That doesn't include salaries of the people who run it - and WMF doesn't just rent servers from AWS, like it's in fashion now, they run everything by themselves from the bare metal up, and they run it over multiple data centers over the globe. This probably also doesn't include the cost of the hardware (this likely is part of "depreciation and amortization" since the cost of a server usually is amortized over several years).
And if you've been on any of those conferences (it's not hard to get in, there are a lot of people invited) - they aren't exactly luxury affairs, they usually happen in a library or university and constitute of a crowd of geeks sitting in groups arguing or staring at computer screens, or sometimes presenting slides to other geeks. Those aren't exactly lavish affairs. Of course, some people get their hotel bill and travel paid for, but given those people help building such enormous site as Wikipedia, it's not that huge of an expense.
This seems like the typical HN meme of looking at something and assuming the technical piece is all there is to it.
Wikipedia's biggest challenge over the years hasn't been building or hosting its website. It's been convincing the world that an encyclopedia that can be edited by anyone on the Internet is a credible and neutral source of information.
One of the ways you do that is by speaking at conferences, traveling to meet with people, etc.
The only reason I checked was because of Jimmy Wales' past controversy regarding use of the foundation's credit card [0]:
> Wool wrote that Wales had asked the foundation to reimburse him for costly items like a $1,300 dinner for four at a Florida steakhouse. Wool alleged that at one point Wales was short on receipts for $30,000 in expenses before settling the matter with the foundation's lawyer and paying the organization $7,000.
> Wool added that Wales' foundation credit card was taken away in 2006.
Are you suggesting that in the face of existing encyclopedia companies, generations of people who had always used print encyclopedias, and the general public still figuring out what the internet is all about, Wikimedia should be restrained from answering criticisms or making a case for their value?
Eh, Wikipedia is 20 years old, and has been a top 20 site for the past 15 years. At this point, they don't have much serious competition. Britannica stopped printing physical encyclopedias in 2012.
The WMF can certainly make their case and answer criticisms, but there are cost-efficient ways of doing that. I don't see much value in funding parties, conferences, etc. for a product that has already dominated in most markets.
Most markets? That’s a very privileged view. Research by the WMF in 2019 shows incredibly low rates of awareness in large parts of the world. I doubly you will read it, nor to I seek to argue this with you, but here are the findings: https://meta.wikimedia.org/wiki/New_Readers/Awareness#Awaren...
Why do you feel the need to use ad hominems rather than simply making your point?
> Research by the WMF in 2019 shows incredibly low rates of awareness in large parts of the world
I wasn't claiming that Wikipedia is near peak global usage. My point was that there is no meaningful competition left, so usage should grow organically, as it did for EN-WP back when the WMF had a shoestring budget.
Can you identify some markets where Wikipedia is losing ground to an inferior competitor with more aggressive marketing? I'm not sure there are any.
The conferences are not to suppress competition, it's to enable cooperation. Wikis are very distributed projects, and conferences are the only place where people could get together, discuss things, exchange experiences and ideas, and so on. And no, it's not the same via teleconference or mailing lists, those do exist but it's completely different mode of operation.
Yes, in the US at least, wikimedia has been very effective. That doesn't seem like "most markets" to me. In fact, not so much the rest of the world at all, and even in the US, now does not seem like the time to stop and let that decline.
Maybe my wording was a bit strong, but the US is a pretty small minority of Wikipedia's usage. If we look at article counts, about 12% of articles are in English (5.8M of 49.3M): https://stats.wikimedia.org/EN/TablesArticlesTotal.htm
Right, so more than twice as much as #2 (2.3M), at least in December 18.
And more recently, the US has more than four times as many pageviews as #2 (4B to Japan's 1B)[0].
So no, I don't think the work is done outside the US, and I think maintaining a leading position doesn't happen automatically, so I don't think the work is entirely done inside the US, either.
How do they spend so little on internet hosting? They're one of the top 10 websites in the world. Are servers and bandwidth really that cheap? It's not like it's all text either - they serve images, audio, and video too.
Anyone know if someone donates hosting resources to them? I simply cannot imagine serving that much traffic on < $3m.
It's because a large portion of traffic to Wikipedia, that is a response to google serving a search request with results involving Wikipedia, is served to the client with a Knowledge Graph box. A large portion of viewers ingest the text of the Knowledge Graph box and don't click through to a page served by Wikipedia's servers.
Is that all that bad considering that Wikipedia is arguably one of the most important organisations on earth?
If it was as corrupt as the NRA (which is why NY is moving to dissolve them), then I would agree with the sentiment but a few million sounds about right (About 9k per employee)
[0] https://upload.wikimedia.org/wikipedia/foundation/3/31/Wikim... (page 13)