Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Wikipedia is swimming in money–why is it begging people to donate?

dailydot.com

Post reply on HN
Print view View original

591–600 of 619 posts

Re: Wikipedia is swimming in money–why is it begging people to donate?

#591

Post by tored »

jiggawatts wrote: post #583
tored wrote: post #263

Earlier quoted context omitted.

Why do you think that MediaWiki is not optimized to handle thousands request per second just because it is written in PHP? MediaWiki repository size is about 1.4 GB and according to phploc it has closer to a million lines of PHP code. Sounds like fools errand to rewrite it.

I looked into the topic in some detail year ago, because I copied Wikipedia's approach for an in-house CMS editor. I looked through the code, did loads of performance experiments, read through related forums posts, etc... The wiki staff themselves admitted that the current parser is inefficient, partly because of PHP, partly because the underlying grammar was not designed to be efficient, and partly because the parse…

You don't need to parse the grammar for every request and as they admitted it was an old implementation. What I understand they have a new one now.

You can parse grammar on save and make an optimized compiled format where static content is already resolved. Next step is to substitute semi-static content (like author name) to a runtime format, this last format can handle dynamic data substitution like current date & time, but probably going to be rare that any substitution is needed for the runtime format (how much truly dynamic data does a wiki have?), most of the time just print it with readfile or similar. Only difficult part here is cache invalidation for the runtime format (I know, it is one of the three difficult things you can do in programming).

3000 request/second in PHP is not hard, especially if you plan for it.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#592

Post by codeecan »

middleclick wrote: post #518
codeecan wrote: post #497

Earlier quoted context omitted.

In the https://en.wikipedia.org/wiki/George_Floyd article, they mention an event where he "held a pistol to a woman's stomach", but leave out that she was pregnant so it doesn't look as bad. Example 2, there was two mass shooting in March: - https://en.wikipedia.org/wiki/2021_Boulder_shooting - https://en.wikipedia.org/wiki/2021_Atlanta_spa_shootings In the Boulder shooting, all victims where white and the shooter wa…

The George Floyd thing is not mentioned because it's not true: https://www.usatoday.com/story/news/factcheck/2020/07/15/fac...

yes, I was misinformed, the report says there was a young child at home, but NOT that she was pregnant. The child being there is not mentioned in the wiki, my original point was wrong, but I think the bias still stands.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#593

Post by Qub3d »

bawolff wrote: post #575
Qub3d wrote: post #562

Earlier quoted context omitted.

The WMF has refused to state how the endowment is stuctured, so we have no proof that they cannot dip into the principle whenever they feel the need/want. Users have tried asking for clarification. Multiple times: https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_C...

I dont think your link really supports what you're saying. One user speculating on a user space page (what amounts to essentially a personal blog) is not the same as actually asking for clarification, and it definitely isn't a large group of users asking for such, as your use of plural would imply. Which isn't to say nobody has ever asked, im sure someone has at some point, just that's not supported by your link.

I use the plural because I know there is more than one editor (Guy) who cares about this, if only because there is also myself. I know there is more than that, however, because of the discussion pages (a sampling of which is linked below) regarding this.

The page itself is a starting point. If you want the meat of the community discussion, here's direct links:

* The original "Signpost" op-ed, a curated meta newspaper of which the user-page is merely a constantly-updated copy of: https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2...

* Requests to the WMF for clarifcation:

https://meta.wikimedia.org/wiki/Wikimedia_Foundation_Board_n...

https://meta.wikimedia.org/wiki/Wikimedia_Foundation_Board_n...

https://meta.wikimedia.org/wiki/Talk:Wikimedia_Endowment#How...?

etc, etc, etc. You get the point.

This is a huge rabbit hole and I didn't want to be accused of "Gish galloping" by linking all this at the top.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#594

Post by akolbe »

checkerscaptian wrote: post #566
bawolff wrote: post #461

Earlier quoted context omitted.

The parent literally used the phrase "editors/moderators". The current bureaucracy is not designed to prevent arguments, its designed to make sure such arguments are productive. Whether it succedes is debatable. Its certainly not without issue.

both of your comments are very interesting, i feel like there should be a way that Wikipedia could auto check facts using an algorithm to flag anything incorrect by referencing up to date reputable sources like NASA, Britannica ect.then use the editors/moderators to pick up anything that was flagged.

The Wikimedia Foundation plays absolutely no role in editing or moderating Wikipedia content. I employs no editors/moderators. All that work is done exclusively by unpaid volunteers.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#595

Post by jiggawatts »

tored wrote: post #591
jiggawatts wrote: post #583

Earlier quoted context omitted.

I looked into the topic in some detail year ago, because I copied Wikipedia's approach for an in-house CMS editor. I looked through the code, did loads of performance experiments, read through related forums posts, etc... The wiki staff themselves admitted that the current parser is inefficient, partly because of PHP, partly because the underlying grammar was not designed to be efficient, and partly because the parse…

You don't need to parse the grammar for every request and as they admitted it was an old implementation. What I understand they have a new one now. You can parse grammar on save and make an optimized compiled format where static content is already resolved. Next step is to substitute semi-static content (like author name) to a runtime format, this last format can handle dynamic data substitution like current date & t…

It's not at all hard! A typical Wikipedia page has the "chrome", which changes based on your logon, and then content, which generally does not. Cache invalidation is trivial, because there's only one atomic "blob" to track instead of the output of some complex query.

This is why it amuses me that people seem to think that Wikipedia "requires" tens of millions of dollars of infrastructure to host. It doesn't require it. It's just that some things are better/easier with more kit, such as maintenance.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#597

Post by kmaher »

ramraj07 wrote: post #29
jonahbenton wrote: post #14

Ridiculous question and even more ridiculous factoids. The assumption that operating as a non-profit implies an entity should stop asking its audience for help advancing the mission, or that it should pay its necessary staff less than competitive wages...please. Give wikipedia more so they can take on the MUCH more well-endowed and entirely opaque private capital taking over nearly every other information outlet in t…

What a ridiculous criticism of a valid article with a lot of facts. I have donated to Wikipedia several times and feel violated now. They might have not lied outright but acting as if you’re a poor Non profit when you are most definitely not poor is outrageous. I don’t know if Jimmy Wales is part of this entire charade but this is exactly what you expect to happen when you employ a CEO who charges north of half a mil…

It wasn't north of half a mil, though that does sound nice. It was $300k when I started and just over $400k when I left five years later, which is all data that is entirely accessible through the organization's public 990 filings.

To inform this conversation a bit, the biggest driver of salaries is the market cost of domain expertise and leadership. Wikimedia's salaries are pegged to a basket average of leading US non-profits, but (particularly for more experienced staff) dramatically below market rates for technology organizations.

You cannot run something at the technical and social scale and complexity of Wikimedia without exceptionally talented people, and you can't compete for talent without some degree of competitive salary. Although Wikimedia employees leave a lot on the table in order to work for a mission-driven non-profit (comparative compensation but zero upside equity), it isn't sustainable (or arguably ethical) to ask people to work for significantly less than the value of their labor.

IMHO, the Wikimedia ecosystem organizations could (and perhaps should) be significantly better resourced than they currently are in order to serve the mission of the organization. Currently most of the funding goes into servicing the existing infrastructure, much of which is dominated by the scale of the largest, largely European-language, Wikipedias.

To truly serve the world free knowledge, and serve it well, Wikimedia would need to continue to invest in increasing its global competences, often in regions/languages/markets where operations are more challenging, with commensurate cost. That would mean scaling up that expertise, whether language engineering or legal. All that costs money, which is why so much of the world is so poorly served by businesses with ROI models.

Fortunately, that's not Wikimedia, and will never be. And hopefully, it will also never be the case that some loud people on the internet dissuade the projects, movement, and organization from investing in the necessary capacity to sustain the remarkable good it does for so many hundreds of millions of people and hundreds of millions yet to come.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#598

Post by 6510 »

CRConrad wrote: post #547
6510 wrote: post #282

Earlier quoted context omitted.

Which is a thankless job.

Sure, but what does that gave to do with anything?

That if you are going to pay a sub set in a mostly volunteer project it [imho] should be the most shitty tasks. Improved moderation would dramatically improve the wiki as people are quick to leave if the mod is unable to do the work required.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#599

Post by kaba0 »

denton-scratch wrote: post #504
kaba0 wrote: post #345

Earlier quoted context omitted.

How often is archived pages edited? It is a very different problem hosting static and dynamic data.

An "archive" is a repository of material that can't be edited. Editing archives is a kind of fraud.

That’s my point. You can easily cache static data, it’s easier to scale, etc. Wikipedia is semi-static at most.

Re: Wikipedia is swimming in money–why is it begging people to donate?

#600

Post by akolbe »

kayxspre wrote: post #586

When I first saw the title (I use HN Digest to send Hot Topics to my mail), I thought this was the same as the story I've read several years ago. I have attached some of them here, from 2012 [1], 2014 [2] and 2015 [3]. [1] https://www.theregister.com/2012/12/20/cash_rich_wikipedia_c... [2] https://www.makeuseof.com/tag/wikipedia-millions-bank-beg/ [3] https://www.washingtonpost.com/news/the-intersect/wp/2015/12...

Yes, and they have four times as much money in the bank today as they did in 2015. Go figure! :)
Post reply on HN