Wikimedia Forum

(Redirected from WM:FORUM)
← Discussion pages Wikimedia Forums Archives →
QA icon clr.svg

The Wikimedia Forum is a central place for questions, announcements and other discussions about the Wikimedia Foundation and its projects. (For discussion about the Meta wiki, see Meta:Babel.)
This is not the place to make technical queries regarding the MediaWiki software; please ask such questions at the MediaWiki support desk; technical questions about Wikimedia wikis, however, can be placed on Tech page.

You can reply to a topic by clicking the "[edit]" link beside that section, or you can start a new discussion.
Wikimedia Meta-Wiki


Filing cabinet icon.svg
SpBot archives all sections tagged with {{Section resolved|1=~~~~}} and sections whose most recent comment is older than 30 days.

New Project Proposal - WikipragmaticaEdit

Hello everyone. Noob here. I've submitted a new project proposal that re-indexes Wikipedia, but also does a whole bunch of other nifty stuff. I look forward to our discussion. I've included the description below.


This proposal presents a paraphrase graph database extension to Wikidata. The intent is to create a directed graph of paraphrase nodes and retained context edges initially of Wikipedia content, but ultimately Internet wide. Initially focusing on unstructured data (text), the vision is to be media agnostic harvesting sentence equivalent information. This foundational construct is intended to support a wide range of uses from indexing to misinformation detection to knowledge representation to enterprise work management. —The preceding unsigned comment was added by DougClark55 (talk) 22:44, 7 January 2021

Project proposal - Feedback - Youth in IG WikiContestEdit

Hi everyone. First of all, let me introduce myself. My name is Pedro Lana, I'm a member of the board of Youth SIG (also known as the Youth Observatory, a global chapter of Internet Society, which is not incorporated as a separate entity), and of Creative Commons Brazil. I've worked with people from the Wikimedia during these last years, but I myself am not familiar with the community, so I hope I am posting this in the right place.

The Youth Observatory promotes an open perspective of Internet Governance, but recently we noticed that most of our members didn't participate in other communities that we consider essential to promote a more open and inclusive cultural environment. We plan to change this by encouraging our target audience (young people in the internet governance ecosystem) to participate in projects aimed at strengthening an open internet in its multiple aspects.

Our rapid grant proposal for a Contest named "Youth in IG WikiContest" is one of our first steps in this plan, and I would like your feedback to advise us on what we can still improve or ideas for better engagement. Other ideas to engage existing communities in Wikimedia projects are also welcome!

The Youth Observatory also remains available if anyone is interested in our support or partnership in something related to youth in Internet Governance.


Wikimedia Enterprise timelineEdit

--Guy Macon (talk) 12:58, 17 March 2021 (UTC)

From the FAQ:
"Will the Enterprise API affect the current Dumps and APIs? No. The system of freely-provided database dumps and APIs remains in place and continues to be supported. They are not being removed or changed as a result of the existence of the new API, and will continue to receive support and development. In fact, independently of the specific Enterprise API, the existing API ecosystem is currently being revamped to provide an improved experience under the "API Gateway" initiative. Part of the reason the Enterprise API is being built separately is in order not to disrupt the existing ecosystems."
...because the W?F is totally going to resist the temptation to not make improvements to the freely-provided database dumps and APIs in favor of improving the enterprise database dumps and APIs that are moneymakers. --Guy Macon (talk) 13:13, 17 March 2021 (UTC)
I am answering solely in my personal capacity and not for the WMF, but also based on knowing what is proposed/planned. It is envisioned that the enterprise-grade dumps and APIs will be made available for free for users for whom the free API and dumps are not adequate (most researchers, for example don't need or want realtime access but bulk access), whose work is noncommercial, and would support the Wikimedia movement (including volunteer content moderation workflows). The concept here is being carefully designed to create a product that *large* *commercial* re-users will find worth paying for - there's no real incentive to also restrict free access to volunteers and researchers and so on. (Google isn't going to say: oh we aren't going to pay for this service that we would find beneficial, because community members devising interesting moderation tools and researchers studying editing habits are getting it for free.) Your cynicism is noted, but it isn't as if this hasn't already been addressed.--Jimbo Wales (talk) 18:20, 17 March 2021 (UTC)
A comment from the Hacker News thread linked above:
"Devils advocate: Google clearly already has a working pipeline to import and format Wikipedia data for its needs. Why would they stop using it and start paying Wikipedia? Will Wikipedia be able to build an enterprise API that's faster/cheaper/more reliable/more scalable than the internal one build by one of the world’s top engineering companies?
No doubt the enterprise API will add attractive value for smaller companies without the resources to process the raw dumps but I’m skeptical that this will convert Google et al into well-paying customers. Unless they start restricting the free dumps... "
--Guy Macon (talk) 13:33, 17 March 2021 (UTC)
Again answering solely in my personal capacity and not speaking for the WMF. In my experience, having talked to these organizations at the highest level for a very long period of time, I do not believe that they will approach this in a simple "build or buy" framework. And in particular, those organizations (all of the ones who I think are interesting target customers) it is the *realtime* access *at scale* which they will find interesting, as opposed to importing dumps, which they may continue to do for some use cases of course. It is true in a sense that the biggest "competitor" for the new product will be the WMF itself with the existing free offerings. I think that's a good thing, and will always be true. But I think there is a pretty obvious set of value-add (*realtime* access *at scale*) which both would cost the WMF money to provide (which is why we don't do much of it now) and will be an interesting product for a handful of large re-users.--Jimbo Wales (talk) 18:20, 17 March 2021 (UTC)
It is certainly true that these big-tech organisations have infinitely more engineering resources (hardware, money, and people) than us. And therefore they do indeed have current workflows - plural - for how they access and incorporate Wikimedia content into their various products. However, especially for these largest organisations, those workflows are unstable and uncertain (from their business-perspective) because we do not a) build explicitly for their scale needs, nor b) offer a Service Level Agreement that guarantees the standard of access, nor c) offer dedicated customer support to them when things go wrong. These are the things those companies - no matter how big they are - cannot provide for themselves. These big companies want to have a contract that they can legally rely upon which says they can access Wikimedia content in a specific way, with a specific format, at a specific rate/speed/uptime, for at least a specific amount of time. They do not have that and - we expect - will be willing to pay for it. Smaller companies, without those immense engineering resources, will also benefit from a more 'level playing field' because we will now be able to provide an 'enterprise grade' service to them too. As described in the operating principles there is no 'exclusives'. I also said 'we expect' a couple of sentences ago for a reason: that this is a work in progress. This is not fully developed and already in operation. This is the moment before there are actual customers and an actual product that the team working on this project are coming to the community to say 'hello, this is what we're building, come and look at the FAQ, principles, technical documentation, and tell us what you think of the progress and the concepts.' 1 year ago it was just an idea that was appearing in the strategy documents and would have been to early to have any meaningful conversation about practical implementation because it would have just been theory. 1 year into the future it would be too late to have any meaningful conversation because anything would already be built and operational. So now is the time when there is something still in 'beta' development that is able to be shown to the community for informed discussion (both technical, and principles). One final thing, I see that the concern expressed on the HackerNews comment that the WMF would limit the existing services (either intentionally, or by neglect) as a way to force companies onto a paid-access model. That would be completely against the operating principles I linked above, and antithetical to mission in general. LWyatt (WMF) (talk) 15:09, 17 March 2021 (UTC)
Thanks! I certainly don't think anyone plans on making the paid access better to the detriment of the free access, just that the temptation will be there. A while back I had a discussion with the foundation about us sending a HTTP referer (yes, that's how it is spelled) when someone clicks on a link. I showed them how doing that can in some cases allow third party websites to tell exactly what pages a Wikipedia user is reading (ask me if you want all of the gory technical details; the method isn't obvious but is well known to website designers). The foundation's conclusion was that the privacy of our readers is less important than our fundraisers being able to prove to potential donors how much traffic our links were sending their way. So I am understandably a bit skeptical about the foundation making decisions that reduce revenue. "It is difficult to get a man to understand something, when his salary depends on his not understanding it." --Upton Sinclair --Guy Macon (talk) 21:28, 17 March 2021 (UTC)
@Guy Macon "HTTP referer (yes, that's how it is spelled)": Interestingly, your link goes to a nonexistent page called "HTTP HTTP referer" 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 22:50, 19 March 2021 (UTC)
Fixed the link. --Guy Macon (talk) 04:30, 27 March 2021 (UTC)
Re Google, their system is extremely fragile: on OTRS and elsewhere we keep receiving reports of incorrect search results (e.g. vandalism cached by Google, and long gone from WMF caches, persisting in their search result page for days). The real-time data is already out there, but they're evidently unable to use it correctly. I'm not saying it's going to be easy to do better, but there's definitely room for improvement. Sure, Google could throw more money at the problem and solve it themselves, but it's probably easier for WMF to provide a consistent/reliable service: WMF has specific expertise (assuming the "normal" WMF employees are going to work together with the "LLC-focused employees") and economies of scale (assuming more than one customer needs the same or similar service, for instance at least two search engines). Nemo 05:40, 18 March 2021 (UTC)
If Google is unwilling to throw money at the problem and solve it themselves, do we have any evidence that they will be willing to throw money at a new LLC that promises to solve it for them? --Guy Macon (talk) 23:13, 18 March 2021 (UTC)
These large organisations already are "throwing money at the problem" - but they cannot solve problems that are beyond their control. The thing that those organisations, even with their vastly larger resources (technical, staff, financial) cannot control is the stability, format, and speed of the 'flow' of information that we publish. There is nothing 'wrong' with the existing APIs of course - but it is insufficient for their specific needs. By metaphor: the electricity that flows into my house, and the electricity that flows into the factory outside my window, is the same - but the infrastructure needed for me to receive it compared to what that factory needs - is quite different. Only the "electricity company" can build the extreme power-supply that factory needs, and the contractual guarantee that the electricity will arrive at the specific flow rate, and specific high voltage, that they need. THAT is what the Enterprise API's customers will be paying for: because they cannot build it, or guarantee its supply, by themselves. LWyatt (WMF) (talk) 11:52, 19 March 2021 (UTC)
Those are all good points, but they do not address the question asked. If, as you say, "THAT is what the Enterprise API's customers will be [not are] paying for: because they cannot build it, or guarantee its supply, by themselves", why then are they not willing to pay to create it? Has any potential customer signed a contract agreeing that if the foundation creates X the customer will pay Y for it? Even if we have no such financial commitment, has Google, Amazon, or Apple given us a definition of what they might be willing to buy in the future, or are the details of the Enterprise API based upon what some foundation engineers are guessing that the customers might want? --Guy Macon (talk) 02:23, 20 March 2021 (UTC)
It was stated on the record in an interview for WIRED that conversations between the the big tech companies and the project team "are already underway" - which is is a simple way of saying that yes, the 'enterprise' team has been in close contact with those companies: finding out what they currently do and how they do it, what they need and can't do themselves, what they'd like instead, and would they be willing to pay for it... This is not being built in a vacuum. LWyatt (WMF) (talk) 04:14, 21 March 2021 (UTC)
Thanks! Very helpful. --Guy Macon (talk) 05:38, 21 March 2021 (UTC)

The Enterprise API: a technical analysisEdit

Based upon my decades of managing hardware and software projects and as a consultant specializing in rescuing engineering projects that are in trouble, here is my technical analysis based upon first principles with no insider information about the Wikimedia Foundation, Google, Amazon, or Apple:

  • Assumption #1: The software that Wikipedia runs on can be run on someone else's' computer.
  • Assumption #2: The content on Wikipedia can be mirrored on someone else's' computer.


Wikipedia:Mirrors and forks,
Downloading the Wikimedia software,
Wikimedia Installation Guide. and
How to mirror Wikipedia examine the above assumptions. Also keep in mind that you can create a mirror of most things that are available on the internet by crawling the website like any ordinary user.

If the above assumptions are true, then Google, Amazon, and Apple can create and maintain complete and frequently updated copies of Wikipedia on their internal servers.

They could also check Wikipedia (we are on the web, after all) to verify that individual pages on their internal server are up to date, with the choice of what pages to update driven by them displaying that content to their users in some form.

At this point, engineers at Google, Amazon, and Apple could reverse engineer the Enterprise API and create an identical API that pulls data from their internal servers. They might even be able to poach some foundation engineers, at least some of whom would welcome a doubling of salary and a promise of better management.

This would give Google, Amazon, and Apple exactly what they would get from the Enterprise API without paying a dime for it.

Engineers at Google, Amazon, and Apple could then attempt to differentiate themselves from each other by making their internal API better in some way than the Enterprise API.

One obvious improvement would be to not update their copy when a new Wikipedia user makes a change unless the edit survives without being reverted for a day or two. This would give them a feed with less vandalism.

My conclusion is that this proposal has a fatal flaw; the potential customer can easily eliminate the middleman. --Guy Macon (talk) 04:31, 20 March 2021 (UTC)

NOTE: A matching comment was published on the English Wikipedia village pump. People wishing to read replies can find them there. LWyatt (WMF) (talk) 04:14, 21 March 2021 (UTC)
That discussion was archived and is now here.
The one question that nobody had a good answer for in that discussion is this:
So why haven't these potential customers made a financial commitment? If, as claimed, they will be willing to pay a substantial amount for such a service sometime in the future, why are they not willing to invest even a small amount towards creating the service? --Guy Macon (talk) 04:59, 10 April 2021 (UTC)

Note, for those following this subheading:
On Friday 16th @1500 UTC - one week from the time the time this message was posted (minus 1 hour) - the Wikimedia Enterprise team will be hosting the April edition of our monthly Office Hours. You are most welcome to join us. Details for how to join are available at our documentation homepages here on meta @ Wikimedia Enterprise.
Equally on Meta are instructions for requesting a separate call for your community group (at a time and technical platform of your choosing); and the videos of the recordings of the two office-hours meetings from March. Relatedly, the team will be presenting at the forthcoming EMWCon Spring 2021 conference later in April. Sincerely, LWyatt (WMF) (talk) 15:56, 9 April 2021 (UTC)

@LWyatt (WMF): could you add the office hours to the Front page of meta - think that will help get the message out. Nosebagbear (talk) 17:04, 9 April 2021 (UTC)
Done. LWyatt (WMF) (talk) 17:25, 9 April 2021 (UTC)

This was covered in The Wikipedia Signpost. In that sin post article I was quoted as saying:

"...DGG is very skeptical about the WMF's need to raise more money and fears "commercial entanglement". User Guy Macon has similar views, believing that raising more money only distracts from the needed focus on the volunteer community. When asked if the WMF could do anything to earn his support, he answered.
' Do it without any money from donors or employee time paid for by donors. Every other for-profit LLC has to secure funding somehow; why should this one be different? Make a pitch to Google, Amazon, and Apple. If they are willing to pay for the service in the future they should be willing to invest in completely funding the creation of the service now.' ".
I stand by that statement. --Guy Macon (talk) 04:59, 10 April 2021 (UTC)
Whoever invests gets to make the decisions and define the operational principles. The startup investment in this project comes from the Wikimedia Foundation, under the direction of the Board of Trustees, and so consequently the operational principles are determined for the good of the movement.
To have outsourced the financing of this [even IF that was actually possible, which is a different question entirely] would have been to also outsource the oversight and design control. Something which would not be appropriate in my opinion. The whole reason we can even have these open conversations about what is appropriate principles etc. is because it is an in-house thing. LWyatt (WMF) (talk) 17:11, 10 April 2021 (UTC)

Line numbering coming soon to all wikisEdit

-- Johanna Strodt (WMDE) 15:06, 12 April 2021 (UTC)

Creating a Navigation BarEdit

Hi Everyone, I would like to know if there is a resources material where I could read on how to create different navigation portal eg. Home|About us|Campaigns|Contacts etc. if there is an easy material that has a lot resources the would enable me learn how to create various navigation tabs for campaigns I would be very grateful. Thanks Jwale2 (talk) 23:08, 14 April 2021 (UTC)

@Jwale2: Do you mean the mediawiki:sidebar? Otherwise which navigation bar are you meaning. Pointers are useful.  — billinghurst sDrewth 08:31, 5 May 2021 (UTC)

Threatened, harassed & bullied by cross-wiki vandalsEdit

Hi, there's some history of cross-wiki abuse against me User:Bédévore/Harassment. There's currently another round of cross-wiki abuse, with threats + sexist & sexual comments on my user pages by single-purpose accounts. I reported the abuses on the steward page for global lock... and nothing happened. I know we are all volunteers but I don't understand what's taking so long. Maybe I should report the abuses on another page to get a reaction about it. Please tell me where. Best regards, Bédévore (talk) 12:50, 15 April 2021 (UTC)

@Bédévore: Stewards are regularly in IRC, and responsive, so you could try popping in with Freenode's webchat function if you don't have an IRC client. Alternatively you can use their other contact means, like SN and their Special:Contact/Stewards. Or talk to your wiki's CU and see if they can escalate to stewards, they have email contact details to raise issues.  — billinghurst sDrewth 08:35, 5 May 2021 (UTC)
Hello @Billinghurst: thanks for the tips. Bédévore (talk) 10:11, 5 May 2021 (UTC)

Universal Code of Conduct Phase 2: m:Talk:Universal Code of Conduct/2021 consultations/DiscussionEdit

The Universal Code of Conduct (UCoC) provides a universal baseline of acceptable behavior for the entire Wikimedia movement and all its projects. The project is currently in Phase 2, outlining clear enforcement pathways. You can read more about the whole project on its project page. There are consultations ongoing at several projects about key discussion topics.

To seek input from participants of projects without individual on-wiki discussions, m:Talk:Universal Code of Conduct/2021 consultations/Discussion is accessible now in several languages and accepting input in any language.

Please let me know if you have any questions. Xeno (WMF) (talk) 16:32, 15 April 2021 (UTC)

Suggested ValuesEdit

Timur Vorkul (WMDE) 14:08, 22 April 2021 (UTC)

Wikimedia thematic organizationsEdit

Where can I change the contents of the section "What are Wikimedia thematic organizations?"? There is a missing word that I want to add. Currently, only the section's title can be changed. Thanks! Nguyentrongphu (talk) 15:53, 28 April 2021 (UTC)

@Nguyentrongphu: If the issue is in the English version, then either just make the edit, or suggest an edit page's talk page. If it is one of the translations, then use the translation tool on the page.  — billinghurst sDrewth 08:39, 5 May 2021 (UTC)
billinghurst I clicked "edit" the page, but only the title can be changed. The contents of the section is nowhere to be found. Can you give me a link to edit the contents of the section "What are Wikimedia thematic organizations?"? Thanks! Nguyentrongphu (talk) 11:05, 5 May 2021 (UTC)
@Nguyentrongphu: In English I see a standard wikipage with a standard edit tab and it opens to a standard edit page, there is not protection, whether I am logged in or as an IP address. If you are not in English, then you need to be using the "Want to help translate? Translate the missing messages" link.  — billinghurst sDrewth 12:26, 5 May 2021 (UTC)
billinghurst I don't see this part in the edit page: Wikimedia thematic organizations are independent non-profit organizations founded to support and promote the Wikimedia projects within a specified focal area. Like the Wikimedia Foundation, they aim to "empower and engage people around the world to collect and develop educational content under a free license or in the public domain, and to disseminate it effectively and globally". Thematic organizations use a name clearly linking them to Wikimedia and are granted use of Wikimedia trademarks for their work, publicity, and fundraising. Currently, 2 thematic organization exists.
Recognition from the Affiliations Committee allows a group to apply for using the Wikimedia trademarks, access to unique grants, and additional affiliate support; however, recognition is not required to do any of the work a typical Wikimedia user group, and groups are encouraged to consider that model if they do not yet meet the requirements of a thematic organization. Nguyentrongphu (talk) 12:35, 5 May 2021 (UTC)
@Nguyentrongphu: are you on the Wikimedia thematic organizations/en or the Wikimedia thematic organizations/vi Vietnamese version? If /vi, then there should be a translate labelled tab.  — billinghurst sDrewth 13:05, 5 May 2021 (UTC)
billinghurst I'm not trying to translate anything here. I want to edit something in the italicized part above, but it is nowhere to be found in the edit page. Although, this is also a problem for those who want to translate. In the translate page, only the titles can be translated and not the contents. Nguyentrongphu (talk) 13:25, 5 May 2021 (UTC)
Ah okay, you haven't deciphered how to interpret {{TCT}} templates (template name it imports and the switch). The text is in Template:Wikimedia movement affiliates/Content though I would not recommend tinkering with that template unless you are really good at templates and not f*cking them up. Much better to suggest some text here.  — billinghurst sDrewth 13:49, 5 May 2021 (UTC)
billinghurst Looks complicated. Anyway, however, recognition is not required to do any of the work of a typical Wikimedia user group. I believe the word "of" is missing. Nguyentrongphu (talk) 16:30, 6 May 2021 (UTC)
  Done  — billinghurst sDrewth 22:23, 6 May 2021 (UTC)

Closing the gap to and between the base communitiesEdit

Read only time on 05-May-2021 at 06:00 AM UTCEdit


Some services will be in read-only for a short time on 2021-05-05 at 06:00 AM UTC.

During the restart time (expected to be around 60 seconds or so) all the components and extensions that use the x1 database will be read-only.

Things that might experience some issues when creating new writes:

  • New short urls cannot be created
  • Email bounces from lists might not get recorded
  • There might be issues with new translations
  • New items on the notification list might fail, some notifications may not be delivered
  • Reading lists might not record new items added to "bookmark" or "read it later" feature

Details: T281212 & T281375

A banner will be displayed on all wikis 30 minutes before this read-only time.

-- Kaartic [talk] 18:40, 4 May 2021 (UTC)