I think he raises a very important point regarding the enormous risk of WMF being purchased by the current generation of tech-giants. I don't think there is any reason to believe that new management of Wikipedia by a company such as Google, Facebook or Amazon would be the least bit benevolent. The implications that Wikipedia being co-opted would have on freedom of information on the internet would be immense. I don't think that this is idle scare-mongering. Of course, there is already a risk of misinformation being propagated on Wikipedia by bad actors, I don't think that there is a very big gap to jump before we lose control over the online record completely.
As a 501(c)(3), it would be illegal for the Wikimedia Foundation to be purchased by any of the tech-giants -- or any other for-profit company, for that matter. In addition, all the content is under freely available licenses.
The most realistic but still bleak scenario I can think of is Wikipedia becoming infested with ads. Anything more extreme, we'd see a fork become popular.
While it can be difficult to sell a nonprofit like Wikipedia to someone like Google or Facebook, it isn't impossible.
There are two ways that the commercial entity could get a court to agree with the sale and both of them start with the WMF screwing up so badly that they run out of money and need to either sell of shut down. We are nowhere near that, but it could happen in the future. I have talked this over with people who appear to know the law on this, but I am an engineer, not a lawyer. If a real lawyer replies, please ignore me and listen to them.
There are a couple of things the commercial entity could do to make it more likely to get approval from a court.
They could simply say that they are abandoning all of the federal and state tax exemptions that a nonprofit enjoys. The easy way to do this would be for the wikipedia.org domain to announce the end of wikipedia and tell everyone that they have accounts with the same usernames and passwords on the new wikipedia.google.com site which runs the exact same software and has the exact same content as the now-frozen wikipedia.com version. Add the usual "we don't plan on changing anything; we just want to save the now-failed wikipedia" language.
Or they could create a new, independent nonprofit with the same rules as the WMF has now but with funding by a generous donation from Google and fork the content as above. As above, the key to making this work is the existing Wikipedia running out of money and shutting down.
So, while it isn't going to happen any time soon, it isn't actually illegal.
"Donate time, not money. Contribute as a volunteer."
My experience in trying to donate any amount of time and effort to wikipedia has been nothing but pain and suffering at the hands of extremely possessive "little hitlers" that do everything possible to defend their "turf".
You are suggesting that volunteering at the corporate/entity level is different than volunteering at the wiki level ?
As a wiki content contributor, it seems like the only easy entry-level approach is to find a category that's interesting to you and largely abandoned. For example, any non-famous books by your favorite author are likely to have stubs or no articles at all.
Of course, if the article doesn't exist at all, you risk a war with the deletionists who've decided certain turf shouldn't exist. And occasionally there's a one-two of one user shredding an article as "excessively detailed", then another nominating it for deletion because it lacks content...
But overall, I've found it a more approachable way to contribute than trying to change or extend any article with an opinionated 'owner'. (At this point, I suspect Wikipedia's factual errors are less likely to come from "anyone can edit" than from possessive and mistaken writers actively restoring errors.)
> that do everything possible to defend their "turf".
In my experience, some of these "little hitlers" appear to be (controlled by) larger corporate entities in some way, shape or form. When the "turf" becomes profitable for some, it becomes very difficult for others to cut the grass so to speak.
For Wikipedia I donate money and not time. I agree, people can be possessive about their turf there, but I think that would only be a real problem if Wikipedia weren't amazingly good.
Wikipedia has a great output. In my view they don't need my time and effort. They're doing great now. When I see a message asking to donate money, I do, because I assume they do need that if they're asking for money.
"The Wikipedia Typo Team is dedicated to improving the quality of Wikipedia by correcting typos and misspellings. If you see any typos (even minor ones), please correct them. This kind of editing is a never-ending job, so we could use your help! Please consider joining our team. All you need to do is start correcting typos."
Re: "When I see a message asking to donate money, I do, because I assume they do need that if they're asking for money." please read
Where did you try to contribute? Over the years I've come to see wikipedia as a long term project. It's generally terrible or controversial on current events or recent history. When people lose general interest and the more academic editors move in it's much better.
> It's generally terrible or controversial on current events or recent history.
Or any person with any kind of ideological charge, especially a boogeyman to some group or other.
The problem with Wikipedia is the fanatics with infinite time, which it accumulates because they're the ones who never give up. The other issue is it's inability to actually deal with problematic users, as the moderation doesn't teach the fanatics to be good, it just teaches them to not be so bad they'll be banned.
I habitually use a VPN, and Wikipedia does not let me contribute (specifically, edit pages), even if I am logged in. So, I don't contribute time anymore, and given the facts mentioned in the essay, I won't contribute money, either.
"I habitually use a VPN, and Wikipedia does not let me contribute ..."
Lucky you.
If you had been allowed to edit, your edits would most likely have been reverted so fast as to be confused with an automatic process. New pages would be up for deletion immediately. A swarm of people with far, far more time on their hands than you could ever have would block you every step of the way.
There is this notion that well meaning, knowledgeable individuals can make small, worthwhile contributions to wikipedia and slowly refine the resource as we all chip in our little bit.
This notion is false.
It's a real shame because I would love to contribute time and effort in the way that people think they can contribute to wikipedia.
> If you had been allowed to edit, your edits would most likely have been reverted so fast as to be confused with an automatic process.
If you know that your edit is at risk of being reverted, you can argue for it on the talk page beforehand. Then if no one replies, you can just make the edit and mark it as "per talk"/"see talk page". If the other person reverts, you can now argue that they're in the wrong because they didn't provide a constructive objection beforehand.
It's a silly game, but it helps provide clarity when editing contentious topics.
Has it become that bad? Back when my password was short and I didn't use a VPN and made the occasional addition or change on Wikipedia, it was all accepted or otherwise incorporated, as far as I can tell.
Re: "Has it become that bad?" No. Not even close. The people who complain about such things are typically newbies who make the mistake of trying to "fix" articles on highly contentious topics with hundreds of edits per day. Just pick a page that hasn't been edited in a while, which is most of them. The history tab tells you how often it gets edited.
I'm a very occasional contributor over the past decade, and not had these nightmares. Of course, I've had my hand swatted a couple of times over obscure rules, but for the most part my edits have gone through without too much fuss, including companies I've launched and products I've built.
"Editors in good standing whose editing is disrupted by unrelated blocks or firewalls may request IP address block exemption, which allows editing on an otherwise-blocked IP address... If you will be editing using an anonymous proxy, including a VPN service, you must send your request to..."
Yes, there are some 370 or so users [1] with an IP block exemption, but my request was denied. Just using a VPN habitually is not a good reason, it appears [2]:
> An editor with a credible editing record who would be affected by this measure may be exempted from the block at administrative discretion, allowing them to edit uninterrupted through the IP address range block.
The conditions for granting this are that:
the editor's normal (non-proxy) IP address will be disrupted by an IP address block placed on a range they usually edit on, through no fault of theirs.
My highlights. At any rate, I'm not going to invest more time and work into being allowed to contribute to Wikipedia.
There are a huge number of trolls and vandals who would love to be able to avoid being blocked by using a VPN. Wikipedia has to balance the need to stop them with the need to let legitimate VPN users edit.
In May of 2018 I posted a request for Comment and got a clarification on who should be granted IPBE.
:It's likely safe to say that administrator judgement on a case-by-case basis can decide whether the provided justification constitutes a legitimate need and whether that need is sufficient to grant IPBE.. it's pretty clear that the community at large likely also doesn't want it to be routinely granted, for lack of a better phrase, just because someone wants it and could hat-collect it. Someone should likely, for example, actually be affected by a block to request it and/or someone should probably be able to explain their need for the additional layer of privacy rather than just, 'Hey I want IPBE because I said the word privacy.' "
That was the decision of the Wikipedia community. If you think that decision was "silly" you are free to post an RfC and see if the consensus has changed.
I've heard bad things about writing articles on Wikis. Lots of ego-wars. Not really sure it's worth my time? If it's a good place to be, I'm happy to help.
EDIT: But I do trust it. So maybe it doesn't matter?
I trusted Wikipedia until I've read articles for which I knew aobut the topic beforehand, too. Now.. not anymore. Pretty sure the crazies with lots of times on their hand have won the editing wars.
Which assumes that the musicians aren't lying, but yes, an occasional lying musician is probably still more accurate overall than Wikipedia's sourcing.
Compared to what? I have the same issue with books in my focus areas that are widely revered, and even some peer-reviewed articles in good journals. I often disagree with things said on Wikipedia, but that's because I often disagree with the novice-accessible versions of things I know a lot about.
Re "I've heard bad things about writing articles on Wikis. Lots of ego-wars." you hear those things from new Wikipedia users who decide that the first thing they should do is either [A] "Fix" the Donald Trump article, [B] try to use Wikipedia to advertise their garage band, or [C] try to correct the biases I list at https://en.wikipedia.org/w/index.php?title=Wikipedia:YWAB
If you pick an article, check the history, and find that it ISN'T getting 300 edits every day, you will have a much easier time improving that article.
My Wikipedia page has a link so you can email me. I will be glad to help if you want to become a Wikipedia editor.
Ego-wars are all over the place, even if you avoid political and BLP articles. Burnt-out long-time editors are the most combative. Due to their connections, they won't get blocked.
Thanks! looks like it was a screwup by the ycombinator.com software. The original link ends with a period, and the software stripped it out -- probably because adding a period after a link that is at the end of a sentence is a common error.
I replaced it with another page on Wikipedia that redirects to the "User:Guy Macon/Yes. We are biased." page.
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer".
Wealth means you can buy nearly anything and that includes people who found a non-profit so one can use them to get to entities like WMF.
The scariest scenario in my opinion is their growing influence on the power of definition.
Like news organizations led by wealthy entities the WMF could just be slightly tuned to spread their alternative truth.
I think this power of defining truth is one of the major battlefields of the 21st century. Look at the annexation of Crimea for example. Was it an annexation or not? For us (people of the West) it seems quite clear but ask people from Russia and a lot of them will tell you another story.
The problem becomes how deeply one needs to research before credible facts can be distinguished from agit-prop / disinformation.
I can't think of many people I know who have the time to become investigative journalists, in effect. And I know that a lot of things I think I know are probably propaganda or spin or just one sided stories.
very big problem. I think you are wise to recognise this, 'And I know that a lot of things I think I know are probably propaganda or spin or just one sided stories.'
How can we flush this from our minds? How do we trust those that purport to speak the truth?
For example, here's an easy one, pi=3.14159 (if rounded to 5 decimal places). Great! I think we all agree this is true. The trouble is without context, this is trivia.
Here is a harder one as we move up the complexity scale. Is Human-driven climate change both 1) real and 2) something we can reasonably combat?
Well, the scientists who study it seem to think it's real and there are ways humankind can reasonably push back against it. But there are many, many who believe differently. Which stance is "true"?
Even harder. Is there a God? I mean, from a philosophical view. Sure by scientific definitions we have to assume there is no measurable supernatural. But does that preclude it's existence? For a significant number of people, G-d of some sort exists. And which is the "truth"????
I'm not soliciting hard answers, just trying to show that as you move to more complex, and in many respects more important, topics of today the notion that there is one singular truth in existence gets fuzzier.
Maths, as I understand it, is merely an a priori deduction from an initial set of assumed-true axioms. In some ways, you can do whatever the heck you like, as long as it's consistent with your axioms and whatever laws govern your deductive process.
The only truth about the world is the world itself. Beyond that, truth is a story which omits and simplifies most while emphasizing some things specificially. Depending on what you omit and what you emphasize you can tell very different stories with "truth".
> Look at the annexation of Crimea for example. Was it an annexation or not? For us (people of the West) it seems quite clear but ask people from Russia and a lot of them will tell you another story.
Perhaps a far aside:
Annexation is a quite defined term, specifically: "the addition of an area or region to a country". Word definitions can change over time, but right now this is what it means. It's not necessarily a charged word either.
This was an annexation, regardless if you believe it was just or not.
And on the latter discussion, I'm much more interested in what people in Crimea think than Russians or us in the West.
And what would sustain the fork? Every time there's talk about the potential to fork a popular web product like Facebook or other social media all that happens is the hobbyists and activists move to it while the mainstream userbase carries on.
Besides, nobody can find the fork anyway. Google banishes such clones to SEO purgatory immediately, from which you are basically guaranteed to never return (especially as a pure clone).
That has been true for a decade now, since the days of Stack Exchange complaining about the clones riding their CC licensed content to easy Google ranking. You can put up a perfect clone of Wikipedia, you'll get nearly zero traffic despite having millions of pages of high quality content.
Yep. Even the RuneScape wiki owner forked their own wiki after Wikia became malicious in terms of ads. It even supported by RuneScape devs themselves. And even then, the original Wikia wiki is still competing on Google SEO after a full year. And this was a real niche. Imagine a big website.
It's true that if you create a perfect fork that search engines will punish you and not even display the results, but that's not to say you can't change that. If the edit-base of a wiki moves with the fork (and this is essential), you can continue to create new content that search engines will index. If you also go back and make tweaks to existing pages, you won't get penalised. It's not a quick process, nor is it simple, but it's possible for a fork can survive, grow and even move above the original.
See https://marc.info , its by far the best mailing list archive, popular in open source communities but it absolutely never appears on Google because it has the same content as massively SEOed crap mailing list archives like Nabble. Google has definitely manually unbanned it a few times but it seems to expire after a while.
Yes, but AFAIK all these archives have nothing to do with their primary source, so while we might all prefer no ads there is no objective way to say marc.info is the authorative source over ad ridden sources.
With Wikipedia or stack overflow, I think whoever gets the majority of participants going forward and keeps activity high could start claiming authority in an objective enough sense, and engaged participants are more mindful of organization ethics than random searchers.
This is the important point. A serious fork of Wikipedia with a large chunk of the community behind it would be dealt with manually by Google; their explicit decision making would be the relevant factor, rather than the algorithm.
But this was in the hypothetical of Wikipedia being sold or taken over. If people don't agree with the practices of the new owners and rally behind a fork trusted by the community, I expect most moderators and contributors to follow.
PR firms are really good at spinning and diffusing and confusing criticism. Not completely, but enough to split a community trying to rally behind anything that is against their interests.
The sale or takeover would start with promises that nothing would change, and then would change slowly enough that each editor would have a different breaking point and thus be too fragmented to move over at a specific point.
Of course there are network effects for moderators. One major reason people volunteer to moderate Wikipedia is because it's in such ubiquitous use.
A fork of Wikipedia with no community is just a content mirror and, because of network effects, you'll have an incredibly hard time attracting anyone to help out with it when they can just go to Wikipedia.
Wikipedia itself. The WMF could continue as an empty shell and sell its core assets. In fact, it would be forced to by law, I believe, if it was badly in debt.
To clarify, Wikipedia comprises not just the stuff which is free of copyright, but primarily the userbase, community, support staff, management structures, and (configured) servers it's hosted on. There's plenty to sell there.
Everything else is easily copied or duplicated, and indeed had been. The only thing you can't copy is the domain name. That's the thing that's worth millions of visitors.
Technically, they could sell the domain & the new owner could maintain the site as-is, but I really don't think that'd happen.
The WMF is mission-driven. A large swath of their funds are spent as grants to other organisations, and they could par that back without affecting their ability to operate.
[disc: Wikipedian since 2004, many friends who are WMF employees]
Just create a dummy-non-profit-company which operates under command of the profit-company. Buy WMF and sellout relevant parts to the profit-company or transform WMF into a profit-orientated organization for the sake of surviving.
The could sell ads, but also edit-permission to articles or cleaning of content, usage of licensed content, user-data, or a gazillion other things which are for the average user harmless in the beginning. Heck, I could see stupid influencers paying $100 for getting their own wikipedia-entry.
Ads and tracking. What you read on wikipedia reveals a lot more about your centres of interest, political opinion and concerns than any commercial website.
There are giant voids in the hierarchy between new and established users and between long term editors and the people above. These are mostly sustained by overworking people which in turn is caused by the crappy software.
That there are rules, eh I mean guidelines, is somewhat of a joke without means to enforce them. Disagreements across the void are more often than not resolved by ignoring the lower ranking editor. The concept of consensus[sic] is truly a joke with no clear path forwards. Money doesn't solve anything apparently.
Re: "In the case of insolvency I don't think the board will have much choice"; this is a key point to understanding my essay. Imagine someone winning a huge lawsuit and the WMF not being able to pay. Or imagine a really bad scandal (something like The New York Times publishing conclusive proof that the WMF is a front for ISIS) and donations drop through the floor, followed by the WMF continuing to increase spending until they run out of funds. None of these doomsday scenarios are likely, but they are not impossible either.
I read it all the way through. It raises it, but doesn’t justify it. As has been pointed out most of their spending is grants to other organisations, which they could cancel at any time. The article gives the impression all of that spending is on bureaucracy, but that’s just flat out not true.
Let's say I bought wikipedia and can manipulate definitions, now what? If I use this power then wikipedia looses credibility and all the darn money turns into dust.
You don’t use it, you let other actors use it while you simply rent it out to them. Maybe some Turkish organizations who are pro-Erdogan use Turkish Wikipedia to manipulate Turkish speakers. All the while, people on the other Wikipedia instances don’t notice and lend credibility to the platform.
When we in the West heard about Dragonfly we didn’t stop trusting our own Google results.
This is already happening. Various local wiki sites (including my country) have become infested by right wing conspiracy theorists, but even though it gains media attention, the Wikipedia "court" labels it as independence by local contributors.
Meanwhile kids are learning that the Holocaust wasn't all that bad.
Imagine this happens, but you can't correct these Holocaust deniers because they bought an edit that will remain for at least 6 months. This is what Wikipedia could be selling.
That sounds like a very poor business model - they can and do that for free already. Publicists have been caught making spin edits numerous times already.
> That sounds like a very poor business model - they can and do that for free already.
I might have been unclear, or I'm misunderstanding you, but charging people for something that used to be free is a great business model, since Wikipedia is so big and is so trusted that it's not realistic to go somewhere else or start a new wiki and hope it gets just as big and trusted.
Not necessarily. They could argue that they're a platform, just like Facebook or Twitter, and claim they want to be neutral. Facebook and Twitter have taken flak, but all in all, people still trust the core features of these platforms. Even if Wikipedia sold the right to edit Turkish Wikipedia to a pro-Erdogan organization I would still probably trust the article about set theory on English Wikipedia.
It’s a risk, but the data and software is available and can live on. Nonprofit governance isn’t trivial, and usually when the growth stops the founders or management get booted by the board, and a new normal sets in.
I don't donate to the WMF because they don't need my money, but I wish they just kept Wikipedia running and made small incremental changes over time for scaling/keeping up to date/adding helpful features. I would of course donate if Wikipedia were at risk of failing (not that it would be that hard to rehost and rebuild it), and they know that and exploit everyone who doesn't realize how rich they are with the annoying begging.
As they say: "The bureaucracy is expanding to meet the needs of the expanding bureaucracy". If you keep giving a nonprofit money, it will keep growing, even if it has more than enough it needs for its earlier charter.
But for how many years can Wikipedia run on that much money? IMO not having a large endowment which comfortably lasts a decade is irresponsible for the kind of resource Wikipedia presents to the world.
I have heard just the opposite: that having a decade-length endowment is generally a sign of hording by the non-profit, and unlikely to be contributing to the mission. What scenario is there where Wikipedia is still providing value and fulfilling its mission but can't raise funds for a decade?
Re: "I have heard just the opposite: that having a decade-length endowment is generally a sign of hording by the non-profit, and unlikely to be contributing to the mission."
That's only true of charities where there is no natural limit on how much they can spend. If you are feeding orphans you can always do more until every orphan is fed. The WMF isn't like that, They have one job; keep Wikipedia and the sister projects like Wictionary online. That doesn't require ever-increasing spending other than to cover actual increases in hosting expenses and essential employees.
Imagine a future Wikipedia that is 100% funded by the endowment, which will always be there even if nobody donates, and which has no fundraising banners, just a small "donate" link. I think that is a goal worth pursuing.
As evidence I point to Boys Town, a philanthropy that quite literally exists to take care of orphans. It amassed a large fortune that exceeded anything justifiable by their core mission. In the 1970s, a tip from Warren Buffett on this won The Omaha Sun (which he owned at the time) a Pulitzer for reporting on the resulting scandal. (He was also an investor in The Washington Post which won a Pulitzer in the same year for reporting on Watergate.)
Your reference to boys town missed the point. Boys town didn't run out of orphans to spend money on. They simply failed to spend a large amount of it on orphans.
A better charity to compare with Wikipedia would be the the Washington Monument Restoration Project.
Once the Washington Monument is completely restored you can't spend any more money restoring the Washington Monument. If donations keep coming in, you shouldn't spend them on giving executives of the charity free ski vacations (yes, the WMF actually did that) You should instead build up an endowment that is eventually big enough that the interest covers all Washington Monument maintenance forever. Then and only then should you tell donors "look, we don't need any more money for restoring the Washington Monument. If you give us a donation we will use it for other things, starting with restoring the Lincoln Memorial".
Your reference to boys town missed the point. Boys town didn't run out of orphans to spend money on. They simply failed to spend a large amount of it on orphans.
Do you have a source for that?
My understanding is that boys town did run out of orphans who met their requirements. There simply aren't enough orphans in the USA with no family to take care of them who slip through the cracks of the adoption system.
But that fact didn't slow their appeals for more money.
That said, my understanding is based on my memories of a book that I don't presently have a copy of.
Ah. I had no idea that they ran out of orphans. If they did, then it would be a perfect analogy. Boys town takes care of all the orphans that it is their mission to take care of, keeps collecting donations and spending on other stuff. The Wikimedia Foundation does everything needed to put an encyclopedia -- the thing that it is their mission to take care of -- keeps collecting donations and spending on other stuff. Thanks for the great analogy, and I apologize for misunderstanding and assuming that there would always be plenty more orphans to take care of.
I am going to think about this, do some research, and see if I can turn it into an essay on Wikipedia about collecting money after the job is done. Thanks!
Didn't the march of dimes also keep taking in donations after polio was eradicated?
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer"
First, you didn't answer my question. You may find the abstract idea of a Wikipedia that is perpetually endowed comforting, but you didn't actually describe a scenario where it was well functioning but ran out of money. I think the far-fetched-ness of such scenarios makes that abstract idea less enticing.
Second, the hoarding worry with charities still applies to those with limited possible scope. The idea is that such charities should not acquire perpetual funds because they become unaccountable; this doesn't rely on there always being more available orphans.
Third, why did you limit things to Wikipedia and Wiktionary? There are a bunch of other similar projects, extant and hypothetical, that the WMF would arguably be the best organization to run.
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer".
I believe that I did "actually describe a scenario where it was well functioning but ran out of money."
From my essay:
"Nothing can grow forever. Sooner or later, something is going to happen that causes the donations to decline instead of increase. It could be a scandal (real or perceived). It could be the WMF taking a political position that offends many donors. Or it could be a recession, leaving people with less money to give. It might even be a lawsuit that forces the WMF to pay out a judgement that is larger than the reserve. Whatever the reason is, it will happen. It would be naïve to think that the WMF, which up to this point has never seriously considered any sort of spending limits, will suddenly discover fiscal prudence when the revenues start to decline. It is far more likely that the WMF will not react to a drop in donations by decreasing spending, but instead will ramp up fund-raising efforts while burning through our reserves and our endowment."
My comment was a response to your comment, not the essay. If you want to debate that paragraph:
* The recession hypothetical is highly unrealistic, as already explored in this thread.
* A lawsuit is exactly the sort of thing that is exacerbated by giving the organization a huge endowment. You want fewer assets to be exposed. Non-profits have bankruptcy protection just like for-profits (in fact, greater), so WMF can resume operations with new donations following closing of the bankruptcy proceedings. The only thing a pile of cash helps with is staving off bankruptcy in the first place, but making WMF a juicy target is not a strategic method of doing this.
* A "scandal" that causes the WMF to lose donor support is exactly the sort of thing that should cause it to lose donor support! Why would you especially, as a critic of WMF, think that the organization should be protected in scenarios where it experiences a scandal so severe that it undermines its ability to raise the minimum funds necessary to maintain operations?
* There is no reasonable political position WMF could take that would cut off donations so severely that operations couldn't be maintained. It's donor base is huge, vastly better than other open-source non-profits, and it only needs a tiny fraction to agree with it.
The distributional impacts of major downturns and their effects on the groups donating to any particular charity aren't identical downturn to downtun; the fact that a particular major downturn didn't hurt WMF donations doesn't mean that would be true of the next one.
What historical downturn are you thinking of, and how much did the revenue of any charity lower purely as a result of that downturn (and not, e.g., because the charity's mission was no longer relevant)?
It sounds like some here are of the opinion that Wikipedia's revenue will never take a nosedive for any reason and that the bubble will never burst. Now where have I heard that kind of talk before?
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer"
What your describing is not just an economic downturn, because no historical downturn has ever come close to lowering donations sufficiently to starve Wikipedia. Rather, it's an end-of-the-world scenario, and Wikipedia's expenses would require a hell of a lot more than just paying AWS fees....
Re: "They're working on an endowment afaik", Yes. They are. which is why I have called on the WMF to "build up our endowment, and structure the endowment so that the WMF cannot legally dip into the principal when times get bad."
The danger I warn about in my essay is the WMF continuing to increase spending even if donations stop increasing. If they do that there is currently nothing stopping them from draining the endowment and spending it.
Re: "Here's another view of the WMF's runaway spending", that same graph is part of my essay, but the version in my essay has links so you can see the sources used to create the graph.
To be fair, some of that runaway spending does result in helpful features - perhaps enough to make it a net positive. Wikivoyage and especially Wikidata are highly successful sister projects that probably wouln't have happened on their own, or if WMF had simply "kept Wikipedia running and made small incremental changes over time".
Wikimedia is contributing a lot of time and resources to other Open/Commons projects as well, at least in Germany: Open Education, Open Science, Open Glam, Open Access, Open Government etc. They also invest in advocacy (lobbying for more Open Culture, Open Access etc), which is an important voice in the political process, since most grassroots initiatives can not sustain this kind of work for a long time. I don't see the expenses as necessary for the upkeep of Wikipedia alone, but as an investment to keep the ecosystem of Open Source/Open Access alive and growing, which benefits Wikipedia as well. While I think it is important to question the distribution of Wikimedia support for projects in the Commons (they probably could do more funding, e.g. a prototype fund) , having a strong institution to foster digital participation and Open Society is vital.
As someone who has edited since 2007 (pseudoanonymously, about 13k edits), I appreciate the continued investments that Wikipedia has made into various initiatives, most importantly technology. It's not cheap or easy to stay on top of technical trends (e.g., upgrading to PHP 7) while shipping feature improvements. For example, I use Visual Editor (https://en.wikipedia.org/wiki/Wikipedia:VisualEditor) all the time.
I am all for transparency in spending, but I don't support the author's suggested guidelines for spending. Sticking to inflation-and page-view adjusted growth in spending is a terrible requirement. It ties the hands of the company to do any sort of innovation. I donate pretty regularly, and I am happy that wikipedia is doing more with my money than to just keep the site running.
Can you name one thing that the foundation did with your money? This might be a trick question, so be careful to not confuse things that Wikipedia does without your money with things Wikimedia does with your money.
I can't, which is why I would definitely like to see more transparency. But you do need a certain amount of trust as well in these things. I don't want it to be the case where people are questioning each and every penny. Instead, I want to have knowledgeable, smart people whom I trust run the company and give them some freedom in investing as they think is right. Exposing too many details will make it difficult to make long-term investments, but I do think we should have more details than what is currently available.
Wikidata is the latest "sister project", has been highly successful in some ways, and the funding for its development was quite substantial (albeit managed by WMDE as opposed to the WMF themselves)
"In May 2015, Risker observed that a team called 'Search and Discovery' was 'extraordinarily well-staffed with a disproportionate number of engineers at the same time as other areas seem to be wanting for them'."
"We also know at some point it was an ambitious project to create a brand new search engine as an alternative to Google."
My essay says "We should limit spending increases to no more than inflation plus some percentage". It doesn't say how big that percentage should be. If the WMF limits themselves to not spending more than ten times what they spent last year they would have done what I ask. Right now there are no limits.
The figures don't look anything like growth rate of cancer, they look like growth rate of a toenail.. 10-15% a year.
Then the author claims that "spending is growing at an ever-increasing rate" like cancer. But it is not. Draw a trend-line over the spending graph, and you'll see that the growth rate is actually damn linear (and well aligned with revenue growth).
Rest of the article is reasons why spending should be reduced. Well maybe it shouldn't. As long as there is revenue growth, maybe it is better to invest the revenue on better tools. When revenue growth stops, you've got the better tools you invested in, and can then reduce spending by not building further new tools.
If that was the case then every company/startup that fails to raise another round would survive on the "better tools" they developed with their previous funding rounds.
I agree it's not a perfect analogy, Wikipedia's model is even worse. Startups generate revenue as well. Wikipedia is nearly fully dependant on funding.
Wikipedia cannot "sell" their product, that's antithetical to it's purpose and mission. Doctors without Borders is not a failure because they rely on donations for disaster relief rather then profiting from it.
No one called Wikipedia or Doctors without Borders a failure. The article states that relying on donations is risky by nature, let alone when increasing expenses every year.
I would love to hear your thoughts on this, do you think there is no way donations may decline leaving Wikipedia in trouble with hundreds of people on it's pay roll?
The risk is that WMF decides, as the article presents, to sell off rather than downsize if donation revenue drops. Wikipedia itself can be sustained on 1% of the donation revenue the foundation is getting each year, so the only threat to Wikipedia is the foundation itself, not that there won't be enough money to keep it maintained, independent, and unmonetized.
As someone who has contributed regularly to Wikipedia over the years, I stopped recently due to another cancer that Wikipedia has, called 'Philip Cross.' Supposedly an individual, he has made roughly 150,000 edits since 2013 without taking a single day off.
Anybody who is not aware of 'Philip Cross' can read about 'him' here:
"According to Craig Murray, whose Wikipedia page has been repeated edited by Cross remarked that "the purpose of the “Philip Cross” operation is systematically to attack and undermine the reputations of those who are prominent in challenging the dominant corporate and state media narrative. particularly in foreign affairs. “Philip Cross” also systematically seeks to burnish the reputations of mainstream media journalists and other figures who are particularly prominent in pushing neo-con propaganda and in promoting the interests of Israel."
Wikipedia management, all the way up to Jimmy Wales are well aware of 'Philip Cross' and yet 'he' continues to operate freely as an editor. Despite the basic usefulness of Wikipedia for non-controversial topics, I decided that I cannot and will not support an organization that allows this kind of astroturfing.
First off, the above comment is an example of hijacking; using a discussion about one thing to try to get attention for something else. You see this a lot with Abortion, Gun control, and US presidential politics. I am not saying that those aren't important topics, but do they really need to be inserted into a discussion about how much money Wikipedia is spending?
I would strongly encourage the person trying to hijack this discussion to start a new discussion
I would strongly encourage all HN readers to not give the poster the attention he wants and to downvote any comments that are not about WMF finances
I also would strongly encourage all HN readers to not respond to this sort of thing and to stay on topic.
For those who are interested, here is what Wikipedia has done about this situation:
Philip Cross is indefinitely topic banned from post-1978 British politics, broadly construed. This restriction may be first appealed after six months have elapsed, and every six months thereafter. This sanction supersedes the community sanction applied in May 2018.
Passed 11 to 0 at 18:34, 26 July 2018 (UTC)
Amended by motion at 20:08, 9 August 2018 (UTC)
Wikipedia's relevant policy states:
"The purpose of a topic ban is to forbid editors from making edits related to a certain topic area where their contributions have been disruptive, but to allow them to edit the rest of Wikipedia. Unless clearly and unambiguously specified otherwise, a topic ban covers all pages (not only articles) broadly related to the topic, as well as the parts of other pages that are related to the topic, as encapsulated in the phrase 'broadly construed.'"
Extremely relevant to Wikipedia's finances, since my framing was why I am not contributing to Wikipedia anymore. People deserve to know what they are funding.
Also not helping your cause, you had to try to negate my argument with a false assertion: Philip Cross HAS NOT, in fact, been been banned, and just last month made 49 edits to James LeMesurier's page (clearly a post 1978 British political figure):
Although you may not agree, this is holistically part and parcel of Wikipedia's ongoing funding saga. Plenty of people have tried to appeal to Wikipedia about this, so it is disingenuous to state that option while dismissing the topic here. We are on HN, not Wikipedia, so one man's 'hijacking' is another person's 'information.'
> The modern Wikipedia hosts 11–12 times as many pages as it did in 2005,but the WMF is spending 33 times as much on hosting
This doesn't make any sense to me. Why would the number of pages and hosting costs be linearly correlated? I can't think of any other scenario besides Wikipedia where this would be expected to be true.
I assume the operative word for him was 'linearly' not 'correlated'. But cost/pageview is sublinearly correlated, so spending growth should be even slower than a linear projection.
I just wanted to point out that the article mentions the number of pages, not pageviews, which is part of the reason why I find that statement so bizarre.
This is the first time I have seen any comment saying that I should have listed page views instead of pages. I hadn't given it much thought until you brought it up, but I don't see any reason which I shouldn't remove the pages statistic and replace it with page views. I do want to make the essay as accurate as I can, and criticisms/suggestions are a big help in doing that. (you never see the flaws in your own writing).
Can anyone think of a downside to me changing it to pageviews?
Sure - I would argue that bandwith costs are only a tiny fraction, and the real cost is in stewarding all the existing content(page numbers) - e.g. more servers = more admins, older content/tools need maintenance, etc.
I think what is happening here is that the server costs argument is weak as a first argument. Maybe x33 expenditure is high, but I know WP has grown, and it's hard to quantify how that growth translates into expenditure - it's easy to imagine higher reliability, more accessibility, more need for maintenance of established products and so on as WP matures. If I happen to like WP and am willing to give them benefit of the doubt, why wouldn't I trust that their engineering team knows what they're doing? You haven't established your credibility as an SRE(or enough facts) to really shake that initial goodwill.
Once I view your initial argument as shaky, I am naturally more inclined to view the rest of your arguments skeptically(or just outright ignore them).
From an outsider POV, your two strong arguments are:
- I want to donate to just WP because I don't care about the rest; As the non-WP fraction of expenses grows, there will be a tipping point where the "waste" will erode my willingness to donate
- Even though there are growing development costs, WP tools have stagnated, core problems aren't addressed, new features aren't being used(another signal of "waste")
I think if you lead your essay with those arguments, the server costs argument would be better received. It's much easier to comprehend "Out of $1 I donate, only $0.50 goes to WP", and as an experienced WP user you speak from a position of authority on the second point. From there, it's easier to convince me that WP is also being wasteful in server costs.
In general though, I think you need more evidence put forward to really convince an outsider that WP is wasteful - more comparisons of overall spending vs just-WP spending, more examples of development stalling or going in the wrong direction. The essay may be convincing to an insider who already is aware of the problems, but from an outside perspective it's light on evidence.
I don't think I have ever hinted that Wikipedia is spending too much on servers. I consider that part of Wikipedia to pretty much be the best managed part, and everything they do increases reliability, helps with disaster recovery, or makes the site faster. I even said "It costs a certain amount to have reliable servers, run a good legal team, maintain the core software, etc. But none of the things that the WMF needs to do require ever-expanding spending."
In my opinion, spending 1,250 times as much to do basically the same job is all the evidence you need. I wouldn't squawk at 5 or 10 times as much, but 1,250 times as much? At that point the burden of proof is on the person who claims that such a huge increase in spending was necessary.
Yes, calling out concrete examples of waste would make it a lot easier to understand if this is about actual mismanagement or about a different ideology were priorities are / how narrow the mission is supposed to be.
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer".
Calling out concrete examples of waste also has a downside. It usually results in the discussion going off into the weeds about whether individual expenditures are justified. There are many people who focus on individual examples of waste. I chose to focus on the big picture -- overall spending growth.
> spending growth should be even slower than a linear projection
Indeed.
In my experience in several organisations, as soon as there is even an inkling of "spare money" in (part of) an organisation, humans will invariably starting coming up with creative ways to spend it. This may end badly.
If you start with the premise: "there is [almost] no budget, what can we do anyway?", then you often get the most creative and useful solutions.
This makes sense - and additionally as higher speed internet becomes more prevalent, it's almost become expected to serve more images and videos and for those to be higher quality.
That's a total guess, but in general I'm of the belief that a higher hosting bill does not necessarily imply waste or that something nefarious is afoot
- What is the $7 million for "Professional service expenses"? That's 8% of the budget and I've got no clue what it might have been spent on. Salaries, travel expenses and depreciation is already accounted for.
- What is the $7 million for "Other operating expenses"? Again 8% of the budget in "other" that doesn't include something that is depreciated and doesn't include travel?
- What is the $2.7 million for "In kind service expenses"? To me that sounds like internal "funny money" where you are paying for one service with another service you perform. But how are they calculating the loss and what account is that money actually coming from?
I worry that millions of dollars are being spent on consultants who are essentially just pocketing donations.
Also, I'd really like to see the operating costs broken down because it's impossible to see what's going on here. You've got $3 million in depreciation, but what is that paying for? Desks, computers, telephones? Do they pay rent, or do they own a building?
The thing that's frustrating is that there may be nothing at all wrong. For example, I can see $38 million for salaries and wages, with 300 employees. But where are the other expenses? Is that the "other expenses" section? If so, how hard would it be to break down that cost for me so that I can understand what the money is being used for?
While I disagree with the approach the author of the page is taking, I completely agree that their financial statements are as clear as mud. If this was a company I was thinking of investing in, I would give it a really wide berth based on the terrible financial reporting. As a non-profit operating on public donations they should be doing a much better job.
>"In-kind service revenue” includes goods and services that the Wikimedia Foundation would normally pay for, but have been donated to us at no charge, such as bandwidth and hosting services and pro-bono legal services. Further detail is available in ...
> "Other operating expenses” include expenses for facilities such as rent and office and non-office supplies, insurance, annual staff meeting, recruiting, staff development, property taxes and Wikidata project funding.
Thanks for that. I was scratching my head at the "in-kind service revenue". For others that were likewise confused, there is a line item on both sides: it's income and expenses and cancels out. So that makes sense.
The detailed audit report indicates that rent is about $1.5 million. The rest is essentially computer purchases.
I'd still like to see a better explanation of the professional services. But it definitely makes more sense.
I suspect that the majority of of professional services spending is on accountants and lawyers. The Independent Audit report itself likely costs a 6-figure sum. Depending on how much accounting is done in-house vs outsourced, some of it might also fall into professional services.
The bulk of it is likely going to lawyers. A lot of the in-kind donations fall into the same bucket:
> In-kind service revenue and expenses recorded in the statements of activities consist of contributed legal services. The amounts of specialized contributed legal services as revenue and expenses are $2,748,512 and $156,191 for the years ended June 30, 2018 and 2017, respectively. The value of contributed internet hosting services for the years ended June 30, 2018 and 2017 are $32,722 and $58,390, respectively.
Professional Service Expenses will be the lawyers, accountants, auditors, etc. That adds up quickly. I’m actually a bit surprised it’s not higher. It makes me wonder what percentage of the day the GC spends giving legal advice vs. haggling with outside counsel about billing.
When you run a large organization you can't exactly nickel and dime every line item. There's a cost to running a business. There's overhead for every employee, usually 25 to 30 percent of their gross pay. That doesn't include office space.
It appears that their biggest expense is the $38,597,407 in salaries distributed to ~350 employees. This makes the average salary $110,278 at the Wikimedia Foundation. I don't really have anything negative to say about that.
However I guess you could question why they need 350 employees. So I looked at the projects they are currently working on:
I can see the argument for say 175 employees and a reduced scope, while building up an endowment to better weather down-years is better for WMF's long-term goals, but there's also the risk that reduced spending would cause reduced donations.
> but there's also the risk that reduced spending would cause reduced donations.
Reduced spending and reduced donations would be a good outcome for Wikipedia readers (especially in the long term). Lower running costs allows putting more current revenue towards that endowment, and means the future is a bit more sustainable in less favorable economic cycles.
Donations also don't exist in a vacuum; the best use of donors' limited money is probably not WMF burning a pile of it.
I agree with everything you say, but presumably those in charge of the WMF believe their goals to be of sufficient importance, so arguing to them that they shouldn't spend money on those goals is silly, but one should rather couch things in terms of how best to achieve those goals.
I'm not arguing to WMF; I'm hoping to raise awareness among donors. If the WMF miss their fundraising targets due to donor displeasure, they have to come around sooner or later. Or who knows, maybe they'll run the thing into the ground. I hope not.
The worst part of it for me is that they advertise it on Wikipedia as if wikipedia were going out of business. When in reality, if you give $1.00 to WMF, ~$0.03 goes to server expenses. Wikipedia is their cash cow that funds a ton of random pet projects and a dozen or so executive staff (which seems huge for a 300 person organization).
If they were honest about that in their fundraising, I think they'd see flat or down revenue. I dislike the dishonesty used to solicit donations and then use them somewhere else. It feels similar to a con.
> [The 2016-2017 report] has numbers for "Grants and awards" and "Professional service expenses" but there is no obvious way of finding out the details of those expenditures
> There has been zero actual effort by the WMF to increase transparency on spending.
Oof. Aside from compute costs, what could justify a more than 100% increase in expenses from 2012 to the present? The platform has barely changed in terms of functionality in that time (at least from my basic user perspective).
> The platform has barely changed in terms of functionality in that time (at least from my basic user perspective)
I’m curious, it sounds like you’re a casual Wikipedia user so - how do you have any idea if this is true? What about editor features, community, moderation, administrative improvements?
This kind of attitude frustrates me - it’s the sort of attitude that kills nonprofit fundraising efforts unless they are for show in a way that laypeople can understand. Surely you can see how inefficient it would be to run a non profit if everything they spent money on had to be highly visible?
I'm a very light editor, small dollar donor, and moderate user and I can say I'm pretty impressed with the platform improvements.
The pace of change appears to be very carefully managed to almost always be positive and minimize pointless disruption. The complaints in this thread are part of why you get those big, flashy redesigns that look great but hurt utility.
Of course as a donor I'd never turn down more good for less money, but I don't think maximizing thriftiness always yields the best value. I'd rather things be just a little plush (if that's how you'd describe the current state) and see Wikipedia continue to succeed as it (largely) has.
Frankly, if a nonprofit can't justify to laypeople why they should have money, they shouldn't have money.
My money is limited, the causes which can benefit from it are not. It seems to me the height of self-absorption that an organisation should expect donations over the many other worthwhile causes without justifying the benefit that money would bring.
Changes to editor tools and processes don't cut it. All of those exist in service of improving Wikipedia as a resource. If I can't see those changes as an end-user, I question whether any of them really mattered. Is a fantastic new user-friendly diagnostic machine justified in a hospital if treatment rates don't go up?
Wikipedia's edits rates have been increasing very rapidly. Tools like those must exist in order to vet them and ensure that Wikipedia remains a trustworthy source. Its hard to see the benefits (nor am I sure how to ensure it shows) as the expectation is Wikipedia is always correct, but without them, the misinformation in Wikipedia would be much greater.
Not claiming to know more than you about this but your point assumes that the changing cost of maintaining the project at scale has some proportional relationship to the number of edits.
As it has attracted global attention and ever more significant influence on people worldwide, Wikipedia's threats are likely more varied and insidious than they were in 2004. The cost of offsetting these threats may not relate to the volume of edits.
The rate of increased page edits is closer and closer to the rate of increased expenses. I personally don't think that's a convincing argument, because I haven't seen this massive improvement in tools to justify the extra costs(most of the extra expenditure doesn't seem to be on development, anyway), but I don't think the argument can be dismissed out of hand.
> Changes to editor tools and processes don't cut it. All of those exist in service of improving Wikipedia as a resource. If I can't see those changes as an end-user, I question whether any of them really mattered.
Honest question: How would we as end-users know if information quality went up due to work-reducing changes to editor and administrator tools?
Being pragmatic I agree that there needs to be some saleswork involved in actually getting donations, but I don't like that this is true. It has nothing to do with the quality of the job being done, but simply visualizations of stuff (representative or not) that potential donors want before opening their purse strings.
I don't doubt that the quality has gone up. I have rather more trouble believing that the quality has gone up 3200 times.
We can talk all we want about the nonlinearity of cost, but that's such an enormous jump that the idea that a, by your own suggestion, possibly unnoticeable increase in quality is worth the cost beggars belief.
Frankly, if a nonprofit can't justify to laypeople why they should have money, they shouldn't have money.
This is a joke, isn't it?
"Make Wikipedia better" is about as explanatory as "Cure Alzheimer's!" yet "Make Wikipedia better" is far more likely to be done and less likely to be funded than "Cure Alzheimer's!"
But here's an easy way to justify:
"Improve the moderation and content-editing tools to encourage more contributions," which has been the justification WMF has used for many recent projects.
It can easily be explained to you, but if that's your criteria, no foundation trying to cure any disease that isn't socially-caused should have your money.
I wouldn't donate to a charity which simply billed that they were aiming to cure Alzheimer's, either. I'd want to know what they'd accomplished so far, what they were planning to use the money for and, critically to the discussion at hand, if expenses seemed out of proportion with the activity I'd want to know why.
I don't claim in the parent comment that Wikipedia don't justify their expenses (although I don't think they do so sufficiently). I'm responding to the specific claim that there shouldn't be an expectation by donators to understand the benefit of their donation.
Process matters tremendously; it is essential to Wikipedia guarding against spam and bias.
To use your hospital analogy, diagnostic quality might be the difference between a hospital that works well, and one that carries our mostly unnecessary medical interventions.
While I don't know that I agree with the analogy between nonprofits and governments, I'm not suggesting that Wiki allow us to vote on actions. I'm responding to the claim in the parent that donators shouldn't expect to understand what benefit their money is bringing.
I think good communications are especially important for nonprofits! If one is donating, they will want to know how their money will be spent.
The things do not need to be highly visible of course -- but they should be at least explained. I don't understand why they are not saying things like "we hired 50 more programmers to keep our servers running, and 25 more to develop new AI to fight spam".
Instead, I looked at "2018 Annual Report" [0], and found those specific points:
- Support researchers who make offline wikipedia
- Advertise itself in Nigeria
- Support participation of women in Wikipedia
- AI to detect vandalism
- Page previews
- Content translation tools
That's 1 user feature and 2 editor features. Did I miss something?
Community and moderation are mostly handled by volunteers. Wikimedia's efforts to moderate the platform have been widely condemned by the community, because they're seen as an infringement on the community's self-rule by an unaccountable bureaucracy.
I don’t understand. It’s pretty reasonable to say that editor features, etc. are not worth the money if they are not noticeable to the basic user, which presumably accounts for the overwhelmingly vast majority of Wikipedia page views.
If you edit using Visual Editor and compare that to the old wiki markup, you will understand. Visual Editor much faster esp. with automatic citation templates. My time is valuable (more than when I started as a poor college student) and my tolerance for bad UX has declined as technology in general has improved. And that's just one tool which has improved. There's an annual community wishlist survey: https://meta.wikimedia.org/wiki/Community_Wishlist_Survey_20...
And yes, it's reasonable. I work in tech as a software engineer and it's not as cheap as people think.
Yes software development is a lot more expensive than people think, but 40 million is an order of magnitude more than the combined funding of all but one (and my current, which I only started this week) of my employers over the past ten years. If spent wisely, 40 million can develop a sizeable amount of software.
But you’re right, plenty more companies burn through that without much to show for so in comparison WMF aren’t doing too bad, I guess. It still sounds somewhat excessive to me though.
Software development is very expensive if you're struggling with a lot of technical debt, and Wikipedia is a nearly 20 year old project at this point, with plenty of 'debt' of their own to deal with. Their software (MediaWiki) is still PHP-based, and for the longest time they were running on MySQL.
The rate of spending growth exceeds actual resource use growth by "lots."
Also, the cost of those resources — bandwidth, storage, etc — have been decreasing on a per-capacity basis over this time period. Bandwidth, compute, and storage are all vastly cheaper than they were 10 years ago or even 5.
Ugh, so is this was they’ve been doing with our donations, building some massive behemoth that burns $100M a year on who-knows-what?
Their financial statements say very little about what they’re doing with it, but it is an exorbitant amount of money to keep a website running, even one the size of Wikipedia.
Other threads state that they’re funding all sorts of side-hustles with the money. If so, that is fraudulent, since when they do fundraising campaigns, they claim to be raising funds for Wikipedia. Really slimy of them to then go and spend it on something else.
So in summary Wikipedia is a responsibly run nonprofit that spends slightly less than it takes in while providing a valuable service to millions (or billions) of users?
The summary is that the Wikimedia Foundation asks for and spends increasingly large amounts of money in donations mostly on things that are unrelated to Wikipedia. The writer seems to want the Foundation to use the money instead to build up an endowment (like universities do) so that Wikipedia is not eternally reliant on donations to survive. He also suggests that Wikipedia itself is fairly well-run, but the Foundation is not.
Re: "Um, the graph shows that the reserves are increasing every year. Money in exceeds money out year after year."
You missed the first part of what I wrote in my essay:
"Nothing can grow forever. Sooner or later, something is going to happen that causes the donations to decline instead of increase. It could be a scandal (real or perceived). It could be the WMF taking a political position that offends many donors. Or it could be a recession, leaving people with less money to give. It might even be a lawsuit that forces the WMF to pay out a judgement that is larger than the reserve. Whatever the reason is, it WILL happen. It would be naïve to think that the WMF, which up to this point has never seriously considered any sort of spending limits, will suddenly discover fiscal prudence when the revenues start to decline. It is far more likely that the WMF will not react to a drop in donations by decreasing spending, but instead will ramp up fund-raising efforts while burning through our reserves and our endowment."
I made it clear that Wikipedia is not in trouble at the present time. But there is a real possibility that there are bad times ahead, and we should prepare for them now.
I think you're reasoning about this backwards. They don't worry about setting harder spending limits because the revenue continues to increase. If the revenue cut back, you call it "naive" to suppose that they could put 2 and 2 together and cut spending. So they should be fiscally conservative and think of the future when they expect themselves to be... profligate and short-sighted? Why?
You started this page in 2017. The revenue has continued to increase at as steady a pace as expenses. And an argument along the exact same lines as yours could have been made in 2009 or 2010 or 2011. Actually, your argument has gotten weaker -- expenses used to be higher than the previous fiscal year's revenue, but in recent years the growth has been much flatter.
There's no compelling reason to think that now is the time to act to make sure that it's not too late to act at some vague future point in time. Sure, something can't grow forever. Granted. But are we at 127 grains of rice or 131,071? How can you tell?
I carefully considered the above objection when I first wrote my essay, but I decided against removing the cancer analogy. I though that making an analogy that would be remembered was important. For what it is worth, I myself am a cancer survivor and even after 10 years of it not coming back I still sweat bullets every year when I do my annual cancer screening.
If people stop worrying about forever offended people and someone announcing that they're offended becomes a headpat instead of attention then we will all be better off.
Putting all the arguments aside, I had no idea Wikipedia was so far in the black. The donation begging had me thinking they were desperately clinging to life. I don't like that. From now on my annual Wikipedia donation will be going to the Internet Archive.
I agree. The cup-of-coffee keeps wikipedia running strategy feels kinda scummy, but hey, it's probably how they got so far in the black in the first place. No incentive for them to stop the campaign.
The Internet Archive is also a profoundly worthwhile charity. But those of us with a little understanding of the space would be remiss to not point out that this article is dishonest and you're falling for it. It isn't uncommon for nonprofits (and Wikipedia is not only a nonprofit but is functionally infrastructure at this point; its survival is important beyond its own self-perpetuation) to try to build up an endowment in order to survive lean times. This is normal. You're falling for an okeydoke. Please don't.
Building up an endowment is fine. Misleading me to think that Wikipedia is close to shutting down due to lack of funds is uncool. I don't care that they spend too much money; I care that they convinced me they're broke.
What exactly is dishonest about the article? The author's criticism is that the Wikimedia Foundation isn't building up its endowment fast enough, and that it instead uses the donations on frivolous spending.
Wikipedia is my charity of choice. I am very happy that they are well in the black. I think it would be a disaster if it was ever "monetised", and I want to keep them comfortable with their current path.
The Internet Archive has been spamming me every year since I've donated. They've apologised once after the first spam (last year) which is fair enough, but just yesterday I've received yet another spam asking me to donate more despite never opting into any of that, and no apology this time.
I highly recommend not donating to an entity that can't even be bothered to respect your time.
If you let me know what your email address is I will ensure we take you off our list and on behalf of the Internet Archive I, and we, apologize to you. Please feel free to send your request to info@archive.org and please reference me, Mark Graham in your email. I will look out for it. We appreciate the support you have given us in the past and are very sorry for not respecting your request to not get more email from us. I assure you we have no desire to send email to anyone who does not want to get it. In fact, just the opposite!
Thanks! I've already replied to the original e-mail and will let your colleagues deal with it; if I don't get any reply I'll get back to you directly.
To clarify, I have never explicitly requested "not getting more email from you" (in fact I'm sure the unsubscribe link works fine), I just haven't opted into any e-mail in the first place besides it being mandatory to provide an e-mail address when donating, but I didn't expect it would be kept for anything other than the receipt and any legal obligations. It's most likely against the GDPR as well if you think your organisation falls under the regulation.
So my problem is less about unsubscribing from the list and more about why I was on the list in the first place which is the only reason I'm making a big deal about it.
Like many successful foundations the Wikimedia Foundation has been taken over by empire builders. These types expand non-core mandates to justify increased headcount and gain clout & self-importance.
The tech requirements of Wikipedia are straightforward. The content comes from volunteers. The software development could be handled via a community development process like Debian's.
I do like his idea of building up investments and running wikipedia off of ROI instead of begging for donations every year. Why is it that WMF hasn't done that yet?
It has indeed been raising an endowment for several years, but every time I ask them to structure the endowment so that the WMF cannot legally dip into the principal when times get bad the answer is "maybe next year".
If the endowment is just one more account that can be drained at will we have no protection from a sudden drop in revenue while the WMF maintains the current spending levels in the hope that revenue will recover.
I also looked into whether the endowment is legally protected against a large payout as a result of a lawsuit. I am not a lawyer, but it looks like the WMF needs to structure the endowment so that the WMF cannot legally dip into the principal to get that sort of protection.
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer".
Seems like a terrible investment for the Gates foundation. It is highly visible and has ample public funding. Additionally, it appears that it has no budget control.
Do the math. Their expenses for 2017-2018 were 81,442,265 (81 million). To live off interest (let's say they pull a safe and solid 5%. If you can reliably do better than 5% with annuity-level thinking, good for you), they'd need 1/0.05 = 20 times their run rate, or 1,628,845,300 (1.6 billion).
They hold just under 135 million in assets, so they'd need 12 times what they currently hold. That's why living off interest wouldn't work. They'll need the telethon til they can make that equation work, or at least come closer.
The point of the article, if I an reading it right, is that WMF was doing about as much good when it was spending 23x less. That means they could spend nearly twice as much as that and still live off interest.
The other point of the article is that WMF has zero willingness or ability to restrain its spending growth (demonstrated over a period of years), much less cut back by 23x.
You can’t “live off interest” like that - you need to let the endowment grow by at least inflation. Or else your endowment and its yearly returns will dwindle in real terms
Letting the endowment grow by at least inflation is already covered in my essay:
"We should make spending transparent, publish a detailed account of what the money is being spent on and answer any reasonable questions asking for more details. We should limit spending increases to no more than inflation plus some percentage (adjusted for any increases in page views), build up our endowment, and structure the endowment so that the WMF cannot legally dip into the principal when times get bad."
"If we do these things now, in a few short years we could be in a position to do everything we are doing now, while living off of the endowment interest, and would have no need for further fundraising. Or we could keep fundraising, using the donations to do many new and useful things, knowing that whatever we do there is a guaranteed income stream from the endowment that will keep the servers running indefinitely."
Looking at what they claim to be working on, maybe the money is just being spent on stuff that americans don't see.
The last few years, the annual plans seem focussed on making Wikipedia relevant in India, South America, and the Middle East.
When I was at EE school, the chip manufacturers gave the school devboards students go to know their stuff. Matlab was cheap too.
I suspect Wikimedia is funding a bunch of journalism and research training in these countries, so that every highschool kid gets pointed to wikipedia at school.
If they are as a charity collecting money ostensibly needing it for their online encyclopedia they should disclose all this explicitly with dollar amounts.
If I remember correctly, the person that wrote this had quite the axe to grind about Wikipedia and her/his major criticisms of it were actually what nonprofits look to shoot for. Wikipedia has genuine issues, but not anything this person says has much merit.
This feels like a very empty response; it doesn't point out any specific problem with the author's facts or reasoning. The basic argument is that, ten years ago, Wikipedia ran a very similar service at a much lower cost, so the current organization is therefore spending money ineffectively. You can say you disagree with that, but why? It makes sense to me.
But Wikipedia is no where near the same service it was in 2012. Dozens (if not hundreds) of features, increased traffic demands and higher availability, it seems obvious that the platform would have to grow.
The point is that while views have increased by 10x over the past 15 years, expenses have increased by 1250x. That leaves a 125x increase in per-user spending that does not appear to be reflected in commensurate increases in user value.
This argument seems shallow because the author already clearly and specifically argued that the growth in cost is not proportional to the growth in services authored. For example, one of the first arguments put forth by the author is that bandwidth costs have increased by about 11x, expenses increased by over 1,000. Even straightforward commodity expenses such as hosting costs have increased more than expected, 33x.
So while yes your argument is true that in fact the platform is growing, that however as an argument hete isn't really demonstrontative of a correct understanding of the problem the author argues.
TBH, that 3x increase in expected hosting cost (33/11) doesn't seem bad to me. Maybe Wikipedia has a lot more redundancy now, maybe their pages load faster in more countries, maybe they're using external cloud providers instead of running bare metal, maybe they host more stuff that doesn't get as many page views, etc. (For example, I'm pretty sure I could host HN off my desktop computer, but I'm sure it's hosted in a safer way than that.)
How many employees is the right number? Is "1" the right number, is "300" the right number? How does Wikipedia's growth compare to other well-known companies?
When you subsidize something you get more of it. Don't donate to WMF. If it crashes and burns it is trivial for anyone else to host the existing material with <10 employees and <100k year hosting costs.
But who would pay the 137,201 active editors? Oh, wait, they are all unpaid volunteers.
OK, who would pay the 1,149 administrators? Oh wait, they are all unpaid volunteers too.
OK, who would pay the developers? Oh, wait, the software was originally created by unpaid volunteers and it works well enough that having three developers evaluating bug fixes submitted by volunteers would be enough. There are plenty of Linux distributions and open-source projects that get by on one or two developers.
What about the 924,632,645 edits, 49,122,961 pages of all kinds and 5,978,255 articles? Oh wait, those are free to copy over along with the software that runs the encyclopedia.
Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer"
I'm a bit astonished how here on HackerNews there are articles all day about how sophisticated and surprisingly complicated and convoluted stuff like hosting, IT security, people management, scaling, updating etc is and yet it is somehow ultra suspicious that a page with massive growth had to change in a non-lineary way over the course of two decades with rapidly changing culture...
Wow employing 300 people to keep up with a massively more popular internet on much more differentiated devices in times of information platforms manipulating national elections is not exactly 300 times as expensive as employing one dude to do odd jobs around the house?! Must be mismanagement!!
Wow administrating a site which went from changing slowly to changing at inhumane pace and serving those pages with media formats which did not even existed at the beginning to a growing global population does not just include changing a checkbox on hostgator?! Must be scammers in their IT!!
I realize this comes off as very salty - I actually don't have much stake in this personally but I despise the pseudo intellectual criticism here in this otherwise very down-to-earth community.
PS: Comparing things to cancer is the Hitler-comparison of biology.
Thought exercise: if the Wikimedia Foundation collapsed today and we had to build a new version of Wikipedia without their assistance, how much would it cost?
Data dumps of Wikipedia's databases, and the MediaWiki software itself are all Freely and publicly available[1], so none of that would have to be rebuilt; we'd only have to create a new organizational and operational infrastructure.
I'm sure WMF does some good stuff, but I'm sure the average donator just thinks they're donating to keep Wikipedia alive and online and has no idea about all of the extra stuff WMF does (and therefore needs money to fund). I'm not necessarily saying WMF's stuff isn't worth funding, but it really seems like they're tricking people into paying for it under the guise of keeping Wikipedia running.
@elect_engineer:
Well, as many here already pointed out, I do not see here cancer grows. The 16,000,000,000 page views per month is something really huge.
For some perspective, I run servers for a commercial company, and with such traffic, we would pay about $7,000,000 per year. But of course, we run on the latest and overprovisioned hardware for best performance and safety, and in reputable, but a very expensive company. In this scenario, $2M for hosting such a big infrastructure globally is not looking too bad. I assume that they need a lot of sysadmins to manage such big infrastructure (maybe 500-600 servers?), and those needs to be paid well. I won't be surprised that this would be much over $10M per year. Maybe this data is available somewhere, those are only my pure estimates.
Anyway, with revenue of $100M and $20M of excess every year, dramatic mails and notices I get from MediaWiki and Wales itself, how bad Wikipedia needs my donation, looks now a little disgusting for me...
How to avoid the fundraising banners on Wikipedia:
Dismissing them by clicking the X works only momentarily. Here is how to keep them gone.
You have to have an account and be logged in. Go to Special:Preferences, click on the "Gadgets" tab, then tick the box next to the statement "Suppress display of the fundraiser banner."
Hello. I am the author of the "Wikipedia has cancer" essay on Wikipedia. After I finish writing this I I will confirm this with a post to https://en.wikipedia.org/w/index.php?title=User_talk:Guy_Mac...] on Hacker news I read HN pretty much every day, even though I don't post very often.
I will be happy to discuss any aspect of my essay on Wikipedia, and I would especially be interested in correcting any errors that may have crept in.
What is the cancer exactly? I have a hard time buying that spending by itself is a cancer. Ignoring all the opinion commentary and looking at only the numbers in your essay, they seem to demonstrate exactly the opposite of your conclusion. The numbers are showing support and net assets are larger and growing faster than spending. Isn't that the definition of a healthy business? Spending increases are roughly linear for a decade, perhaps even slowing slightly based on your plot, but Wikipedia usage and value has grown as it becomes a global resource.
This just my take on it from a distance, I don't know the ins and outs of the foundation or the web site, but all the commentary in your essay seems to obsess hyperbolically on separating the absolute spending numbers from any context. I'm failing to see evidence of an actual problem based on what you wrote.
In a general sense, I can completely agree with the abstract idea that spending should have transparency and a budget, that's always a good idea.
Wikipedia seems to me like it's still dramatically overperforming given its expenditures, that its value currently outweighs its costs, when you compare it to healthy for-profit websites of the same size & usage. Could all the increased spending be a symptom of the fact that Wikipedia 10 years ago was undervalued and managing by miracles to survive on a shoestring, and now that it's enormous the costs are correcting to where they should be?
Please stop with the giant disclaimer in front of every single reply and just post something like "(I am the author of the original essay)" and your response.
This actually looks like a healthy, growing organization. You want your revenue to exceed expenses and have a bit of a nest egg to give you buffer for an uncertain future.
I usually check ratings for a charity by reviewing it on Charity Navigator (which is itself a 501 (c) (3) and in need of donations as well). The rating for the Wikimedia Foundation happens to be really good: https://www.charitynavigator.org/index.cfm?bay=search.summar...
I use Wikipedia for reference on various topics on a daily basis and I know my kids do it at school as well. Hence, I donate to the Wikimedia Foundation, the approximate cost of my yearly Netflix subscription as they provide huge amounts of valuable information without any ads and to everyone without paywalls. There aren’t very many free services on the web that run without ads.
Charity Navigator is a fine organization, but they do have the flaw of having to take the charities word on what the core mission is. If the WMF says they need 350 employees -- or 350,000 employees -- to accomplish the core mission, Charity Navigator takes their word for it. If the WMF says that flying people to exotic locations for Wikimania meetings is needed to accomplish the core mission, Charity Navigator takes their word for it. If the WMF says that building a search engine to compete with Google is needed to accomplish the core mission, Charity Navigator takes their word for it.
My position is that Wikipedia was accomplishing that core mission just fine in 2008. I was an active editor in 2008 and I did not notice any pressing needs that were not funded because we were spending 4.3% of what we are spending now. What, exactly, are we doing now that we were not doing ten years ago that justifies us spending 23 three times as much money to do essentially the same job? I could see spending 5 times more, but 23?
In 2019, with a budget of 100+M, if the WMF was really devoted to keeping wikipedia sustainably online, it would have found a way to store it in a P2P fashion.
> In 2008 Wikipedia had over 5 million registered editors, 250 language editions, and 7.5 million articles. Wikipedia.org was the 10th-busiest website in the world. We had already started Wiktionary, Wikibooks, Wikinews, Wikiquote, Wikiversity and Wikispecies, we had already opened chapters in multiple countries, and we had already moved from Florida to San Fransisco.
No neat charts showing trends on these numbers? I'd want to see that comparison, at a bare minimum, before sounding alarms.
The fact that wikipedia, the world encyclopedia, is centrally hosted is itself a problem. Wikipedia should be censorship resistant, borderless, distributed and resilient to disasters, similar to Bitcoin. So far it's none of those things, so it doesn't deserve my money. If I could contribute by hosting articles on my computer I would be more than happy to offer up a few gigs of harddrive space in exchange for access to the world knowledge bank
Oh wow, this is awesome! Also looks like there is an active proposal for decentralizing wikipedia[1]. But IPFS seems to more or less offer a decent solution.
And this should include resistance against deletionism.
To me, the idea of an open wiki died as soon as a group of people started bullying other contributors and telling that the thing they wrote about shouldn't be written about.
So in your view, it's okay if there's a Wikipedia article about my pet cat?
My problem with Wikipedia is the campers, who prevent improvements and even corrections to certain articles because they have a specific view of "the truth". Any changes to their corner of Wikipedia is frowned upon, taken as an offensive and hostile implication that they were unable to maintain their space correctly, and, if not reverted, at least made as difficult as possible.
Having come into conflict with them, I closed my own account and try to limit my current dealings to mere typo/formatting changes. Nothing substantial. Although from time to time I permit myself to do so - and then go away.
> So in your view, it's okay if there's a Wikipedia article about my pet cat?
I wouldn't have a problem with it. I think we should build tools that empower people to involve themselves in the things they care about (and filter out the things they don't care about) instead of having moderators/deletionists/nanny state decide what's good for you.
I agree with your sentiment, but I am curious what the real-life tradeoffs would be?
For example, what would happen to a distributed wikipedia in relation to say maps and borders or other controversial subjects like trump, scientology or vaccinations?
Maybe these are the same questions people had when wikipedia first started in relation to traditional encyclopedias, and we already have the answers?
I suppose all nodes would have to be in sync with a parent node, but indeed that is a good question. What happens when one goes rogue and starts bending the truth. There is nothing stopping criminal governments around the world from doing it, in such a set-up.
It's a symptom of the centralized nature of the web. To serve everybody, you need a boatload of servers.
I recall hearing about this cool free editable encyclopedia in the early 2000's, but I could never get a single page to completely load. It was unusably slow. It's nice to think Wikipedia only needs $50K a year in operating expenses, but you can't reasonably develop and host a top-10 website on that.
The hosting expenses associated with the Wikimedia foundation are, according to their financial report [0], 2.3 million USD. Hosting their servers is actually their smallest expense item, other than "Special event expense, net". As a comparison, they spend 52.1 million on salaries, wages, awards, and grants.
I'd disagree with your statement that they could not reasonably develop and host a top-10 website with the amount of money that they're raising, and if they needed to, they could significantly cut down on expenses and still keep everything running.
Wikipedia does about 200 billion pageviews a year.
At $80,000,000/year, one penny buys them 25 pageviews.
Yeah, there's room for some efficiencies here.
I say this as someone who donated $1,000 a long time ago and then realized that the WMF, despite what it's begging appeals would have you believe, is doing just fine. It's just a matter of what they choose to spend donor money on.
I have no problem with them raising money and spending it, but I find their fundraising campaigns very misleading.
Wikipedia, to a first approximation, is basically a collection of static webpages, especially in comparison to other top websites. Relatively few pageviews need to actually hit the central database (compare this to Google, where essentially every pageview needs to be customized specifically for the request), which means that the actual cost to service the traffic is going to be much less than other websites of comparative pageviews.
Isn't this always the case with open source / "free" platforms? Loads of volunteers put the effort in, while someone somewhere has enormous financial benefits off of it?
> If we want to avoid disaster, we need to start shrinking the cancer now, before it is too late. We should make spending transparent, publish a detailed account of what the money is being spent on and answer any reasonable questions asking for more details. We should limit spending increases to no more than inflation plus some percentage (adjusted for any increases in page views), build up our endowment, and structure the endowment so that the WMF cannot legally dip into the principal when times get bad.
Year-over-year, "revenue" and "profitability", or at least the not-for-profit version thereof, have been increasing for Wikipedia.
Every company in the world is trying to be in that exact situation.
Economics is not biology. Stasis is very hard to maintain. Typically, a business is either growing, or it is declining. Our current problems tend to revolve around companies that are too big to fail, not companies that are too big to succeed.
I would be much more concerned if Wikipedia's revenue and expenditures were flat over the past decade.
Wikipedia is a non-profit, not a company. Wikipedia's "revenues" are donations from readers, who are told that they have to donate for Wikipedia to stay online. Since almost everyone already uses Wikipedia, "growing revenue" must mean getting more donations from each user, who then might donate less to other charities.
>I would be much more concerned if Wikipedia's revenue and expenditures were flat over the past decade.
Why? The thing of value in wikipedia is the encyclopedia that was Built off of FREE LABOR by VOLUNTEERS. The increases in spending seem to be going to white collar office jobs that aren't actually necessary for The Wiki to exist and thrive. All wiki needs is servers. It shouldn't need to grow.
Wiki isn't a capitalist entity. It doesn't need to play by the rules of capital. And that it is mirroring the pattern of capitalist companies is tremendously worrying for the future of the project.
I think you hit the nail on the head. For a charity that is trying to feed starving children, they can always use more money as long as there are any starving children left. Same with most charities. But if a charity exists to but an encyclopedia on the web, they don't need to keep increasing their spending other than what is needed to cover any increases in the cost of running the servers plus a reasonable amount for support staff.
