Hacker News Reader

Hacker News

Comment by toomuchtodo | Hacker News Reader

Full thread

toomuchtodo

As of September 2021, the Wikimedia endowment stands at >$100 million.

https://wikimediafoundation.org/news/2021/09/22/wikimedia-fo...

(sourced from https://en.wikipedia.org/wiki/Wikimedia_Foundation)

View on HN

It seems that this is approximately how much they spend each year and have ~500 staff.

I like Wikipedia but does it really take that many people and that much money to run it?

Having grown with startups from a very early stage to mid-stage, it seems pretty clear that 500 people never do 10x as much as 50 people, more like 2x if that.

That much spend seems unsustainable long term.

There has been some concern about their inflating spending for a while. I'm honestly not well versed in if this is some potential disaster or problem like it can be made out to be, but it does seem like Wikimedia will generally find ways to spend money they get, whether it's a good use of the funds or not.

https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_C...

It's one of the most visited site on the planet, and people are constantly writting on it. It's on the same scale of facebook, youtube or twitter, activity wise.

And I wager those sites have more than 500 people working on them, for a lot more money. And they are for profit!

Same scale? We're talking about almost static pages, with some videos / audio. At heart wikipedia is very simple and did not changed much from 15 years ago.

For comparison, OpenStreetMap has roughly $800k in reserves and their expenses are a couple hundred thousands dollars per year.

I agree there are probably more justifiable expenses at Wikimedia than that, but that there is a middle ground between the high water mark they're targeting and what other non-profits are operating with. The Internet Archive's annual budget is ~$10 million.

EDIT: @dylan604 (replying here due to HN throttling): Section 230 shields Wikipedia from most liability/exposure: https://www.eff.org/issues/cda230/successes/wikipedia

What's the legal expenses associated with Wiki? I could see some people wanting to sue for information linked to/about them. OSM probably doesn't have that same type of exposure.

Alexa rank for wikipedia is 13, OSM is 4550. It's not on the same planet. No, not on the same galaxy.

Wikipedia is not only read and written by the entire planet of humans, including the smallest 3rd world countries, it's also a bot feast, and busy powering APIs all over the world.

It got hundreds of links in some articles, which are automatically connected to all other articles in a gigantic graph of data. Oh, each article may also exist in a hundred languages, times each version, thanks to the history of all edits ever performed or suggested.

It's not just links though, now articles have plenty of metadata that generate connections, tables, ranks, toc, indexes, hierarchies, and listings as well.

All that is edited, rollbacked, debated, diffed and rendered almost on the fly for millions of articles, while respecting the permission system, and of course, the fun part, detecting the thousands of attempt at abuses by second.

In the blink of an eye. Because wikipedia is _fast_. It's one of the fastest site around.

Of course now it is also serving billions of images, sounds and videos. And the foundation is also maintaing:

    MediaWiki

    Meta-Wiki
 
    Wikibooks

    Wikidata

    Wikinews

    Wikiquote

    Wikisource
 
    Wikispecies

    Wikiversity

    Wikivoyage

    Wiktionary

Not to mention animating workshops all around the globes, helping NGO to provide offline wikipedias for poor countries and providing and maintaining the wiki as FOSS.

Thanks, it’s always so easy to forget how much work has to be done so that things seem easy and effortless.

you turned me from Wikipedia cynical to wanting a job there

"MediaWiki is free software. No guarantee or warranty of any kind is provided." Sounds about right!

We've run our internal corporate wikis on it for years without any major problems. It's a great example of free self-hosted software that just works.

Section 230 means you eventually win the lawsuits, but you still have to pay lawyers to defend them.

Correct me if I’m wrong, I was under the impression that Section 230 means most lawsuits are summarily dismissed. Without it you have the first amendment which as you say means you probably eventually win but it’s a much bigger hassle.

>>> As of September 2021, the Wikimedia endowment stands at >$100 million.

>> It seems that this is approximately how much they spend each year and have ~500 staff.

> It's one of the most visited site on the planet, and people are constantly writting on it. It's on the same scale of facebook, youtube or twitter, activity wise.

How much of that money and staff goes to actually running the site and maintaining the software? IIRC (and correct me if I'm wrong), a large fraction actually goes to other efforts.

Between 2005 and 2015-2016, page views grew by ~11x, articles by ~12x, hosting expenses ~33x, and total WMF spending grew by 1250x.

Hosting and actual sysop expense is quite limited. I think 15ish years ago when Brion Vibber started to do it full time, he ran the whole Wikimedia operation pretty much on his own, with an annual spend of low six digits (and that included his salary, hardware, hosting, the whole shebang).

Spending $100m a year on dev work, conferences, grants, D&I staff etc is a luxury, the site could be kept running for less than 1% of the cost (and dozens of big companies would do it for free just for the PR benefits).

Maybe the only other function apart from hosting/sysop that is essential to WMF's operation are the lawyers that prevent it from being shutdown in entire countries by lawsuit-happy individuals.

Well, the lawyer factor is probably most important.

A lot of people are willing to spend big money to shut down truth.

"It's on the same scale of facebook, youtube or twitter, activity wise." {citation needed}, as they say. Some parts of the Wikipedia "global" movement are so small, they didn't even realise for example, most of "Scots" Wikipedia was pure gibberish. The so called movement is a galaxy of tiny rocks revolving around one massive black hole, English Wikipedia, and even that project is so small as to feel quite parochial. As a result, there are backlogs everywhere, especially in important but difficult work, like copyright problems.

> I like Wikipedia but does it really take that many people and that much money to run it?

It takes a lot of people and money to raise $150m/year ;->

According to their 2019/2020 statement they spend 5M$ in "donation processing" and an additional 5M$ in wages for the donation campaigns.

10M$ in expenses to get 120M$ really isn't so bad.

[1] https://upload.wikimedia.org/wikipedia/foundation/f/f7/Wikim...

I wonder if that also includes all the work done simply to figure out the best way to solicit for donations?

Thanks! I checked on Charity Nav https://www.charitynavigator.org/ein/200049703 It says they spend 9 cents to raise every dollar, almost exactly your numbers. Overall, they are rated superbly.

Between content moderation, dev, security, infrastructure and such, then the support needed, I can easily see how Wikipedia can grow to 500 people. They're offering a stable product that has a presence in almost _every country in the world_, serving an insane volume of data. This is not a random startup, I'm actually amazed they pull it out with just 500 people

> Between content moderation....I can easily see how Wikipedia can grow to 500 people.

IIRC, nearly 100% of all content moderation is done by Wikipedia volunteer editors, so that's not actually an operational expense of theirs.

True, but I don't think you can keep the volunteers if you're in maintenance mode. The scale of volunteers needed requires some responsiveness, some support.

> True, but I don't think you can keep the volunteers if you're in maintenance mode. The scale of volunteers needed requires some responsiveness, some support.

Some, but not much I'd wager. My point is it's not like they have a big line item for paying full-time 100 moderators or something, so "moderation" doesn't belong with the other items on that list.

It's the best kept secret of Wikipedia that the volunteers generally despise the Foundation, and their long history of being unresponsive to their concerns and slow to act on their priorities while forcing the volunteers to adapt to what the Foundation finds things to spend all this money on, is a big reason for it.

With an Alexa rank of 13 and very real competitors that are large tech companies (every search engine now tries to answer your question instead of presenting Wikipedia anymore) seems to be quite appropriate to me.

>I like Wikipedia but does it really take that many people and that much money to run it?

Maybe we could ask Mozilla. /s

At least Wikipedia is not actively working on sabotaging their flagship product.

I wouldn't be so sure. Many flagship software projects launched the Foundation have been greeted with outright hostility by the volunteer community, precisley because they don't seem to be of much use, and in some case often hamper, the basic task of writing and maintaining English Wikipedia.

> I like Wikipedia but does it really take that many people and that much money to run it?

Just take a look at how fast people around the globe are editing it.

https://wikistream.toolforge.org/

Not that I like those tactics to beg for money but, yes, keeping it running while ensuring accuracy requires a lot of staff.

Keeping Wikipedia accurate is the volunteer's job. You can tell by how easy it is to find errors.

It is worth noting that the endowment is about half of Wikimedia's assets, and is not included in that tally. See https://www.dailydot.com/debug/wikipedia-endownemnt-fundrais... for more.

That's not a lot of money in context.

It's quite a lot of money. https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_C...

I read the page. It's very angry about Wikipedia's finances.

I still don't resent their asking and I still donate.

Wikipedia is a truly worthwhile service and I'm happy they have much more money than they need to scrape by.

You seem to feel that the amount of money they should have is just above what they need to get by. I think that is a bad way to run any operation.

My thoughts too, "living paycheck to paycheck" doesn't sound like it ensures Wikimedia's survival. We'd all be shouting "Save Wikipedia!" if they have a bad year were expenses are up and donations are drastically down and they end up going offline due to not having enough reserve to help it.

But Wikipedia is living paycheck to paycheck despite getting a lot of donations. That is what Guy Macon's article is complaining about.

Yeah. The paycheck is rising exponentially every year, and yet they find a way to spend it all every year.

The vast majority of those expenses are not Wikipedia.

Wikipedia's cancer is the Wikimedia organization.

Cancer? Wow. Personally, I expect they have a lot of good they can do in the world. And I believe they need resources to do it. What do you do that is so virtuous that you can call Wikimedia a cancer?

> What do you do that is so virtuous that you can call Wikimedia a cancer?

That's a very caustic statement, and it's not your place to judge, but if you're inviting me to soapbox:

Personally, I give my charitable dollars to Against Malaria Foundation to buy insecticidal bed nets. They've raised $342 million in total since their founding and distributed ~137 million nets. GiveWell estimates the program costs about $3000-5000 per life saved; or to put it another way, they've saved roughly 68,000 lives.

The WHO estimates that 409,000 people died of malaria in 2019 alone; there's a lot left to do here.

WMF has raised more than double that (~$800 million), over a similar time frame. Of that, Wikipedia's current hosting expenses and compensation for authors and editors totals something like $2.5 million annually.

If you think that's a good use of your charity dollars, great. We can agree to disagree. If you have any spare capacity, though, I would highly encourage giving to Against Malaria Foundation, or any of GiveWell's other top charities.

The $100m Endowment was their hedge against donations drying up, being a big enough cash hoard to allow them to stay online indefinitely if donations dropped to zero. It is now fully funded. I wonder how many people even realise this is the case, and will carry on donating under the false impression Wikipedia needs their help to survive.

Yeah. I always wonder if the complainers do the same deep analysis and whining about how for-profit companies spend their income, or if they just plunk down $120 for Xbox Game Pass and leave happy. Is Wikipedia worth $20 a year to me? Yeah. So they can have it and spend it how they like. Just like Microsoft can with my Game Pass subscription.

You aren't paying for a product. It's crazy people don't even realise that Wikipedia isn't free because they're just nice people who prefer prefer run on donations not capitalism. It's free because they don't own a single word of their encyclopedia. That was how it was set up. They knew they couldn't charge for it, because they weren't paying anyone to write it, and aren't offering any guarantee that you can trust it. It can be copied and hosted freely by anyone. Wikipedia can do what it likes with your money, except, of course, pay someone to write and maintain a trustworthy encyclopedia. Not by choice, by law.

> You seem to feel that the amount of money they should have is just above what they need to get by.

Not at all. The only reason Wikimedia do not have enough money to run Wikipedia in perpetuity is because of Wikimedia's exponential spending growth in non-"Pedia" projects.

I would love for them to spin off a Wikipedia-only endowment that funds Wikipedia forever. It would (likely) not need to take any future donations. The level of money Wikimedia take in is far, far in excess of "just what [Wikipedia] need[s] to get by."

The problem I have with Wikimedia's fundraising is that it is disingenuous.

Wikimedia's gross mismanagement of the donation revenue threatens Wikipedia's long-term viability, which is something donors should probably care about.

It's about as much money as Signal has. One of those two serves a lot more people though!

It reminds me of the Mozilla Foundation.

Yale University has an endowment of $30 billion. That’s an endowment 300 times bigger than Wikipedia. https://investments.yale.edu/

Apple, in comparison, has $200 billion in cash reserves. https://finance.yahoo.com/news/15-companies-most-cash-reserv...

What’s a hardware manufacturing private company have to do with the Wikipedia foundation?

If Wikipedia is the most successful non-profit tech organization, Apple is the most successful for-profit tech organization.

Okay. Then why do you compare cash to endowments?

Reply on news.ycombinator.com