Should you give money to Wikipedia?

Readers of this blog might already have noticed that links to Wikipedia are a frequent occurrence. The reason for this is simple; both me and Tove are avid Wikipedia readers. In fact, we think Wikipedia is about the best invention since sliced bread.

Still, I belong to the 98 percent who scroll past Wikipedia's occasional fundraising ads. This might seem odd from a die-hard Wikipedia fan. But there is a method to the madness. Since my financial means are rather limited I want my money to go as far as possible. And as far as I have been able to find out, Wikipedia does not make the most of its money. 

Many commentators have noted that Wikimedia Foundation, the organization behind Wikipedia, is rich beyond means. This, however, is not my main criticism. Ample reserves can be justified if you run a critical operation that you want to survive for a very long time. The more acute problem is the way Wikimedia Foundation spends money.

Auditors ready

Being very much a public entity, Wikimedia Foundation publishes its finances on its homepage. The financial statements for the broken auditing year 2020/2021 is available here. This transparency makes it rather simple for would-be donors to assess the actual need for donations.

Firstly, Wikimedia really has a lot of money. Financial assets are listed as $205 million. This is a big pile of money to have tucked away. But actually not that much money considering that the same financial report state that Wikimedia had total yearly expenses of $112 million. The financial resources, which look abundant enough, actually cover less than two years of expenses.

The more interesting question here is why does the Wikimedia Foundation need to spend more than $100 million a year? It is the seventh most visited website in the world (the link goes to Wikipedia, of course) which should be costly. But the financial report states that Wikimedia only spends $2.4 million on internet hosting. Since Wikipedia’s content is provided by unpaid volunteers and its codebase is also open source and provided by volunteers the cost of running the website should really not be more than the internet hosting.

This raises the question what the other $110 million in expenses are for. The financial statements are not entirely clear. Some things are clearly stated. For example $9.8 million goes towards “awards and grants” which presumably is money spent on worthy projects outside the Wikimedia organization. Another well stated expense is almost $1.6 million spent on rent for Wikimedia’s offices in central San Francisco. The wisdom of a non-profit organization paying for some of the world’s most expensive office space could be discussed. But it is still a minor expense in this context.

The main expense is salaries. $68 million is spent on this yearly. For this sum Wikimedia gets the labor from 550 staff and contractors. The average salary of $124,000 a year seems high to a low-income earner like this writer, but it probably reflects the salary level in San Francisco.

The better question is what all these employees actually do. Looking back in history just 3 years, to 2018, Wikimedia Foundation had 300 staff and contractors. The headcount almost doubled in 3 years. Personally, I cannot say I noticed any difference in Wikipedia during this time that could explain a doubling in personnel.

Going even further back in history gives some perspective to the current organization. The very first available full-year financial statement from Wikimedia Foundation is from the 2005/2006 financial year. The foundation then had total expenses of $791,907, 0.7% of last year’s expenses. 15 years is of course a long time. But I used Wikipedia regularly in 2006 and as far as I remember it looked and functioned about the same way as it does today.

It is not only the size of the expenses that differ between 2006 and 2021. The composition of the expenses has also changed dramatically. While Wikimedia today put 2% of its money on internet hosting, in 2006 no less than 24% of the budget went on hosting.

Money for nothing

The financial statements are precariously short on details on actual spending. Not even the Wikipedia page on the Wikimedia Foundation has more than a single sentence to say about the actual expenses. To get something resembling a picture of the expenses one has to go to Wikimedia’s own homepage where they tell the public something about the work they do. The following six areas are listed as their core operations:

·        Host free knowledge projects

·        Deliver grants for good

·        Grow community around the world

·        Support education and learning worldwide

·        Partner for change

·        Open the Knowledge

The first two points, internet hosting and grant giving, are both included in the financial statements and stand for 2% and 9% of expenses respectively. The Partner for change operation seems to be merely an intention to work with other actors rather than an actual program, which should mean that its costs are negligible. The lion’s share of the expenses should thus be from the three other operations, where, conveniently, Wikimedia lists a number of their own projects.

Grow community around the world mentions several projects most likely involving multiple employees from Wikimedia. The Community health initiative intends to limit the use of Wikipedia for harassment through the development and implementation of a range of anti-harassment tools. There is also a mention of an AI project, supposedly launched in 2015, to help Wikipedia editors find and correct faulty edits. Since I have never heard of this before, despite being an occasional Wikipedia editor, I assume it never got off the ground.

The Support education and learning worldwide operation is working for greater use of Wikipedia in the education industry. The Wikimedia homepage mentions numerous lofty ambitions for more Wikipedia use in the classroom but no information on actual money spending. One line from the homepage goes: “At the Wikimedia Foundation, we help teachers and students of all ages access and use the full potential of Wikimedia projects for their learning objectives.” Which could be only cheap talk or an indication of expensive education information projects.

The sixth and final area of operation, Open the Knowledge, is also the best documented. The objective seems to be to spread the Wikipedia project to as many disadvantaged groups as possible. Or, seen from another perspective, the objective could be to ingratiate Wikipedia with as many politically correct groups as possible.

Project Rewrite has the lofty ambition to close the gender gap on Wikipedia. The Inuka project strives to “deliver [a] compelling experiences to Wikipedia readers in emerging markets” (it seems to be a complicated way of saying mobile first). Wiki loves women is a slightly mis-named project (in fact it only concerns African women) that aims to increase Wikipedia entries of and by African women. To its credit it presents its accomplishments fairly openly. For the price of 3 Wikimedia employees (probably full-time but it does not really tell) it has made 1922 Wikipedia edits. The Wiki loves pride project is quite self-explanatory.

Mission creep deluxe

It is hard not to associate all this with a fair dose of hubris. The original purpose of Wikipedia, still listed on the Wikipedia site, is to be a “widely accessible and free encyclopedia”. This ambition did Wikimedia fulfill already in 2006 with a budget a fraction of today’s.

It could be argued that Wikimedia is doing an essential job in enhancing the Wikipedia encyclopedia by widening its base of contributors. After all, an encyclopedia with the ambition to cover all human knowledge needs the cooperation of as many humans as possible. This seems to be Wikimedia’s own reasoning and its raison d’être for the multiple outreach projects they are funding.

The problem with this argument is that it is impossible to choose your outreach objects in anything resembling an objective manner. Wikimedia is surely correct in noting that African women are underrepresented as Wikipedia contributors. But Norwegian fishermen are most likely also underrepresented (the total lack of an article about Norwegian fisheries indicates as much). That the former is subject to intense Wikimedia interest and the latter is totally unknown is surely not a rational decision as much as a political one.

Wikimedia still accomplishes its core task of keeping Wikipedia (and its sister wikis) online. But by now this is only a minuscule part of its workload. The major part of its mission today is to spread the ”Wikipedia spirit”, something rather intangible that nevertheless seems to correspond strongly with a certain set of political views.

If one happens to share the political views of the Wikimedia Foundation there is of course good reasons to support it financially. If one does not support the political views or, like me, are politically dispassionate, there is no good reason to give any money. Wikimedia has more than enough money to fulfill its original purpose of hosting Wikipedia in eternity.