Tired of the constant wikipedia naysayers. Yes, it has problems, but it's still a resounding success. People seem to ignore the idea that something can fail in some area but still be strong overall.
'edits are declining' does not mean that the wiki is declining. The low-hanging fruit has been largely been done. I imagine also that vandalism control is improved, leading to less vandal edits and less corrective edits. I think the 'edits are declining' comment is far more complex than simply 'editors are fed up and leaving in droves'.
People have been naysaying Wikipedia since the day it started, yet it's still in the top 5 websites by visits.
While he is wrong that Wikipedia is a failure, I wouldn't dismiss this guy's rant out of hand. The editor trends he cites are real. But more importantly, experiences like his are becoming more and more common.
If you think of the product that Wikipedia delivers as being editor satisfaction, it's currently succeeding more in spite of itself than because of its virtues. Part of the impetus behind the GUI editor project is to indirectly broaden the editor base.
I think a more direct solution may be a kind of karma system, perhaps modelled on Slashdot "meta moderation". Most Wikipedians, even the ones that really piss other people off, are usually acting in good faith. If you investigate these disputes, invariably the complainer actually did do it a little bit wrong, or at least there is a legitimate difference of opinion. But the problem is, Wikipedians start a conversation with you by reverting everything you just did. At best you are told to go read a few hundred pages of policies to understand why you were wrong. It would be nice if we had some system whereby the greater community could gently nudge these revert-happy editors, and remind them there are more constructive ways to engage. Just as new Hacker News contributors get downvoted to oblivion when they post memes or GIFs.
I know for a fact that some people at the WMF are interested in a karma system. It's not easy for 5 or 6 people to stay on top of what 10-100,000 people are doing. One concern is that the WMF may be paying too much attention to soul-sucking trolls who rant on Foundation-L, and not enough to awesome people who are quietly doing great work. I don't work at the WMF any more, but maybe some hack day I'll do this....
It isn't a failure at this moment, but as he rightly points out, it will gradually start failing in the years to come, unless something changes.
I like producing overviews of subjects of explanations, but I get the most satisfaction out of knowing it's worthwhile to others. If my contribution can be distorted and vandalized at will, without me possibly being able to revert all the changes, I'm just not interested in investing the time.
I gladly engage in debate about facts and more subjective things like writing and educational styles, but only with reasonable people. If I'm editing the article on some specific part of Quantum Mechanics and have to deal with pseudoscience believers that keep putting in nonsense, I'm not interested either. They can of course have their own page.
I doubt I'm the only one with these thoughts. Just like Athens can't currently stand against the vandals, neither can Wikipedia.
Well that's the problem with committees, unless you are all reading off the same sheet, the music ain't going to sound so sweet.
Regardless, the answer is not "don't bother", the answer is still to sort out the fundamental root cause of the issue. It is the same truth for everything else in life.
A karma system might make the game more fun for griefers. There's nothing in an automated system that prevents a subtle troll from levelling up to mod and hurting wikipedia more efficiently.
I don't think moderation rights should be granted on the basis of karma. In fact, I'd be happy if the karma was visible only to the user in question, and maybe to higher-level administrators. That should remove most reasons to game the system.
There are other reasons why I am interested in the meta-moderation concept. One is that it allows utter newbies a peek at what Wikipedians do all day. Once you have a sense of how others make edits, you may learn to do it yourself.
wow. it's a pretty bold statement to call the largest and most accessible repository of knowledge ever collected to be a "failure". the author of this article needs to get his ego in check, just because your reports of vandalism weren't dealt with in the way you wanted doesn't mean you need to throw a hissy fit.
the worst thing about wikipedia is the mess of egos and politics that happens behind the scenes. i'd love to see some stats about the quality of edits that these so-called editors with their tens of thousands of positive edits make, versus what the average user edits in. i'm willing to bet that the majority of the actual content and information comes from the one-time contributors.
Here's the problem imo: most of Wikipedia's content is currently unmaintained (as far as I can tell).
If there's any political or nationalistic controversy to the content, there's 50 wikilawyers obsessively monitoring it. But if the article is 'boring', anyone can go in there and write whatever they want and (unless a bot thinks it's vandalism), nobody will care.
This is natural - it's fun to write an article and collaborate on edits. Then you move on, because no normal person wants to be the content janitor. And what happens to the article 5 years later? How about 10 or 20 years? I think if Wikipedia doesn't get a handle on this, on the structural level, the site is going to be mostly random agrammatical garbage sooner-or-later.
Wikipedia is already at the point where it seems like every article has content warning banner on top of it. Most of them have been there for years, and nobody cares.
>most of Wikipedia's content is currently unmaintained
"Most" by what measure? By bits? By number of articles? Possibly. By pageviews or importance, no way.
I cannot remember googling something, checking out its wikipedia entry, and finding it unhelpful. It's been at least 5 years since that has happened.
Sure, there's a bunch of low-quality articles with no maintainers, but to claim that that makes wikipedia bad is to claim github's bad for the same reason.
I'm certainly not saying the articles are unhelpful, I'm noticing most of them are slowly getting worse.
Next time you try your google experiment, read the article closely. I can almost guarantee you'll find dubious claims inserted in odd places, atrocious writing, and all sorts of wikiwarts indicating untrustworthy content. Now check back in six months. Chances are the article will not have improved, it will include more of the above. In the long-term that doesn't bode well for the content.
>"Most" by what measure? By bits? By number of articles? Possibly. By pageviews or importance, no way.
Are you positive not by pageviews? Sure, the articles that get a lot of pageviews are maintained, but I imagine there's a pretty big long tail of articles that get small number pageviews each, and not all of those are maintained. I don't think they're necessarily going to fall into disrepair, though.
> But if the article is 'boring', anyone can go in there and write whatever they want
Well, in general what you can do is go in there and write whatever you want - perhaps a well-researched article about a subject which you know well - and then have it deleted within a few seconds.
Conversely, if you try to edit down the nonsensical word salads wikipedians tend to create, your edit will likely be reverted with 'Valuable content which could be improved'. The whole thing is becoming impenetrable.
Let's not get hung up on how literally he intends his rhetoric. He's expressing his opinion of its doom given current dynamics (including many that are not evident to most readers).
He may be right, he may be wrong, but it's more honest and interesting for him to give this sort of bottom-line judgement starkly. As generous readers, we can fill in all the implied qualifiers ("from my experience", "in ways that aren't obvious", "if current trends continue", "unless my recommendations are followed", etc.).
As generous readers, we can fill in all the implied qualifiers ("from my experience", "in ways that aren't obvious", "if current trends continue", "unless my recommendations are followed", etc.).
Wikipedia editors are probably unconsciously conditioned against weasel words... but in counterpoint, when you are presenting an opinion piece, they can be appropriate.
I think this is why I'm such a recluse, I just don't understand human social behavior. Had ->I<- written what he wrote I would have gotten my ass handed to me. When I make statements like that on HN I either get downvoted or my ass handed to me in a reply. This is why I have a psudonym HN account. When I do it, it turns into an arguement, when others do it, people are understanding. sigh I just don't understand you humans...
from the small amount of time i've spent browsing wikipedia edit histories it appears that most of the useful cleanup (of vandalism and the like) is done by bots, not by users. the users who delete stuff delete it not because it's bad, but because it's 'not notable' or some other opinionated metric like that.
I notice that the peak in activity is about the same time as the advent of the Notability guideline [1]. I think the two are related, but I only have one data point to back my opinion.
For me, the zealous application of notability was when Wikipedia changed from being an attempt to document everything to an attempt to replicate what already existed in paper, and I lost the fire in the belly.
It's got to be one of the most contentious issues in the project, with half considering it the saviour of Wikipedia and the other half considering it the death.
The reason why editors stagnate is more like this:
Wikipedia will die because noone real can contribute. Write a new article and it will be shot down. Write an article about important programming languages and it will be killed with some strange acronym.
Exception: a several thousand word article about what Homer did eat in one 8 year old Simpson episode will stay.
His “on the list” idea is totally Wikipedia directorate style and a.) will be implemented b.) will accelerate stagnation
Wikipedia will stay relevant - and I still use it - for some trivia I want to know, it's usually sufficient complete there. But it creates a mirror universe that reflects the thoughts of all editors, not of the online community.
Yep. Wikipedia rejects primary sources as a matter of policy. No original research means that the only "legitimate" content can be secondary or tertiary. Mistakes can't be corrected by people in a position to know the truth- whatever information must be published somewhere first. This is completely backward from how quality content must normally be created.
> This is completely backward from how quality content must normally be created.
It's precisely how content for encyclopedias are generally created. Encyclopedias are compilations and summaries of information about a variety of topics, not compilations of original research by experts from a variety of fields.
The problem may be the encyclopaedia format then. High-school textbooks, traditional encyclopaedias and other printed resources are hardly immune from regurgitating long debunked theories about aerodynamics and Haeckel diagrams.
You're being a bit unfair. Wikipedia does have a goal, and ignoring other valid goals is not really a valid criticism. Wikipedia doesn't strive, for example, to end poverty or cure cancer. Their goal is to build the best and most freely available encyclopedia. You just seem to have an issue with what an encyclopedia inherently is, so I suppose Wikipedia becoming the best instance of an encyclopedia doesn't appeal to you.
Wikipedia is an encyclopedia and not a research journal.
An encyclopedia is more like a dictionary. I cannot just add a word to a dictionary just because my colleagues and me thinks its a great word and we use it often or because we are great in our own field.
No; although a lot of editors do tend to reject them out of hand without actually reading the policy (this really annoys me). Nothing actually precludes using a primary source - so long as it is on a purely factual basis and is not "interpreted"/analysed etc.
"This page in a nutshell: Wikipedia articles cover notable topics—those that have gained sufficiently significant attention by the world at large and over a period of time, and are not excluded for other reasons. We consider evidence from reliable independent sources such as published journals, books, and newspapers to gauge this attention."
Tabloids aren't considered a good source. As to "science papers published on a university website", that all depends. I remember a few months ago that one joyous postgrad student was hosting a pretty off-the-wall theory on his personal university space and trying to get it into Wikipedia as "published by the university". On the other hand properly peer reviewed papers get used all the time/
I'm not enough into american culture to decide if a daily tabloid is a "newspaper[s]" or not. Taking the biggest German tabloid "BILD", although I do not endorse their way or content, I'd classify them as a (bad) newspaper nevertheless.
This is why the word Tabloid is problematic :) because in the UK it means "trashy newspaper", whereas in other places it means the format. Clearly tabloid, in whatever meaning, is a newspaper.
But what you have quoted is a nutshell summary. I can see how one reading of it could be that newspapers are always a reliable source (and as tabloids are a form of newspaper). But I have feeling you've seriously stretched the reading to prove a point.
Reliable sourcing is linked; and if you read that policy it will make it more clear.
FWIW I never like labelling sources, as some editors tend to, in certain categories (newspaper, journal, etc.) because our policy explicitly sets out criteria to apply to sources on an individual basis.
So in that respect, I agree with you that the nutshell is a problem.
Sorry, deleted my comment as it was in error - I thought I caught it before any replies came down the pipe. Basically I misread it as being about a single language.
I do think that there is some merit to his defense (if there's not enough references or the references aren't really about the topic at hand...) but I think the balance would be better left with leaving the article in place and sticking the 'references aren't great quality, please improve' banner at the top.
I'm continuously mystified as to why people who are heavily involved with WP seem mystified as to why the trends in contributions are going into the toilet.
It's really very simple:
1) It's the highly problematic notability guideline abuse
2) It's the deletionists who think it's their job to lobotomize WP and delete knowledge out of the public sphere
3) It's the super editors who enable the deletionists and there's no reasonable mechanism to police either them or the deletionists.
I agree that there's a lack of oversight. The self-licking ice cream cone of the WP community built bureaucracy is finally starting to come apart because there's no reasonable way to participate in the bureaucracy unless you are an established part of it.
In the end, WP will be an interesting curiosity, populated by people drunk on the power their carved out fiefdom in the tangled mess of WP processes gives them, typing furiously about how great their contribution was to the great edifice of WP while their deletionist minions slowly burn the castly down around them until all that's left is a slightly dog eared copy of the Encyclopedia Britannica.
edit reading through the rest of the comments, this idea is repeated over and over and over again. It's sad really, there are vast communities of smart people actually telling WP what the problem is, and every once in a while a clueless editor pops his head out of the vault and asks "why are edits going down?" when it appears the rest of the planet outside of WP...the potential editor pool...already know the answer, but WP won't accept it.
It's also the huge number of juvenile editors using automatic revert tools (eg Twinkle) who think making many reverts and slapping warning banners on user pages is a useful way to clean Wikipedia.
You get the feeling they're on some adolescent power trip.
The whole "VANDAL PATROL" badge thing reeks.
See also the antipathy towards IP editors, who provide many useful edits and who could be encouraged to return but who are often treated like dirt.
See also the bizarre over-concern about usernames. WP has software limits on user-names, and extensive filters. Usernames need only gentle filtering, yet there have been more than one notice board for user name action.
Most of Wikipedia is great, and it's possible to ignore a lot of the meta-stench, but sometimes quiet productive editors get caught up in drama-mongering idiocy.
I used to clean up and correct small mistakes or vandalism, I don't have an account.
My corrections were usually reverted almost instantaneously and I had to insist to push them, even when just cleaning dirty words by bored teenagers. Does someone even read edits before reverting them ?
Nowadays, I just correct nothing and don't bother.
"Does someone even read edits before reverting them ?"
I don't think they do. I'm convinced that they won't even read them through the entire appeals process. In the world of WP, the reversion holds more weight than the actual content that was reverted.
Pure sour grapes. Wikipedia does not need more elitism and douchebaggery than it already has. The experiment is a success because of its pluralistic foundation, but 'top editors' (academics) keep forgetting this and attempting to turn Wikipedia into a 'real' encyclopedia. It isn't nor will it ever be. It's a repository of facts, opinions, lies, and yes, some vandalism. In other words, a snapshot of the state of the internet, maybe even more than that.
The main problems with Wikipedia from my outsider perspective:
1. Lack of a viable long-term business model. Begware will never flourish.
2. Lack of an efficient impartial mechanized/algorithmic resolution system.
3. Lack of expert business leadership - see #1.
As personal appeals go, I'm just not sure. On the one hand it's almost impossible to sympathise with the guy's crazy predictions or hurt feelings. It definitely needs a big watery-eyed photo of him across the top or something. On the other hand, "don't bother" is a very powerful call-to-action -- so powerful that I nearly didn't bother reading the rest of the post.
He seems to be sure the number of active contributors is going down due to poor oversight and 'troublemakers'. In my experience this is wrong - far more important is that people can't be bothered to get their head around wikipedias obtuse structure and ever-more-complex editing methodology, and when they do, their contributions can be overzealously edited out by the wikipedia police...
Nice article, but it'd be nice to see what the author's proposed fix to this would be. Stopping multiple accounts from the same person is a nice idea in theory, but breaks when you realise it's technically impossible (IP checks are easy to work around, and ignore the fact that many people can share IP addresses).
My vote would be for a strong reputation system (i.e. your edit must be approved if you are not a trusted editor); it may hurt one-time contributors but i think it would be for the best in the long run.
If they could set the system up to incentivize quality rather than quantity, gamifying it could lead to more good edits as well.
I'm not sure it's plausible for Wikipedia, but it seems to work for StackOverflow. That said, StackOverflow does not have any seriously controversial material on it.
The problem with this is it conflates expertise with wanting to play little karma games. They may (sadly) be equivalent on StackOverflow, but in some fields I imagine the experts have better things to do. The wiki would start to reflect experts at the game rather than experts in the topics covered.
Why is updating and extending wikipedia pages not part of normal school or university work.
E.g. teams of two students get topic and have to create a value contribution to wikipedia. Not only recite current knowledge but go beyond that and extend the "common denominator" of knowledge called wikipedia. It would teach about research, team work and to some extend about editorial quality and processes.
I proposed it once in college. The teacher suggested to bring it to an multimedia course. Maybe the world would be more open today.
It is; there are various outreach efforts to bring Wikipedia editing to the classroom (and other similar arenas). The GLAM and Wikipedian in Residence schemes work with various cultural institutions and ones such as Campus Ambassador are active on college/uni campuses.
Most only just getting into their stride, but I think the world is more open like you hope :)
It seems to me his appeal is mostly about his personal problems with the organisation. He felt unsafe, felt like his privacy was void, and that nobody listened to him. Also he calls it a failure because of a few core problems (which I, as outsider, don't see going wrong as badly as he says it is), but he doesn't acknowledge what Wikipedia has archieved and what it is to many people.
I personally think it's a great resource to learn the basics about mostly any subject. I never donated though and I'm not sure I ever will, but I don't see how a non-profit organisation needs millions to run. Sure they need money for hosting, but not millions. And if they do, they should look for a model to get some money. I'd rather see ads than Jimmy Wales' face printed hugely on top of all articles for months.
I don't think Wikipedia is going down the toilet, nor do I think it was a failure. Personally, I don't think the graphs are surprising: first Wikipedia becomes a big hit, and loads of people sign up. Then, as time passes, some people become unactive. In the end, I think we'll still have a pretty large baseline editing Wikipedia.
The biggest problem I see with Wikipedia is that it's only reliable as long as you're looking up non-controversial issues. Thus, it's fine when you're looking for definitions and trivia, but when it comes to the interesting stuff - pretty much all of which is controversial, because that's where the good bits of science happen - Wikipedia is horrible. In these cases, it is much better to check out (for example) Google Scholar and search for scientific articles directly.
Ah, if only there were an encyclopedia that would directly list all the appropriate references for a given subject. I'd pay good money for that.
Would it be possible to create a bayesian spam filter-like system where new users edits are analyzed and matched to old accounts? That way, admins could get a heads up for a new user that is likely to be a pseudonym of an old user. Would it be technically feasable?
Maybe there are social consequences to this as well? What about false postives?
While I agree with the article that WP is in trouble, I disagree entirely as to what the problem is. WP is not too bad at the low-level policing part, to do with stopping vandalism, enforcing its biography of living persons policy, &c.
It has mostly failed at the higher-level policing mission, of ensuring that those editors who are trying to write articles that are soundly sourced and informative are able to get on with this task without being harassed by people pushing agendas. The more value a resource Wikipedia becomes, the more valuable it is to such agenda pushers: cranks, lobbyists, corporate PR, &c, who often can devote considerable energy to this task.
For example, the editor FT2, who earned money through his NLP practice, pushed pseudoscientific doctrines on many Wikipedia pages. he was expert at playing Wikipedia politics, and rose to become a member of Arbcom, and only came unstuck when he abused this position to punish editor Orangemarlin, an editor who campaigned against pseudoscience and who was a nuisance to FT2.
The way WP now works is that the enforcement of the low-level policing is done at the expense of the higher-level issue of neutrality. By playing the politics well, as FT2 did, agenda pushers can use the WP rules to waste the time and wear down the patience of good content editors. I left WP following the fiasco of the Arbcom ruling against Peter Damian.
Ed Buckner wrote an excellent summary of the problem with agenda pushers in Skeptica Adsversaria (apologies: the formatting in this copy is awful 400px left and right margin hard coded into every paragraph: OK if you have a very wide screen or can overide the margin somehow): http://www.btinternet.com/~musicweaver/wikipedia-by-ed-buckn...
>My personal well-being (as well as my privacy and, I truly believe, my safety) are not worth the risk and grief.
Jesus. Melodramatic much?
I've been a low level Wikipedia editor for many years. I have edited and written a smattering of articles, and it has never been a waste of time, and certainly not dangerous to my SAFETY.
I don't know if anyone has gotten any value from my articles, but I have. I have written articles to help study for exams back in college, I've written articles to help my keep track of the TV schedule of a favorite but unpopular TV show, and I've written articles about favorite but unpopular books. I've also written and edited articles that I use as references for my work.
I never got involved in the "politics" of Wikipedia; I just care about making the content better for personal use. Maybe this is where this guy went wrong. Simply editing articles is not a waste of time, even if you're the only one that reads them. At the very least it helps you.
>I never got involved in the "politics" of Wikipedia; I just care about making the content better for personal use. Maybe this is where this guy went wrong. Simply editing articles is not a waste of time, even if you're the only one that reads them. At the very least it helps you.
I'm also a low-level wikipedia editor; you have been extremely lucky to have not had this trouble, my friend. In 98% cases, it has been great; but when there is a disagreement, the debate can quickly get out of control if the sides aren't willing to compromise. For instance, try making an edit on the page for the Monty Hall Problem http://en.wikipedia.org/wiki/Monty_Hall_problem.
"Statistics already show a steady decline in editor activity on Wikipedia, and analysts have myriad theories as to causes. But I think it all comes down to a handful of interconnected problems:"
Isn't there a more obvious reason? Most subjects are already covered, and already written well, leaving only subjects which less people care about.
I definitely agree with you. There are some pages written better than more "real" encyclopedias would do. Ok, other pages could be made better, but the most part of "arguments" have been already treated.
That "naive openness" of Wikipedia that the author describes is a huge part of its success. If a top-down form of editing is sought, there are encyclopedias for that. Wikipedia already has its groups of elites (deletionists come to mind), and we've had plenty of articles posted to HN about them. But I can't see any good reason to call Wikipedia a failure, as the author attempts to do. If Wikipedia were an obscure site, with few articles of interest, then one could entertain the thought of failure.
In one possible future Wikipedia editors are vetted by an independent auditing commission consisting of topic experts, an Institutional Review Board and a jury of their peers. Once vetted they are given a cap and gown that they must wear while committing their edits. They would also be allowed to put the letters W.E.C and the number of pages they are considered arbiters of after their name and on their facebook profile. All Wikipedia Editors Confirmed would be allowed to park in handicap parking spots and be given preference in matters of tenure at all educational institutions. Order would prevail, knowledge would be defended, und so weiter.
In another possible future a collection of bots will textmine wikipedia edits in realtime and many wikipedia janitors will rely on botscores to determine which edits are fraudulent. Advanced wikipedians will become cyborganisms devoted to spotting evidence of bad intent, their skills will be in high demand as corporate negotiatiors and wikipedia will languish as the cognitive bandwidth of it's most effective members is coopted into helping facebook fend off it's antitrust lawsuit and making the democratic parties stranglehold on national politics unassailable.
In another possible future it is revealed that Chris Poole is actually Jimmy Wales father due to bizarre accident involving a time machine and an ice cream sandwich, and that the entirety of the wikipedia project is in fact an elaborate troll designed to spell out a very rude word in babylonian when viewed from the right angle from three light years away. (that is a very disturbing future indeed).
These days I find myself going to tvtropes rather than wikipedia, even for real-life things. When I was looking up a country I was about to visit, the wikipedia page contained more facts but tvtropes was a lot more informative. When looking up an obscure band, wikipedia didn't even have a page. It's not without its problems, and seems to be getting worryingly formalized lately, but it feels more like wikipedia of old than today's wikipedia.
Citizendium is pretty clearly a failed project at this point. The founding principles of the site (Real Name editors only, strong bias toward credentialed "expert" editors) are pretty much an amplification of the exact features that Kevin Forsyth is complaining that Wikipedia is displaying!
where can we get statistics for the last two years of editing?
Every time I see graphs showing the decline of editor contributions they always seem to stop in 2010 (I expect the trend to continue, I'm just curious).
The decline of existing editors has stabilised a bit - the trend downward is now fairly slow (and there are a few upswings). New editor pick up has recently slumped a bit more dramatically.
"....Another editor cheerfully tutored me in what this means: "Wikipedia is not 'truth,' Wikipedia is 'verifiability' of reliable sources. Hence, if most secondary sources which are taken as reliable happen to repeat a flawed account or description of something, Wikipedia will echo that." http://chronicle.com/article/The-Undue-Weight-of-Truth-on/13...
It makes sense to me. Going with the majority opinion is the only way to have stable content. Trying to resolve controversy on Wikipedia is what would actually destroy it. Democracy can cause horrible injustice, but we've learned from history that it's the only irrefutable way to decide things, unless the evidence is even simpler than counting ballots.
It doesn't mean Wikipedia is a failure, it just means Wikipedia is not the oracle of truth that some people expect it to be. It's a repository of majority views, which is usually pretty valuable, even if it's sometimes wrong. We have other venues for challenging popular fallacies.
You better have stronger reasons than those presented in the article to doom Wikipedia to be a "failure"
The problems stated by the author are valid however they are solvable. Definitely, Wikipedia must have a scoring system to allow editors with high points to have more credibility than those with lower points, also a system to eliminate the trouble makers.
'edits are declining' does not mean that the wiki is declining. The low-hanging fruit has been largely been done. I imagine also that vandalism control is improved, leading to less vandal edits and less corrective edits. I think the 'edits are declining' comment is far more complex than simply 'editors are fed up and leaving in droves'.
People have been naysaying Wikipedia since the day it started, yet it's still in the top 5 websites by visits.
http://en.wikipedia.org/wiki/Tall_poppy_syndrome