make it clear to Jdforrester that this will need prior clear approval at enwiki (preferably through an RfC) with indications why this e.g. doesn't violate the three blockers for the NWE? At the moment, it doesn't even seem possible, from my limited tests, to use the NWE at mobile, so to simply switch the current editing environment to the NWE seems a rather premature idea, and considering the just-closed RfC one that might generate significant backlash. Fram (talk) 07:39, 31 March 2017 (UTC)

I haven't been able to find any evidence of a timeline for this Phab task. There might not be an actual "plan" at this point; stuff sometimes languishes in Patch-for-review state for years.

I'm not sure that asking the core community here is the best approach. Experienced editors are almost always desktop-only editors; if they edit on a mobile devices, they're usually using the desktop site (and so wouldn't be affected). It would be ideal to be able to ask the people who are actually affected by it (a group that doesn't appear to include either of us). I'm not sure how we could reach that group efficiently, though. If they build phab:T89970, then maybe it needs a way to limit a question by site. Whatamidoing (WMF) (talk) 17:14, 31 March 2017 (UTC)

"I haven't been able to find any evidence of a timeline for this Phab task. There might not be an actual "plan" at this point;" Perhaps actually "liaise" about this and contact the person who filed the task or created the patch or who is responsible for the team? They should know what they plan to do and when, no? As for "people not affected by it", I didn't use VE but still was heavily affected by it (negatively). I don't put on blinders and care only about what I see and encounter, I also check e.g. the vandalism shown to mobile viewers that desktop editors didn't get to see. Many desktop editors already have experience with the NWE, so asking them about how smart this plan is may give some meaningful insights (if the just-finished RFC and the blockers listed there and the promises given there aren't enough...). Fram (talk) 07:01, 3 April 2017 (UTC)

If you use a reasonable definition of plan, such as the one given in the article Plan, then your assumption that every single project has a plan behind it is invalid. He seems to have a general notion of what will (someday) happen, and I am asking for more information, but it's unclear whether the characteristics of an actual plan (e.g., a timeline and resources) will develop soon.

I'd be happy to hear more about how other people using NWE could significantly affect you. Whatamidoing (WMF) (talk) 16:46, 3 April 2017 (UTC)

"your assumption that every single project has a plan behind it is invalid." What assumption? Where did you get this idea that I assumed that every "project" has a plan (but only a plan according to your definition of course, just like it is only a blocker if you use the Mediawiki definition) behind it? I asked what they plan to do, not what their Plan is. Perhaps read Planning instead; when they created the Phabricator task and the Gerrit Patch, they were certainly "planning" to do something with it, I hope? There was some forethought, some intelligent process at work, and they had a specific goal with that task and patch? Or do the WMF people still blunder about blindly, doing random tasks without any coordination and meaning? If I am to believe you, yes, but then again... Anyway, next time perhaps stick to "I am asking for more information" instead of going of on irrelevant ramblings which don't help the discussion one bit and only make the WMF look like headless chickens.

"I'd be happy to hear more about how other people using NWE could significantly affect you." So you've forgotten the early days of VE completely, and how it did affect people not using VE but having to cleanup the mess it left behind anyway? Yuor confidence that enabling NWE for mobile won't have similar results is charming, but baseless. Fram (talk) 07:16, 4 April 2017 (UTC)

I remember the problems that Parsoid created (ranging from horrific bugs to minor differences in formatting conventions) quite clearly. I'm not sure why you think that's relevant, though. The wikitext that an editor types is the same no matter which of the dozen-plus editing tools that person is using, no?

On your other theme, sometimes a Phab task is just a reminder that something will need to be done eventually, or a way to note that a plan needs to be created. Projects do not always spring forth fully formed. Whatamidoing (WMF) (talk) 19:43, 4 April 2017 (UTC)

Sometimes, yes. Usually things the communities ask for, these can linger in phab for years and years. But this one, like I said right from the start, already has a Gerrit patch which is ready for review. This is not some vague "perhaps one day" issue, but something almost ready to be deployed, pushed through by Jdforrester, not some low-level volunteer. Fram (talk) 06:57, 5 April 2017 (UTC)

Whose exact words about this patch, BTW, were "Not soon". It sounds like they won't even take a decision about whether that's ultimately going to be the right approach for months (alternatives include at least various permutations of beta features and staged rollouts). I think that a fair summary of the situation is "no actual plan yet". Whatamidoing (WMF) (talk) 17:42, 5 April 2017 (UTC)

Hey

Don't know if you paid mind to the recent kerfluffle about use of Wikidata labels in mobile views of en-WP content but it appears that for now that has been stopped. See here.

The Reading team apparently wants feedback from the en-WP community about "blockers" and that is a discussion that should be teed up.. carefully, well-advertised, and somehow minded by someone like you who understands how both the editing community and the reading team think and operate. A chance to do this right. I don't know if you are interested or have time, or know someone else who might help. Jytdog (talk) 20:45, 30 March 2017 (UTC)

Hi Jytdog,

Thanks for the note (and the compliments ;-) . I saw the first RFC note at VPT; I haven't had time to follow any of the other village pumps. I'm fully committed until m:Tech/Server switch 2017 is over (3 May 2017, if all goes well). None of my teammates are embedded in the Reading team at the moment, so I'm not sure if there's an obvious person to pass this to. I also don't know what's in Reading's annual plan, so even if a perfect technical blocker [1] was identified, there might not be anything that could be done about it for another 15 months. How soon do you think this should be addressed?

[1] Just for the record: "Never put anything from a sister site on my wiki" or "Make another community be just like us" aren't technical blockers. And that's another layer of problems, because I suspect that "it's separate, it's different, and I don't understand them" is, in reality, the most objectionable aspect for many editors here. It's not an unreasonable POV, but it's not a technical blocker, either. Whatamidoing (WMF) (talk) 17:27, 31 March 2017 (UTC)

Just for the record: as far as I can see, Jytdog wasn't discussing technical blockers, but things the enwiki editing community considers "blockers" full stop. The focus on "technical" blockers only is not a good aproach. Points like "text added / changed through external sites, which isn't affected by our blocks, protection, ... is fundamentally unacceptable" is just as much a blocker as "preview doesn't work". Your "... and I don't understand them" is typical WMF disdain for the very real and fact-based problems many people here have with things like Wikidata (or at least some aspects of it), and previously had about things like Gather and so on. The only thing many of us don't understand is the company culture at the WMF and the utter disregard they seem to have for essential things like testing, community input, keeping promises, and so on. While some things have improved, some things clearly stay the same. Fram (talk) 06:56, 3 April 2017 (UTC)

The idea of "blockers", as used in that discussion, comes from a specific set of advice pages at mediawiki.org, and the idea there is exclusively about actionable technical blockers. Basically, if it can't be fixed via coding (or the equivalent in UI design, config changes, etc.), then it's not a blocker for the technical project. (Also, under that model, it's the devs who decide whether a complaint is actually a blocker for a technical project, not a community.) Whatamidoing (WMF) (talk) 16:36, 3 April 2017 (UTC)

The idea that things the enwiki community finds blockers will be ignored by the WMF because they are not technical may well be correct, but will only backfire against the WMF. The devs can try whatever they want, in general if enough people here don't want it, we'll find a way to disable it anyway, by brute force if needed. WMF policies stand above local policies (e.g. a community without a BLP policy, like Wikidata, cannot simply ignore the WMF stance on this, and globally locked editors shouldn't be allowed to edit locally anyway); but the idea that WMF devs can decide what is acceptable for communities like enwiki and dewiki and what isn't is theoretically probably true, but in practice has turned out to be a very bad idea each and every time.

Most people in this enwiki discussion have no idea of the existence of that mediawiki.org advice page or of the definition of "blocker" you are using, but use the wikipedia definition: "A blocker is someone or something that blocks." This may be a technical issue, but also a person or organisation, or a philosophical difference in approach, or ... Whatever enough people decide is a blocker, is a blocker, no matter if it is technical in nature or not, no matter if the devs agree or not. I thought this would clear to you by now, after all the things that got blocked at enwiki (Gather, AFT, Mediawiki Viewer, now the wikidata descriptions, ...) and the significant friction this caused between enwiki and WMF. Mediawiki is free to set up any model of blocker they want, but it simply won't fly in reality. Fram (talk) 07:25, 4 April 2017 (UTC)

Had the parallel discussion on my Talk page here User_talk:Jytdog#.22Blockers.22. Jytdog (talk) 17:00, 4 April 2017 (UTC)

As the docs say, "Ultimately, the responsibility to define whether a problem reported is a blocker or not belongs to the development team." If you plan accordingly, you are more likely to reach a solution that everyone can work with.

(I'm not sure that I buy the claim that Wikidata has insufficient BLP-related policies. Isn't their policy that everything should be sourced to top-quality ["authoritative", not merely "reliable"] sources? There is certainly some work to do to reach that point, but it seems to me that their baseline policy is stronger than ours. If memory serves, the main point of their BLP proposal was to say that they didn't want to include home addresses for a bunch of celebrities. The sourcing issue is already covered at WD:V.) Whatamidoing (WMF) (talk) 20:00, 4 April 2017 (UTC)

The last paragraph is wrong. They are a very young project and have almost no policies that I have found. WD:V is just a proposal, as is WD:LP. Even in practice, they don't think about V - as I have pointed out several times at WT:MED, their training materials and talks including ones by Lydia, never even mention sourcing. With respect to attitudes about data about living people and V in WD, this RfC from last summer is instructive as is this current discussion at their Chat. Jytdog (talk) 20:42, 4 April 2017 (UTC)

A site which thinks Quora and the like are good sources is not aiming for top-quality, authoritative sources at all, they are setting the bar considerably lower than enwiki. As for your first sentence, I couldn't care less what the docs at mediawiki say. If something that is considered a blocker by the enwiki community is disregarded by the devs, it usually backfires quite badly. If they and you would recognise this and change their docs and approach accordingly, "you are more likely to reach a solution that everyone can work with." Fram (talk) 06:53, 5 April 2017 (UTC)

They don't seem to have any sourcing policies. See Wikidata:List of policies and guidelines. SarahSV ^(talk) 07:20, 5 April 2017 (UTC)

Does a policy have to be written down and stamped with a template that says "policy" at the top, for a community to have a policy in practice? (If so, then Commons doesn't have a policies on either living people or sourcing, either.) Whatamidoing (WMF) (talk) 18:03, 5 April 2017 (UTC)

When I last checked, Wikidata bots were adding unsourced text from other wikis, so (assuming that continues) there is no sourcing policy, written or otherwise.

Wikidata is nearly five years old. By the time enwiki was five, the core content policies were in place, and without the benefit of a similar project having done it already.

Sourcing policies are what make things difficult. It's fun otherwise to roam around adding a few words here and there. It's not so much fun when your work is expected to be accurate and attributed to an RS. It looks as though Wikidata has chosen to keep things easy to attract contributors, and that's where the danger of it lies. SarahSV ^(talk) 00:50, 6 April 2017 (UTC)

But - if you look at entries of the qualifiers for an entry - there is the room for 'verification' links - each claim requires a reference - so maybe not advertised as a policy - maybe it can be seen there is a process in place for verifiable proof for a claim - if that was checked - maybe 20 or 30 % of wd actually has properly verified data ? JarrahTree 09:33, 6 April 2017 (UTC)

Special:RecentChanges gives no bot edits during the last month, in any namespace. (Perhaps they just don't use a bot flag?) But if you think about it, they ought to have bots that import some specific information from the wikis: interlanguage links and Commons categories, to give just two examples. (One also does hope that information taken from this wiki is sourced here, so 'came from this English Wikipedia article, which ought to contain the original sources' is not exactly the same thing as "unsourced".)

The concept of a "reliable source" doesn't map well to that project. The key features of a reliable source – impartiality, editiorial oversight, not self-published, etc. – aren't as relevant for the data that they collect, which is usually more like "What's the URL to the Google Knowledge Graph for this subject?" or "What is the population of this city according to its country's election board?" than "What is a fair description of the literary merits of this poem?" They actually have a reasonable point about wanting "authoritative" sources: they need the URL to that Knowledge Graph itself; having it mentioned in a typical third-party "reliable source" is irrelevant. I suspect that we'll get nowhere if Wikipedia editors keep asking them to use our kinds of sources. Whatamidoing (WMF) (talk) 16:38, 6 April 2017 (UTC)

You need to show registered users to get to see the bots, since bots are also registered users (perhaps this needs a phab task a swell, as it is counterintuitive as it stands). [5] And your "reliable source" "taken from enwiki" conclusion goes against what enwiki (and I think all other wikipedia languages) state, i.e. that another wiki is not a reliable source, and that you have to use the source directly, not think / hope that the information will be sourced on that other wiki. Every wiki needs to improve its sourcing, we don't achieve that by copying information from one to the other, and certainly not by sourcing the central, supposedly authoritative repository from the standard wikipedias. Your idea about what kind of data they have and why they take a different approach to reliable sources doesn't match reality either. This is not about wanting reliable sources for the endless and largely useless list of identifiers they show at the bottom; this is about reliable sources for the many properties they show at the top, which are things very similar to what wiki articles have and is the kind of information they try to fill infoboxes or create pageholder articles with. So yes, we should continue asking them to use such sources, and refuse to use their data if they don't have that kind of sourcing and demand that kind of sourcing in policy (certainly in BLPs, but in general as well). Something like the number of deaths from the holocaust (according to Wikidata for the last 15 minutes: 0[6]) needs an impartial, not slef-published source with editorial oversight and the like. Wkidata is stuffed full with information that needs such sources. I guess we'll get nowhere if even the WMF supports their misguided idea that this isn't needed for Wikidata, I thought only GerardM defended this but I am clearly mistaken. Fram (talk) 06:49, 7 April 2017 (UTC)

Well, the holocaust again happened on Wikidata after only 1 hour of vandalism, great. What about something like [7], a BLP of a 12-year old without any sources at all (nor any links to other wikiversions)? According to your description, all one needs to do is add an authoritative source, say his Twitter account, and this is good enough and other wikis shouldn't complain? No thanks. Fram (talk) 08:26, 7 April 2017 (UTC)

I'm not saying that enwiki is a good source; I'm saying that an entry that is actually sourced – even if it is sourced to a weak source – is not actually "unsourced". "Information<ref>en.wikipedia.org/wiki/Page</ref>" is a sourced statement; "Information" is an unsourced statement.

On your Holocaust example, I don't think that "number of deaths" is actually what they want. I think that they actually want "number of deaths according to Yad Vashem" (which reports five to six million) and "number of deaths according to US National Archives" and "number of deaths according to the German government" and "number of deaths according to Dr. William Hoettl's court testimony" (who said six million Jews were killed or died) and "number of deaths according to Wolfgang Benz" (5.3 to 6.2 million) and all sorts of other numbers. And then they need qualifiers that indicate whether the figure is supposed to include only approximately six million deaths of Jews, or if it should also include other groups, such as the approximately one million Romas that were similarly killed.

I don't know if Wikidata considers a Twitter account to be authoritative for the existence of a person. I suspect that they would accept that as being authoritative only for the existence of the Twitter account. Whatamidoing (WMF) (talk) 16:51, 21 April 2017 (UTC)

"I'm not saying that enwiki is a good source; I'm saying that an entry that is actually sourced – even if it is sourced to a weak source – is not actually "unsourced". " I already responded to that strawman argument above, no idea why you reiterate it here. I don't care at all if something is unsourced or not, since sourcing it to enwiki, or some unreliable source, or my neighbour, is just as untrustworthy as not being sourced at all. That's what about everyone in this discussion is trying to explain at length, so to go back to "but it isn't unsourced!", as if that somehow is relevant or convincing, is basically ending this discussion with you on the same note as many previous ones. That's even ignoring your ramblings on the Holocaust. Yes, they can include multiple numbers, from multiple sources, which will al be in the same range. Having the number at 0 is vandalism, plain and simple. Just like multiple wikilanguage versions who take their infobox data from Wikidata (improvement! The future! Every house should have one!) stated last week for 1 1/2 day that Kurt Cobain was born in Denmark and died from anal sex. This again is not some obscure, rarely visited article, but an article with on enwiki some 10,000 pageviews per day, and hundreds of watchlists looking at it. Using Wikidata is inviting vandalism, extremely poorly sourced or usually unsourced "facts", and all kinds of problems we have been trying to eradicate here for 15 years now. I hoped some discussion with you was possible, but it seems that in all three cases here (the reliable sourcing, the vandalism of things like the Holocaust article, and the BLP problems with articles like Cash Hudson) you fail to grasp even the most basic problems surrounding it miserably (or try to deflect the discussion with strawmans and non sequiturs, that's also a possibility). Is this an actual requirement of working at the WMF, to deflect all criticism of their pet brainchilds at all costs? Or just a characteristic of some of its employees, no matter what their job description and superficial helpfulness says? I have no idea why I still hope, after all this years, that you will actually be a useful community liaison instead of this. Fram (talk) 08:55, 24 April 2017 (UTC)

You keep hoping because it turns out to be true, at least on occasion.  ;-) I've been asking around, and a lot of bots or bot-like scripts are unflagged at Wikidata, which means that there is no easy way to identify how a particular edit was made (just like AWB and Twinkle edits here often get counted as "wikitext editor", even though it's not true). However, a guess of around 4% "fully human" (1 in 25) changes to items at Wikidata seems to be reasonable. For comparison purposes, here at the English Wikipedia, about half of all edits are believed to "fully human". Whatamidoing (WMF) (talk) 17:42, 24 April 2017 (UTC)

Twinkle/AWB being counted as wikitext editor is only important to the WMF becaus they make the VE use stas seem even lower than they already are, which is a sad thing for the hyped default editor. Bot edits vs. not bot edits is important on a whole other level, as bot edits are not checked but made automatically, without thought or control. They have the side-effect of boosting the number of supposed Wikidata edits (just like many "edits" at Wikidata are actually edits on another wiki, e.g. moving a page), but the main issue is that flagging things as "bot edit" makes it clear that they are fully automated and easily fooled. Fram (talk) 07:08, 25 April 2017 (UTC)

I think the reason to track specific tools is to get resources fairly allocated. If 50% of edits here are done with bots or semi-automated tools such as Huggle, then maybe there ought to be some grant money or a WMF dev whose job is to keep the more popular tools working.

I'm not sure that "easily fooled" is a very relevant consideration for importing a database. You'd need a human to be thoughtful about how to set it up, but if your database is, say, PMIDs and their exactly matching doi's, then it ought to be imported by bot: it's not possible to get fooled for this kind of work, but a human is going to introduce typos or paste something in the wrong field. We have made this particular edit with a bot here at the English Wikipedia for years; I can't imagine why Wikidata shouldn't also use a bot for it. Most of their edits are along these lines. Whatamidoing (WMF) (talk) 15:25, 25 April 2017 (UTC)

No, most of their edits are importing unreliable sources, mainly wikipedia versions. Fram (talk) 17:23, 25 April 2017 (UTC)

I just checked the entire first page of RecentChanges. I found sourced additions such as this and this, and many, many changes to the Wikipedia interwiki links, but I found zero edits that cited Wikipedia. What have you found? Whatamidoing (WMF) (talk) 03:13, 26 April 2017 (UTC)

[8] and all creations by that bot? Sourced to tools.wmflabs.org (brilliantly reliable source, that), and when I open that link[9], I get "source "//sv.wikipedia.org/w/index.php?title=Lista_%C3%B6ver_fornl%C3%A4mningar_i_Essunga_kommun&oldid=29976091"". Another editor is adding items sourced to Wiki Loves Monuments Italia[10]. Most human-created entries are totally unsourced, of course. This bot adds unsourced items based on enwiki. This bot does the same based on e.g. wikispecies or dewiki. Fram (talk) 07:30, 26 April 2017 (UTC)

no content translation submenu item off contributons menu item

Tracked in Phabricator
Task T163344

I am endo999 and I don't have the Content Translation system submenu item from the Contributions menu. I also don't have any beta Content Translation notice in beta tools of preferences. Hadrien B. is having the same issue.

Endo999 (talk) 00:48, 20 April 2017 (UTC)

Hi Endo999,

The database behind ContentTranslation had some problems during yesterday's m:Tech/Server switch 2017. I haven't heard the latest news on it, but fixing it was their top priority yesterday. Whatamidoing (WMF) (talk) 16:59, 20 April 2017 (UTC)

Cursor jumping

Hey there, following up on this. I tried the "New wikitext mode" beta feature and hitting shift there can move the cursor as well. Haven't done testing to see when exactly it occurs. Ed ^{[talk] [majestic titan]} 01:37, 30 April 2017 (UTC)

Weird. Thanks for the update. Whatamidoing (WMF) (talk) 19:14, 1 May 2017 (UTC)

Archived AllPages VPT thread

• Wikipedia:Village pump (technical)/Archive 153 § Special:AllPages disabled

Hi there. I was looking through the VPT archive and spotted this thread. It does not necessarily answer your question completely, but the user's latest talk might allow you to draw some conclusions or inferences. It seems worth mentioning this to you, as it could be of broader interest to the WMF. Murph9000 (talk) 11:40, 3 May 2017 (UTC)

If someone is interested in making a copy of the English Wikipedia (whether private or mirrored on another website), then I cannot imagine why the WMF would want to stop him. The WMF provides tools to do this (a fact that seemed to be unknown to him, based on his questions at VPT). I probably should have just assumed that's what he wanted to do and pointed him at Wikipedia:Database download and https://dumps.wikimedia.org/ to begin with. I'll post those links on his user talk page now. Whatamidoing (WMF) (talk) 16:27, 3 May 2017 (UTC)

Sure, it's out there for anyone. The particular POV push just seemed noteworthy combined with the past question and current events. Murph9000 (talk) 16:45, 3 May 2017 (UTC)

Happy Day!

Doc James has sent me a great translation tool that will speed up my work by x100. I am pretty excited. There is one problem. When the translation of an article from English to Kreyol Aysien is finished and appears in the ht wikipedya, the title appears in English. I can move the page to a new one with a title in Kreyol but now I leave behind a trail of empty articles with English names. So, I would like administrative status to be able to remove these empty pages when I have finished translating an article. I anticipate being able to create dozens, especially since my summer break is coming up with more free time to pursue my hobby.

Best Regards,

Barbara (WVS) ✐ ✉ 11:19, 6 May 2017 (UTC) aka Barbara Page

We have to take this through the Stewards. It's not super complicated (you may recall that when we elected User:WhisperToMe, the hardest part was getting a few people to show up in the discussion to endorse him), but there are procedures to follow. w:ht:Wikipedya:Administratè is the place for the discussion. Which account would you like to use for the admin bit? WhatamIdoing (talk) 06:03, 8 May 2017 (UTC)

WhatamIdoing, I found that an English redirect is created to the Haitian version of an article. This happens when I move the page to the new Haitian title. So the problem doesn't really exist any more. If you still think that it would be useful, then let's use the name MdmPaj. Therefore, I don't have to be an administrator if it is going to be a bother. Best Regards,

Barbara (WVS) ✐ ✉ 12:29, 8 May 2017 (UTC)