In fact, Wikipedia's edit rate has dropped significantly since its high-point in 2007. In May 2007, it took about 5 weeks for 10 million new edits to be added. Presently, it's 9 weeks; the number of edits per unit of time has approximately halved.

The rate of edits per article per unit of time is a fraction of what it used to be. Basically, many articles are fairly static compared to ten years ago, when new content creation was at its peak.

And you're right; the vast majority of old article revisions are never looked at.

Yes, and if you spend x% more each year, then naturally you must ask for x% more money next year, just to have a big enough reserve again. And then you can spend more again, and again, ad nauseam. :) That's exactly what's been happening. WMF asks for and spends about 30 times as much money now as they did ten years ago. If WMF follows that logic for another ten years, it will require 30 times as much money in 2027 as it does this year, just to keep the reserve high enough. That will be $2.4 billion. And after another ten years, $72 billion. I think something will give before then. :)

Wikinews is practically dead. (English Wikinews, at any rate; I don't think it's any different in the other languages.) So are many of the other Wikipedia language versions. A slide shown at Wikimania 2014 said that of 284 Wikipedia language versions, 12 were "dead" (locked), 53 were "zombies" (open, no editors), and 94 were "struggling" (open, less than 5 editors). 125 were described as "in good or excellent health" (that number included every Wikipedia language version that had 6 or more editors).

In my opinion, a Wikipedia language version that has 6 volunteers working on it could not be described as in "good health".

But Commons content has grown significantly, and it does have large files. As far as I recall, it doesn't account for very many pageviews though, compared to Wikipedia.

Pagecounts. Admittedly, traffic was somewhat less in 2008 (that's as far back as that graph goes), but it has more or less plateaued for the past four of five years (while annual revenue more than doubled over the same period).

Then you differ from many people. There are countless expressions of concern online from people who've seen the fundraising banners. Moreover, many Wikipedia volunteers over the years have expressed concern that the fundraising messages make it sound like there is a financial emergency when in fact there isn't. Over the years, it's been a recurrent topic of conversation on the Wikimedia mailing list, every December.

I am okay with that too, though I draw the line at severance payments of this magnitude. YMMV.

And all of that amazing content is brought to you by unpaid volunteers.

There is little need for money to fuel Wikipedia content production. Ten years ago, when content production was at its peak, the Wikimedia Foundation had 11 employees and a twentieth of the budget it has today. Wikipedia looked and worked much the same then as it does now ...

People, by and large, donate "to Wikipedia" (but in reality to the Wikimedia Foundation) because they believe there is a shortage of funds to keep Wikipedia up and running and, like you, would not like to see it disappear. But the Wikimedia Foundation isn't in financial trouble; it is swimming in cash, and has been less transparent about many things, including executive compensation, than it could be.

In my view the WMF could do more to demonstrate that it is spending these increasing amounts of money on things that actually benefit readers and volunteer contributors in some palpable way (including how much was spent on each of these). Cost/benefit statements, so people can see that their money has been put to good use.

There are many reader- and contributor-facing things the WMF could do, but doesn't, to my knowledge. For example, they could pay to provide volunteers with free access to paywalled sources, to enable them to cite better references, and create more reliable content (present initiatives in this area seem rudimentary). They could provide readers with tools enabling them to gauge the trustworthiness of an article, based on its sourcing, or how much healthy community involvement it has seen (what information there is now is so impenetrable that no casual reader can make sense of it). They could communicate more openly about known problems in Wikipedia projects that readers should be aware of. Example. Things like that.

Many volunteers – content writers – are quite jaded about the WMF, feeling the WMF get free money off the back of their volunteer work and spend it on stuff that doesn't really help. Spending money in ways that produce little benefit has been an acknowledged problem in the past.

It is difficult, because both contributors and readers are an amorphous mass, and the WMF has perhaps tried to listen more of late under the new CEO. But when I see managers with a checkered work history receiving six-figure windfalls, or wanting to spend $32 million of donated funds on building a Google competitor, or the WMF clamming up and being unresponsive to reasonable questions, or putting out misleading fundraising messages as they have in the past, I am not convinced that this does justice to the mission people gave money to support. The money given to the WMF is given to them in trust, and in my opinion they need to do more to earn it. That's what this is about, not whether Wikipedia is useful or not.

You gave your time freely. Donors gave their money freely, believing Wikipedia to be in financial trouble. That's generous.

You say you "did this expecting nothing in return, not even gratitude". Doesn't it strike you that the attitude of WMF managers, involved in building the same project as you, yet asking for $200,000 over and above the rightful compensation they received for their work – all paid from those donations – is strikingly different from your attitude?

In 2007 Wikipedia was a top-10 website with much the same traffic as today and got by on revenue of $5 million. Today it asks for – and gets – 16 times as much. Content creation costs nothing and is done by volunteers, who also retain legal responsibility for the content they contribute. The WMF itself has always been protected from liability by Section 230(c).

As for the "unstable donation-based revenue stream", revenue has been on the up and up for every year of the foundation's existence. And whenever revenue has increased, spending has increased proportionally. It looks to me the spending is not driven by need, but by the availability of cash, including cash to pay managers the payments disclosed in the Form 990.

It takes 20,000 people donating $10, in the belief that this money is necessary to "keep Wikipedia online", as Wikipedia fundraising banners have put it, to pay one manager $200,000.

To me, asking for that kind of payment seems sharply discordant with the generosity of volunteers and donors contributing freely in the belief that they are helping to build a common good.

Tell this to the unpaid volunteers who write the content that you and others appreciate so much. They get nothing under this arrangement, while the WMF sits by a faucet spewing money.

Given that contributors donate their time, given that readers donate money, isn't there a moral case to be made that departing managers should not ask to be handed a six-figure sum upon leaving, enabling them to do nothing for two years?

You get kids donating a bit of their pocket money to Wikipedia, believing the site is in financial trouble. You get people in developing countries donating $5, which to them is a lot of money. It takes 40,000 such people making that sacrifice, believing they are contributing to a better world, to pay one manager that extra $200,000.

In my view, it stinks.

Given that all of the content of Wikipedia, Commons etc. is contributed and curated by unpaid volunteers, the question is how much of the "value" is due to the paid staff. Because the staff take no part in writing or checking content.

In 2007, for example, when Wikipedia was already a top-10 internet site, the WMF had less than a dozen employees (compared to something approaching 300 today). How much value have the $350-odd million in donations and the hundreds of employees added since then, over and above the content freely contributed by unpaid volunteers, and was the money spent efficiently to create the most value for readers and contributors?

In 2005, Jimmy Wales was proud to tell people:

“So, we’re doing around 1.4 billion page views monthly. So, it’s really gotten to be a huge thing. And everything is managed by the volunteers and the total monthly cost for our bandwidth is about 5,000 dollars, and that’s essentially our main cost. We could actually do without the employee We actually hired Brion [Vibber] because he was working part-time for two years and full-time at Wikipedia so we actually hired him so he could get a life and go to the movies sometimes.”

It's a very different animal today.

Could you explain this in more detail? The US Department of Labor states, "Severance pay is often granted to employees upon termination of employment. It is usually based on length of employment for which an employee is eligible upon termination. There is no requirement in the Fair Labor Standards Act (FLSA) for severance pay. Severance pay is a matter of agreement between an employer and an employee (or the employee's representative)."

This seems to contradict your assertion that employment law dictates severance pay on an increasing scale.

The Wikimedia Foundation does not participate in content production at all.

Content production, curation, quality checking etc. is all left to unpaid volunteers. That is by design – the WMF doesn't want to be a publisher or arbiter of content. In part, that's because they would then become potentially liable for defamation, errors, etc., whereas now, they're just an online service provider protected from any such liability by Section 230(c) of the Communications Decency Act. Contributors themselves are legally responsible for any content they add to Wikipedia.

So I don't think you'll ever see the WMF intervene in content production.

Well, Wikipedia is entirely written by unpaid volunteers (excepting the odd paid PR writer).

The money the WMF raises does not go to the contributors who create the content. (Okay, some volunteers apply for grants for something or other, and get them, but that is a very, very small percentage of contributors.) The broad mass of volunteers does not get anything.

The WMF does not write the content, and does not check it. They don't even purchase accounts for volunteers to access paywalled sources. (There is the Wikipedia Library, but as far as I know, that relies on donated accounts.)

Actually, internet hosting costs them only $2 million a year. See page no. 3 in this document. Total support and revenue, for comparison, stood at $81.9 million last year. So 2.5% of the fundraising income is spent on hosting.

Quite so. The disadvantage of a semi-crowdsourced system like the one Wales proposes is that the volunteers are anonymous, so you never know which bias has been applied. At least with Fox News or CNN you know where you stand.