[bawolff]

It should be noted that WMF had a significant management shake up since then. The lila era management is basically all gone.

However im glad that this is the article posted, and not the rediculous "cancer" article that compares amount spent on servers when the site was smaller, slow and constantly going down to the modern day site when its fast, stable and a bigger site.

[Disclaimer: used to work for wmf. Do not anymore. My opinions are solely my own]

Edit: i initially just read the intro to the article. The article is long and says a lot more things as well other than just lila/knowledge engine.

[zozbot234]

It's more than a bit ironic that people complain about the knowledge engine initiative, when Wikidata is now one of the more successful Wikimedia projects and pretty much the same thing.

[bawolff]

Wikidata already existed during the knowledge engine fiasco, so if it was the same thing im not sure what they were doing.

I suspect a lot of KE's failure related to inconsistent messaging and project scope. I'm pretty sure even now, nobody knows what KE was planned to be or what its value proposition would be, or "what" it actually is beyond buzzwords

[zozbot234]

The first phase of Wikidata (the MVP so to speak) was purely as a common repository for interwiki links, not a general knowledge engine/knowledge graph. The latter is a lot more recent.

[bawolff]

General statements were added in 2013 which is still before KE.

The query service (the SPARQL endpoint) is probably the most KE like part, and that was later, in 2015. Still before KE but much closer. I think (hard to say because nobody gave a straight answer) that KE was more going to be about federation and searching non wikimedia (Free) resources.

[sundarurfriend]

What were the problems with the "lila era management" in your opinion? The article made it sound like there were just some well intentioned decisions received poorly, that led to that person being sacked. But there's often a festering hoard of management issues beneath what's apparent to outsiders like us, and I'm wondering if that was the case here.

[bawolff]

Ultimately there was a lack of confidence in Lila's leadership. The how & why doesn't matter if you can't effectively lead people. See also https://www.mollywhite.net/wikimedia-timeline/

[wombatmobile]

> The how & why doesn't matter if you can't effectively lead people.

On the contrary, how could any CEO lead an organisation that is dysfunctionally structured in the first place?

I'm not commenting on Lila, of whom I know nothing. Rather, I'm looking at the structure of WMF and the Wikipedia product and seeing disconnects and ambiguities.

Wikipedia's value comes from its content, which is provided by contributors who are not paid.

Wikipedia's budget goes to WMF staff, and capex approved by... I don't know whom. But who directs the effort of staff? This is where the org's structure appears to have resulted in years of frustration and dysfunction for all concerned.

How is WMF's strategic direction determined?

What is the role of the founder and public face, Jimmy Wales?

What is the role of the WMF CEO?

What is the role of the Funds Dissemination Committee?

What is the role and influence of large doners?

How are conflicts resolved between these levels of stakeholders?

Those are the key questions that I would want to understand before considering whether I would donate to WMF.

I've just started by reading the below articles. If anyone can add to this list, including with personal observations by insiders, I'd be grateful.

Wikipedia Foundation exec: Yes, we've been wasting your money

https://www.theregister.com/2013/10/08/wikipedia_foundation_...

Wikimedia timeline of events, 2014–2016

https://www.mollywhite.net/wikimedia-timeline/

Wikipedia:Wikipedia Signpost/2016-02-10/Special report

https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2...

[bawolff]

These questions have the same answer as basically any other company. Most internet companies do not pay their content creators or if they do (e.g. youtube) they pay them a relatively small pot of the pie that is really only significant for the super popular people (whether that's ethical is totally different question).

Anyways, some of the info you seek is probably at meta: https://meta.wikimedia.org/wiki/Grants:APG/Funds_Disseminati... https://meta.wikimedia.org/wiki/Strategy

Jimbo is largely a public figure-head. He has limited influence on WMF priorities beyond what any other board member would have. He has significant moral influence in the english wikipedia volunteer community.

[To be clear, im just answering questions. I dont care whether you donate. Do with your money what you think is best, its none of my business]

[wombatmobile]

Thanks bawolff.

I read this wikipedia article about KE.

https://en.wikipedia.org/wiki/Knowledge_Engine_(Wikimedia_Fo...

I infer from the article that KE was, in part, a defensive initiative to stop google leaching traffic away from wikipedia.

It sounds like JW was a key strategist, and that wikipedians were left out of the loop, which lead to management turmoil within the ranks.

It sounds like a difficult situation, to fight an external threat using internal broadcast democracy, when you know the information is going to go public right away.

Can you imagine Gates, Bezos, Jobs or Google consulting 50,000 employees every time a competitor looked like gaining traction?

I can imagine the internal transparency and consultation practiced at WMF could be beneficial for many aspects of the operations. But at a strategic level, particularly when threatened, that model could have liabilities for an organisation - particularly one which commands such a significant place in what is a lucrative commercial domain.

To complicate matters, the funding for KE was coming from an external grant source, which may have had its own agenda.

Difficult.

[ShakataGaNai]

People should donate to where they think it will do the most good. No NGO is perfect.

[ Same disclaimer :-) ]

[MattGaiser]

Did someone think that the higher server bill meant you guys were embezzling money?

[dx87]

The article claimed that they were wasting money on high paid executives and software that nobody used, meanwhile the pages are maintained by unpaid volunteers. I don't know if it's true or not, but I've also never donated money to them, so I don't really have a dog in the fight.

[wombatmobile]

What was the work culture like inside of wmf?

Gender balanced?

Cooperative?

I'm curious because my experience of editing, and that of other people I heard from at an editor meet up, was that the culture is heavily male dominated, and unnecessarily adversarial in a bullying way. That said, the CEO at the time (2007 - 2014), Sue Gardner, seemed enlightened, progressive, smart and highly capable.

[bawolff]

> Gender balanced?

Fwiw, when i was there, engineering teams generally had more men than woman. No tech teams and management tended to have more equal representation. However i would say the gender demographics of eng teams were fairly on-par for the industry at large. I never saw any sexism, but as a man, if it was present it probably would not be directed at me, so i may just not have noticed.

[toomuchtodo]

<This sentence removed and more context in a reply below because it was not accurate enough to stand on its own>. Is the non profit to gate people from editing if arbitrary diversity requirements aren't met?

The best part about Wikipedia is that anyone can edit it, anyone can run it, and nobody can lock up its content (for whatever reason). Look no further to Mozilla to see what happens when a non profit fails to meet it's mandate and is stuck with "peacocks" (folks who prioritize status or signaling > substance) instead of practitioners.

[geofft]

> The best part about Wikipedia is that anyone can edit it, anyone can run it, and nobody can lock up its content (for whatever reason).

This is true in the same sense that anyone can track down and fix bugs in the Linux kernel.

Now, I have certainly done so, but I've done it as an employee for a big and profitable company who had plenty of time to spend on it. Wikipedia is similar - it is superficially open to everyone, and sure, you can probably add some detail to an article about William of Normandy's first cousin once removed just fine, but if you're trying to document, say, whether or not there was a death camp in Warsaw (https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2...), you're up against people who have more time and resources than you do.

[bawolff]

I think OPs point is if you don't like the rules on wikipedia, you can download a dump of articles run a mediawiki instance, and make your own wikipedia.

Which in practise also requires skill and effort. Even importing the xml dump file into a mysql database in reasonable time requires special skill when we are talking data in the 100 gb range before decompressing.

https://dumps.wikimedia.org/enwiki/20200920/

[bawolff]

Sue was the ceo (executive director) before lila. Lila was the ceo where things went south.

[toomuchtodo]

Excuse my sloppy comment. Lila did a terrible job, but Sue was running Wikimedia while there was a slow decline in editors since the mid 00s. Having donated to Wikimedia previously, I no longer do, because senior leadership (across "administrations") is more concerned with issues completely disconnected from their core role ("To unlock the world's knowledge"). Those dollars go to the Internet Archive now (which stores copies of Wikipedia distinct from Wikimedia's control).

Non profit dollars are hard to come by, and the greater issue and frustration is watching orgs like Mozilla and Wikimedia incinerate them on causes that are not their mission (in stark contrast to efficient, nimble orgs like Let's Encrypt and the Free Law Project).

[bawolff]

There is some research to suggest that (en) wikipedia's growth pattern (a peak followed by slow decline as beurocracy becomes ossified) is common in similar projects. So maybe they didn't do anything wrong they just failed to fix it. That said i dont think anyone really knows what to do about the editor decline.

[wombatmobile]

> i dont think anyone really knows what to do about the editor decline.

How was this problem presented and discussed inside of WMF?

As an editor, I had an opportunity to discuss it with Sue Gardner at a meet up with other editors. The feedback from other editors was so strong, I'm baffled that WMF couldn't address it.

Although, I understood from Sue that WMF's role was only as a software company that made tools. She said quite clearly that the editor community ran itself.

But if dwindling editor participation is an existential threat, surely that's a core WMF problem? I'm curious what this looked like from inside WMF.

[toomuchtodo]

> That said i dont think anyone really knows what to do about the editor decline.

Pay them? Tens of millions of dollars in donations, what's the excuse not to?

[wombatmobile]

"Gardner was instrumental in raising Wikipedia's warchest and WMF's staffing. In 2011/12 - the last year for which figures are available - the Foundation raised $38.4m, up from $5m in 2007/08"

Wikipedia Foundation exec: Yes, we've been wasting your money Editors should get dosh, bureaucrats get too much, says outgoing fundraising chief

https://www.theregister.com/2013/10/08/wikipedia_foundation_...

[TheNorthman]

> Is the non profit to gate people from editing if arbitrary diversity requirements aren't met?

Notice how you're the one suggesting `diversity requirements' and then proceed by going on a tirade about it. There are many things to be done about diversity and not all of them requires coercion.

[toomuchtodo]

The comment I replied to mentioned gender about a cohort of participants ("I'm curious because my experience of editing, and that of other people I heard from at an editor meet up, was that the culture is heavily male dominated"), hence where my statement comes from. I want to see quality work regardless of who is doing it, and the folks doing this are uncompensated volunteers.

EDIT: @TheNorthman I cannot reply to your comment as I'm throttled by HN. Put bluntly, the gender of contributors does not matter and shouldn't even be considered. Let contributions stand on their own merit.

[wombatmobile]

You are right, toomuchtodo, that anyone can edit Wikipedia. If you want to see "quality work regardless of who is doing it" then it would pay to look deeper into the dynamics of gender.

Why?

Male editing culture tends to work differently to female editing culture.

How?

One tends to be hierarchical. The other tends to be cooperative.

It's possible to develop a system that uses both of these modes to sharpen the editorial process, but that doesn't happen through self organisation when the starting point is a large gender imbalance.

Resolving this conflict is one of the biggest challenges faced by Wikipedia.

The difficulty is, however, that WMF isn't directly involved with Wikipedia content, which comes from the community. WMF is just a software company responsible for the tools the community uses for publishing.

[TheNorthman]

Sure. I'm not disputing them bringing up gender diversity, I'm simply stating that nobody but you were talking about requirements and keeping people out to meet certain `diversity requirements'. You were, simply put, attacking your own straw-man.