Jump to content

Wikimedia Foundation Inc. is corrupt and bad

From Meta, a Wikimedia project coordination wiki
(English) This is an essay. It expresses the opinions and ideas of some Wikimedians but may not have wide support. This is not policy on Meta, but it may be a policy or guideline on other Wikimedia projects. Feel free to update this page as needed, or use the discussion page to propose major changes.
Translate

Wikimedia Foundation Inc. is corrupt and bad. The organization has bloated to become unwieldy, unaccountable, and it has little to show for the hundreds of millions of dollars it has wasted and continues to waste.

Fundraising

[edit]

Wikimedia Foundation Inc. should ideally be very small, be limited to managing the trademarks and technical infrastructure, with chapters around the world should be doing most of the work and have almost all of the budget. Having a single non-profit foundation based in San Francisco with a budget of 150,000,000 USD.[1] and a staff size of north of 500 full-time employees[citation needed] is ludicrous.

The Wikimedia Foundation raises so much money by intentionally misrepresenting the state of its finances and lying to donors.

If you talk to friends, neighbors, colleagues, and family members, they all have the impression that Wikipedia is desperately in need of money to continue operating.

Why do they think this? Because Wikimedia Foundation Inc. has spent millions of dollars in cash (and billions of dollars in screen estate) to run intrusive and obnoxious advertising on Wikipedia telling readers that this is the case.

Much of this money goes into the Endowment, which is a slow-brewing mess, with way too much money and way too little oversight.

The English Wikipedia agreed that the WMF's fundraising emails were misleading by a !vote of 45-3 in September 2022.[2]

Strategy process

[edit]

Closing for the Benefit of Openness? The case of Wikimedia’s open strategy process. The researchers examined the 2010 strategy process, including interviews with many staff and community members. The Foundation wanted to run an open and inclusive process, but the researchers found that the Foundation utterly failed to do so. There was an open invitation for the community to submit comments, but all decision making was kept under tight control of Foundation staff and other Foundation-appointed representatives. The researchers found that the outcome of the process became progressively disconnected from community input at every stage. The study predated Strategy/Wikimedia movement/2018-20, but the same conclusions apply. The newer strategy outcome is almost entirely a fabrication of the individuals put in charge of the process. The number one strategy item to get more money for the Foundation, hardly a community priority. However more significantly, one strategy item proposes a radical agenda to undermine community content-quality policies such as Notability and Reliable Sourcing. That has intense and near unanimous opposition among the community, but it was unilaterally established as a strategy item by the fringe individuals running the process.

Specific problems include discussions scattered in dozens of subpages, which are difficult for any volunteer to follow and staffers paid thousands of dollars to respond to comments which are ultimately ignored in favor of slides made by some Wikimedia Foundation Inc. exec in Google, which the board signs off without any critical feedback.

Reading/Web team

[edit]

The Reading/Web team is constantly enraging communities with poorly thought out and poorly implemented software, such as Gather, burning community goodwill and making it way more difficult to implement ideas in the future. There are a worrying number of self-merges and quasi-self-merges. For example, with a small team, the manager, or even team members, can make bad changes that people outside the team would never approve. When you're the manager, people won't say no, even if the changes are bad. This a problem with +2 rights generally, but specific teams are especially damaging.

They are also wasting effort on Vector 2022 when nobody asked for Vector to be versioned.

Plan to replace Wikitext

[edit]

In 2011, the Foundation decided on a strategy "to deprecate wiki syntax" with "The long-term goal of...HTML-only wikis".[1] This resulted in the Foundation building Visual Editor and Flow with no native support for wikitext. They instead rely on an intended-to-be-temporary Parsoid engine to perform round-trip translation of Wikitext into a variant of HTML and back. This strategic decision was made without consulting or informing the community. This round-trip translation system is enormously complex, fragile, buggy, and slow. It has resulted in the single most disruptive Foundation time-sink between 2011 and 2022 and beyond.

Feature requests that should be implemented but aren't because of the above wastage

[edit]
  • Vector or rasterized image editor for a media repository.
  • Automatic suggestions for categories when making a new page.
  • Visibility into what users are searching and not finding.
  • A blame feature, so that finding when text was added to an article does not require manual bisection.
  • Better tools for managing categories.
  • Native implementation of numerous workflows currently dependent on bots run by volunteers.
  • Slideshow support.
  • Dark Mode.

Unorganized notes

[edit]

This is a brain dump and needs to be cleaned up.

  • Focus on the post-Lila era to simplify?
    • Tho you could write so much more about how bad the organization is if you focused on just the Lila years.
  • Specific examples would be helpful.
    • Paying an SEO consultant company hundreds of thousands of dollars.
    • Useless branding efforts, paying companies thousands of dollars to create one-time use branding media.
  • Board is completely useless.
    • Has fundamentally abdicated its role, relies almost entirely on the CEO.
  • Board elections are corrupt.
    • Canceled during covid?????????? Nobody understands the seat selection process, keeps changing.
    • Election managers can't translate enough, can't anticipate having many candidates, etc.
  • What's not going okay?
    • Community Wishlist Survey 2022/Results is just sad, process is very clearly broken.
    • MobileFrontend sucks, should not exist.
      • A couple volunteers made MonoBook responsive and it's 1000 times better than MobileFrontend. Oh look, it shows categories and page notices and doesn't collapse all the text to ruin in-browser search and so on.
    • MediaWiki should natively handle different devices.
    • Mobile sites (en.m.wikipedia.org and friends) should not exist, why would we want this, they're so bad.
    • MobileFrontend extension re-implements core features for no good reason.
    • User identity issues continue, can't easily rename yourself, etc.
      • Focus now is on some bizarre plan to make IP addresses less accessible, but also continue to use them???????? Who knows, makes no sense.
        • Such a massive waste of time and energy and it's only going to be IP masking?????????
    • Very poor GIF support, can't paginate through, can't pause.
    • Sound logo stupidity.
      • All of the finalists are too long and they're all really bad.
      • A ton of time and energy wasted on this project that nobody asked for to support a very questionable use-case.
      • Main project page has broken English and lists like a dozen communications staff who worked on this. What were they doing exactly?
    • Flow is horrible; still infecting mediawiki.org and probably other wikis.
    • There's a mobile app but is it actually used for the one use-case it would be good for: uploading media to Commons?
  • What's going okay?
    • Reply tool is nice.
    • Pinging is nice.
      • Both came from actual community members, liked by the community, etc.
    • Page view stats exist.
  • The Funds Dissemination Committee got disbanded after a decade —— why?
  • Continued lack of transparency regarding directed and undirected grants.
    • Previous commitments from executives to publish grant agreements, but they've never been published.

References

[edit]