Wikimedia Foundation Board noticeboard

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search
Board of Trustees Board noticeboard Archives
Welcome to the Board of Trustees' noticeboard. This is a message board for discussing issues related to Wikimedia Foundation governance and policies, and related Board work. Please post new messages at the bottom of the page and sign them.
  • For details of the Board's role and processes, see the Board Handbook.
  • Threads older than 90 days will be automatically archived by ArchiverBot.

13 years[edit]

I would like to draw the Board's attention to this thread started by @Guy Macon:. Guy sets out the problem - and the lack of effort from the Foundation in addressing it - very clearly. I have asked if anyone on the Board has a responsibility for equalities issues such as this. I would expect an organization such as this to have a lead member assigned for equalities. DuncanHill (talk) 19:48, 10 July 2019 (UTC)

Hi there. The Board recently approved the WMF Medium Term Plan, where Platform evolution is very much a top priority. The Core Platform Team explicitely includes increasing accesibility and usability among its outcomes. Kind regards, Raystorm (talk) 18:36, 11 July 2019 (UTC)
@Raystorm: That's a 3 to 5 year plan. So hopefully 18 years after becoming aware of the problem WMF will have a fix. I don't think that's an acceptable timescale. Pinging @Guy Macon:. DuncanHill (talk) 21:38, 11 July 2019 (UTC)

On 03 February 2006, it was reported to the WMF that our CAPTCHA system discriminates against blind people. See phabricator T6845. This appears to be a direct violation of the Americans with Disabilities Act of 1990 and leaves Wikipedia open to the possibility of discrimination lawsuit.

In particular, National Federation of the Blind v. Target Corp. was a case where a major retailer, Target Corp., was sued because their web designers failed to design its website to enable persons with low or no vision to use it.

So why, after 13 years of inaction, do we not have a set of software requirements (including a testable definition of "done"), a schedule with milestones and updates, and budget and staffing information for solving this?

And no, I will not accept any proposed "solution" that lacks the name of an WMF employee who has been given the assignment of fixing this, a budget that says how much the WMF expects to spend on solving this, a deadline that say how long the WMF expects it to take to solve this, and a way for an independent third party to look at the results and verify whether the requirements were met.

Regarding hiring someone else to fix this, I would very much like the idea to be given careful consideration rather than being dismissed out of hand. The WMF is great at running an encyclopedia. Nobody else, anywhere on earth, even comes close. However, running an encyclopedia does not magically confer the ability to create high-quality software, and the WMF has a pretty dismal track record in this area (Examples: Visual Editor, Flow, 13 years of failing to making an obvious but boring improvement to accommodate blind people.) I realize that this will anger some people, but why should it? Olympic-level athletes don't get angry when you tell them that their athletic ability does not magically confer the ability to repair automobiles or do astronomy.

Comments from phabricator:

  • "This doesn't just effect addition of external links, it also prevents new users from registering, requiring them to use ACC to request an account."
  • "There is no one currently assigned to this, so no one is taking it upon him to fix this at this moment. It's also not something that any team at the foundation is responsible for, so it's not likely to be prioritized from that end."
  • The only thing stopping us from having an audio captcha is that nobody's put the work into implementing it yet." --Source: Chief MediaWiki developer as of 2008
  • "So the question is why has work not been put aside to fix an issue of recognised high importance that will, 13 years after first being raised, resolve an issue that results in us discriminating against people who are (in many jurisdictions) a legally protected minority?"

---Guy Macon (talk)


@Raystorm: Thanks for the update. Would you be so kind as to answer the following questions?

  • What is the name of the WMF employee (or employees) who has been given the assignment of fixing this? Alternatively, when will that employee be named and by who?
  • What is the budget -- in other words how much does the WMF expect to spend on solving this? Alternatively, when will the budget be created, and by who?
  • What is the deadline -- in other words how long does the WMF expect it to take to solve this? Alternatively, when will the deadline be decided on, and who will make that decision?
  • Where will the software requirements be published, and how does the WMF propose that an independent third party can look at the results and verify whether the requirement were met?

I can help with that last bit by lining up several people who are vision impaired and use different screen readers and have them test the solution if and when it is ever completed. --Guy Macon (talk) 22:51, 11 July 2019 (UTC)

  • Since, a Trustee says that CPT has the reponsibility to look into this issue:- Paging @Tim Starling, Legoktm, and Anomie:-Can you answer the above queries and/or share anything else around the locus. Regards, Winged Blades of Godric (talk) 08:16, 14 July 2019 (UTC)
    I can't speak about prioritization or resourcing. But I think trying to fix CAPTCHAs is just a dead end. They're fundamentally broken, bad UX, and don't do that good of a job of keeping spambots out. So-called "accessible" ones are broken even more. Except based on our last testing, they do stop some spambots, and that was a significant amount of them (I can't find the link to when we last tried this, I'll look later). So I'd rather figure out some % of spambots they're blocking, figure out how to improve our antivandalism/antispam tooling to improve productivity or success by that same or better % (so in theory we're not adding an additional workload to patrollers), and then just turn off the CAPTCHA. Legoktm (talk) 17:07, 14 July 2019 (UTC)
    I think you mean mw:Extension:ConfirmEdit/FancyCaptcha experiments, which was 5 years ago, but that test didn't prove anything. Most of the additional bot edits were easily prevented by abusefilter, they were just not yet covered by existing rules. We were supposed to repeat the test after adding at least some minimal abuse filters (which is now easier with Global AbuseFilter). Nemo 18:56, 14 July 2019 (UTC)
    Yep, that was it. I think repeating it in 2019 for some pre-determined time with sysops prepared to add more AF rules would be a good start. Legoktm (talk) 15:58, 15 July 2019 (UTC)
[Edit Conflict] Any discussion of how to fix this is a distraction and a huge waste of time. That discussion should happen after the WMF decides to fix it and gives me straight answers about staffing, budget, and schedule. I don't care how the WMF stops breaking the law. All I care about is that the WMF does stop breaking the law.
Regarding what percentage of spam bots CAPTCHA stops, imagine for a moment that we suddenly have reason to create a National Federation of the Blind v. Wikimedia Foundation page with content similar to Federation of the Blind v. Target Corp. Do you think that the court will look favorably on a "but it stops X% of spam, reducing the workload of our patrollers!" defense to the charge of purposely discriminating against blind people? Also imagine that you tried spamming Wikipedia and the CAPTCHA stopped your spam bot. Why wouldn't you simply buy a spam bot that defeats the CAPTCHA? Just counting blocked spam doesn't tell us how often the spammer gives up and how often he adapts to the countermeasures --Guy Macon (talk) 19:31, 14 July 2019 (UTC)
Sorry, I have no interest in engaging if you're going to be making veiled legal threats about theoretical lawsuits. Legoktm (talk) 15:58, 15 July 2019 (UTC)
I deny making a legal threat, and invite you to report me at ANI if you seriously think I have made a legal threat. While you are at it, you might want to report everyone on our legal team, because every one of them has, at one time or another, said something like "we should do X or stop doing Y in order to comply with the law and/or reduce our risk of being sued."
Did you at the very least "get it" when I told you that your "here is how I think we should fix this" comments are not helping? Again I tell you, any discussion of how to fix this is a distraction and a huge waste of time. That discussion should happen after the WMF decides to fix it and gives me straight answers about staffing, budget, and schedule. --Guy Macon (talk) 15:48, 16 July 2019 (UTC)
  • Since the CAPTCHA was first deployed, I have supported simply removing it. I filed T6845 as a position statement along these lines. The task description has since been edited to say that this would "invite too much abuse for our communities", which I don't believe. "Too much" is subjective. I think we should address spam and vandalism by other means. At the time, I suggested configurable heuristics as an alternative to CAPTCHA, and this has since been implemented in the form of AbuseFilter. So what is left to do, on the technical side? In my opinion, it is a policy question, not a technical question. -- Tim Starling (talk) 00:24, 15 July 2019 (UTC)
    @Tim Starling: That was me, and it is based on conversations I had with I think Brian, where they tested this and had to roll it back because it was holding back way more crap than they had expected... I can't exactly find the history of that, but i'm pretty sure I remember this correctly. —TheDJ (talkcontribs) 08:19, 16 July 2019 (UTC)
  • My name is Evan Prodromou; I'm a product manager in the Core Platform Team. I'm probably about to step in it, but... I think we're missing out on an opportunity here with the Captcha. Google, for example, uses reCaptcha as a mechanism for training image classification AI for self-driving cars ("Click all the images with cars/crosswalks/traffic lights/storefronts"). I wonder if there are classification tasks for media on Commons (images, video, and audio) that we could do with a Captcha? Perhaps we could even make a widget that 3rd-party websites could include, similar to how they use reCaptcha...? I'm going to follow up on this issue, both for the immediate and long-term. Feel free to contact me at eprodromou *at* wikimedia *dot* org for follow-up on CPT. --EvanProdromou (talk) 15:06, 15 July 2019 (UTC)
    That's phab:T34695. Legoktm (talk) 15:58, 15 July 2019 (UTC)
    EvanProdromou and Tim Starling, be careful that you don't make the error of talking about how to fix the problem while the WMF has not assigned anyone to fix the problem. I have been talking about this for years, and every time the conversation gets sidetracked into discussions about how best to fix the problem by people who have zero authority to actually change the Wikimedia software. Following that path leaves us without any actual improvement being made. We don't have a "we don't know how to fix this" problem. We have a "The WMF refuses to assign someone the job of fixing this" problem. Once we get someone who's job it is to fix this, we can discuss how to fix it with that person. Until then any discussion about how to fix this leads us way from solving the real WMF inactivity problem. --Guy Macon (talk) 16:12, 15 July 2019 (UTC)
    Disabling FancyCaptcha takes about 5 min of developer time, if the community decides so. Nemo 16:22, 15 July 2019 (UTC)
    Right. I don't know why Winged Blades of Godric is pinging me and other engineers. I've made my opinion on the matter perfectly clear, but I'm not empowered to change it. If there's a legal argument for disabling the CAPTCHA, then talk to the legal team. If their advice is to turn it off, it could be done in a few minutes. -- Tim Starling (talk) 05:45, 16 July 2019 (UTC)
    I have sent this to the legal team at least 3 times over the years, and have never received a reply of any kind. I encourage anyone interested to contact them and see if they have better luck.
I strongly oppose pinging or contacting any engineers or developers. Unless you are dealing with one person working out of his garage, engineers don't get to decide what projects to work on.
As I have repeatedly stated, we don't have a "we don't know how to solve this" problem. We have a "WMF management has not given anyone the job of solving this" problem. We need someone in WMF management to name an individual who's job it is to fix this along with a budget and a schedule. --Guy Macon (talk) 15:37, 16 July 2019 (UTC)
  • EvanProdromou: the existing captcha is already reasonably good for accessibility, since it is easily defeated by OCR. I tried the OCR extension described on this page against our existing captcha, and it worked 4 times out of 5, with short solving times. If the idea is to optimise for a false sense of security for editors while still allowing blind users to create accounts, the existing captcha is a pretty good solution, and presumably much better than image classification. The main problem with it is that it is unwelcoming. Installing obscure Chrome extensions is a barrier to entry. That would not be improved by switching to image classification. -- Tim Starling (WMF) (talk) 06:23, 16 July 2019 (UTC)
    That's all fun and all, but before we know it we end up in another WMF mega project that drowns in its own ambitions.. KISS.. —TheDJ (talkcontribs) 08:20, 16 July 2019 (UTC)

So, will anybody from the WMF be answering the questions at the top of this section any time soon? (...Sound of Crickets...) --Guy Macon (talk) 01:08, 17 July 2019 (UTC)

(...Chirp...) --Guy Macon (talk) 01:25, 2 September 2019 (UTC)

Fifth column[edit]

On 26 April 1989 the official organ of the Chinese communist party spoke out against democracy. As the thirtieth anniversary of the massacre approaches Wikipedia has been cut off - we've not heard from Anna Frodesiak since 23 April. In this country, expect people like Nigel Farage to continue speaking out against democracy - Leavers secured less than 30% of the vote in the EU election, but to hear him speak you would think they had secured a landslide.

Proposal to amend bylaws regarding Founder's seat vacancy[edit]

The current bylaws, section 3 (F) reads as follows:

Community Founder Trustee Position. The Board may appoint Jimmy Wales as Community Founder Trustee for a three-year term. The Board may reappoint Wales as Community Founder Trustee for successive three-year terms (without a term limit). In the event that Wales is not appointed as Community Founder Trustee, the position will remain vacant, and the Board shall not fill the vacancy.

I propose amending the bylaws such that if Jimbo is unwilling, unable, or unavailable to fill the position, that it convert to a community elected seat. In fact given the significance of the position, I would propose that it convert to two community elected seats.

Just to clarify, I have no desire for the Founder's seat to become vacant. It merely occurred to me that it would be a Bad if the board were to shrink in such a fashion, losing such a critical voice for the Community. Alsee (talk) 20:45, 11 September 2019 (UTC)

Bulgarian Wikinews: Please do not delete[edit]

Regarding bg.wn I ask you to consider recruiting volunteers from BG.WP and any relevant affiliates to delete existing fake news and check new submissions. I think doing this would be in line with the Wikimedia Foundation policies on language inclusiveness and supporting growth of free content in foreign languages.

I have started this discussion in a section titled 'Question from Gryllida' at the 'bg.wn' page which I linked above. --Gryllida 22:43, 15 September 2019 (UTC)