Slashdot Mirror
Larry Sanger on Wikipedia and World
Phoe6 writes "MIT Tech Review is running an article on Larry Sanger, an epistemologist and the co-creator of Wikipedia. It is very interesting to know his views on Wikipedia. He says, 'To build a public encyclopedia, you don't need faith in the possibility of knowledge, What you have to have faith in is human beings being able to work together.'"
What he seems most upset about is the problem of "revert wars" happening whenever an author wants to be the absolute authority on a topic and regularly patrols their article to undo any edits that are made to what they consider their "perfect" work?
What could they do to defuse these situations with a moderations scheme that encurages contributors but discurage this kind of abuse?
Wikipedia article on Larry Sanger
;-)
It had to be done
-- If you try to fail and succeed, which have you done? - Uli's moose
mod article down (Score:-1, Troll)
wikipedia is a bunch of hyporcritical censoring liars. they love to censor certian types of poltical articles that don't match their agenda but they let opinion and bias sneek through if it fits their agenda
Apparently the same applies to Slashdot mods...
He says, for the whole thing to work, humans have to work together and help one another. Now, I like Wikipedia, and I think it works, but not because people are "helping" as much as they can. It works because there are lots of people with big egos that want to show off their knowledge, and moderators that aren't afraid to ban an entire subnet (any computers from the Calgary Board of Education, where I go to school, are prohibited for making changes; take two guesses why, the first one doesn't count). It still works, but the assumption that it is because people are "helpful" to each other is slightly flawed. I'm sure there are some, but as with most other facets of life, I imagine they are vastly overshadowed.
"What you have to have faith in is human beings being able to work together."
:(
Seems that most of our greatest achievements have been by individuals. People working together usually create destruction.
Wikipedia is doomed.
Wow dude, has anybody seen the last article - Yellow Dog Linux? It's become a GNAA nest, I think. To be more on-topic, this dude does have some interesting views on wikipedia, but I think the core of it lies in the article summary - "it's all about having faith in people's ability to work together...". For a cynic like me, I don't have that faith - at least for something like an encyclopedia. There are enough people who like destruction for the sake of destruction, see previous article on Yellow Dof (and 9/11) for an example. What does it cost somebody to revert a wikipedia article and totally trash it? It's a teency-weency bit harder to do the same thing on a FOSS project. As it is, skeptics/cynics like me take all encyclopedias with a pinch of salt, be they online or on dead trees.
My Favourite Meme
In order to deal with the 'reliability' aspect constantly brought up, Wikipedia's appointed management, could use an audit to ascertain the quality of the project.
My rough idea is, pick the 10 most popular articles, 10 random articles of moderate-to-high traffic, 10 random articles of low traffic and then do a compare/contrast against 'reputable' references. Then, check those references (and Wikipedia) against primary source references (if they exist, like journals/textbooks, for medical facts..etc). It will provide a good, quantified metric of the quality, acting as a rough indicator of where Wikipedia stands.
> So, you poor Wikipedians, finish your sorry library project and create something much more unique.
I don't know... I decided to take a quick look at everything2 because I had not before, and wikipedia impresses me every time I check it out. I happen to come from Marion, OH, home of Warren G. Harding. (Sure, the rest of the planet thinks the guy was a terrible president but in Marion he's a home town boy and worshipped!) Anyways, I looked him up on both. I have to say that the Wikipedia Article looks much nicer than the the Everything2 article, contains more info, links, etc... I'd be happy to see a counter-example though!
http://en.wikipedia.org/wiki/Mad_cow_disease
Maybe all that doom training will be worth something!
If Tyranny and Oppression come to this land,
it will be in the guise of fighting a foreign enemy. -James Madison
Everything2 was alright at one time. But since the advent of Wikipedia, e2 has been displaced and somewhat made irrelevant.
e2 has lots of idiotic entries, which include personal rants, crackpot theories, and at times - 5-word entries making sense only to its writer. It is slow as hell. The search feature is terrible. Entries are in a form of a forum conversation - there's little collaboration going on. Your entry never gets modified in order to be made better. And lastly, e2 is not indexed by google.
For all intents and purposes, Everything2 sucks. I used to be a major fan. But after Wikipedia, e2 is no longer desired.
If the first Foundation - I mean Wikipedia falls, the second Foundation - I mean Wikipedia - on the other edge of the galaxy shall prevail. Encyclopedia Galactica?
Everything I need to know about copyrights I learned from Slashdot.
well, sort of.
I did a search for "revert wars" and came up empty.
So I created an article (sort of)
http://en.wikipedia.org/wiki/Revert_wars
Lets see if we can't get this puppy fleshed out a little.
Rethin
why not have a moderation system like slashdot?
Require that 5 editors approve of a content addition/change before that modification is applied.
Track the editor's moderation record. Make negative modding count both against the negative moderator as well as the moderated.
This way only by getting 5 positive mods in x number of editor views can an addition get approved.
There certainly has to be a way to handle the vandalism and pettiness. slashdot's moderation system seems to do a great job of handling just that.
I mean, as an example, cruise slashdot at +5 and you get some good meat. drop to +4 and you've got your side of fries (or potatoes), +3 to eat your vegetables +2 for fiber +1 for garnish and 0&-1 for a dark alley to purge yourself in an anorexic fit.
Just cruise the first couple posts on this thread and take a gander at what allowing anyone to post anything brings...
I know there are problems with the slashdot moderation system - but as a whole it's a good system which tends to bring the most relevant and informative posts to the top of the heap. I would venture to say the slashdot moderation system is one of the most effective user-based moderation systems in existence.
Now, since I'm not familiar (and like to read the contributions of individuals), tell me; how closely does the slasdot moderation system currently relate to the wikipedia moderation system?
as an afterthought and to browse off topic (further?) since the inception of politics.slashdot.org I have contemplated the idea of something like a debate.slashdot.org
It's quite a tricky notion to convieve - how could you setup something akin to a formal debate in the form of a web forum? I mean, it seems all the lego pieces are here, robust moderation system, informed parties abounding with great skills at backing claims.
Would you somehow create opposing teams by using a vote system? how would you determine the representative for the side of the debate?
mark my words. With slashdot and wikipedia we have only begun to see the possibilities of massive contribution of free thought.
Sanger says participants often become embroiled in "revert wars" in which overprotective authors undo the changes others try to make to their articles.
In my experience, this is not at all what revert wars are about. They're not about pride of authorship, because that's an impossibility on Wikipedia. They're about controversy. You get an article about, say, messianic judaism, or Ronald Reagan, which then becomes a battleground between believers and skeptics.
Find free books.
Larry plays a mean Donegal-style fiddle in the local Irish music sessions.
Having text subject to a moderation period for hours or maybe a day or two in a discussion area (with some sort of indicator or flag) would be a LOT better than instantaneous posting, IMO.
I contributed to the entry on Internet Explorer (specifically, removing it). A while back, some editors at Wikipedia (I'm not attributing--I'm sure this time lack of attribution will make them happy) were continually deleting the section on removing Internet Explorer from Windows. The kept changing criteria... First, they wanted the passage on removing IE to say exactly who recommends it. Then, it had to meet Neutral Point of View and attribution criteria. Then, another Wikipedia editor asked what computer security experts recommend IE removal. It finally ended; they deferred and named the three experts in the field.
Per the article: Nonbias is a difficult ideal to live up to. Indeed, the most common complaint against Wikipedia is that it is unreliable; since anyone can publish or edit any article instantly, theres nothing except the diligence of other contributors to keep favoritism, misinformation, vandalism, or sheer stupidity out of the encyclopedias pages. I'd argue that so-called nonbias is not the problem.
The problem was that these dedicated editors were not deferring to the actual experts (in this case, me--the guy who has a site on removing Internet Explorer from Windows 2000, and ignoring the creators of XPLite and nLite). If the editors don't like something, all they have to do is claim that it violates the holy grail Neutral Point of View and you'll have to beat them over their heads to get your text into the Wikipedia. Moderation is a lousy way to get at the exact truth, but eventually, it comes to light (seems to here at Slashdot, anyway). No, obviously the truth isn't what everyone thinks, but it would sure help with those editorial battles. An article might have a comment that Hydrogen caused the Hindenberg disaster, and it gets modded +5. Eventually, you can bet the comments pointing out that it was the zeppelin's skin (paint) will also get modded +4 or +5. The key is with the Wiki, with moderation, potential authors wouldn't have to have month-long running debates and editorial beat-downs.
Is sad. When anybody can change the entire entry without anybody noticing.. the "Douche" entry was insulting some girl with first and last name for about a week or two before it was changed.
Without a serious review system, I can see it becoming a nest of crap that no one will be able to use.
Not a Twitter sockpuppet... but I wish I was.
think of how dumb the average person is...
Now realize thay 1/2 the world is even dumber than that.
Much of Wikipedias last funding drive was pushed through by Ayn Rand supporters. Their motivations are unclear, however.
Slashdot:
- once you post it's set in stone
- everything is moderated by default
- mods have low power as individuals
- moderation is recursively cliqueish; moderator approval feeds back into modpoints
- system designed to force some semblance of signal into a high-noise community
- unavoidably encourages groupthink and modwhoring
Wikipedia
- everything is mutable
- moderator intervention is rare, the normal way problems are resolved is via discussion and edits
- moderation is a private club with significant power
- system assumes most people are "signal" and that "noise" is rare
- encourages discussion, reason, and NPOV
But Wikipedia's "staff" of volunteers is "better than any full-time staff you could imagine, because there are so many people involved," Sanger says. Any malicious or mistaken entry "is going to be instantly noticed" and corrected.
Bzzt. Wikipedia has a LOT of articles...430000+ in the English version alone, with varying ranges of popularity, of course. Vandalism that happens on some articles will be corrected immediately, vandalism on others could take days to languish. I've seen insidiously biased and incomplete articles that take far longer than an instant to get fixed.
What's more troubling is that people think Wikipedia is an end-all of knowledge. I wish it were, I really do. The problem is, a vandal or somebody just flatout misinformed could easily change some obscure date from like 1342 to 1324 and nobody except an expert could possibly notice and correct. From this we can derive a major problem in Wikipedia: The number of bad edits to good editors can be incredibly disproportionate, and everyone else in between won't usually know the difference.
In a perfect world, we'd seek out that information three times over before using it, and change any wrong edits back, but humans are just naturally lazy and not inquisitive enough when it comes to information on Wikipedia. In some sordid way, Wikipedia really does reflect the sum of all human knowledge. It's just that humans aren't perfect.
When someone uploads a patch to an opensource project, you have a pretty good idea of the effectiveness of that patch--it'll either do what it says, or it won't, if the new source will even compile. Bugs can be found by the sheer number of people using the software, and they're usually a lot more apparent than an unfact on Wikipedia. No information compiler exists, and it doesn't spit out warnings when you've mispelled somebody's name, transposed a digit in their birthyear, or just die when you've got something completely wrong.
I think Wikipedia would do well to perhaps remove editing by anonymous users, or perhaps introduce some sort of moderation system like those discussed elsewhere in the thread. The problem with these solutions is that knowledge is very fleeting--sometimes somebody just won't care long enough to create an account before an edit, or they might be a rare holder of some tidbit of knowledge that can't be verified by a moderator. And who's to say the moderator's correct?
Wikipedia has a vast amount of potential. Their pursuit for freedom in both beer and speech of human knowledge is remarkably admirable, and I consider them one of the best Internet charities around. Regardless of the inherent problems, I will continue to be an editor and support them in other ways as time goes on.
--sean
"[T]he single essential element on which all discoveries will be dependent is human freedom." -- Barry Goldwater
Another angle would be to allow authors to block edits of their text, but to allow others to put dissenting links in it pointing their own articles.
I wouldn't do that, because the "author" is not necessarily any more an authority than the dissenters are. And the NPOV thing on Wikipedia is very specific about *not* treating all points of view equally, or letting a very vocal minority make itself seem like an equal player with commonly-accepted ideas.
At the moment, I can't think of a better way they how they do it -- it's not chaos, because they actually do lock down articles that have become wars, and they do include reference to fringe ideas (but clearly label them as fringe).
If you haven't read their bit on the neutral POV, it's very mind-opening stuff; there's no need for the chaos, and there should be no "winner" of the edit war.
I've given this some serious thought since my earlier post, and while I'd love to see every edit moderated in some way, I don't think it's in any way practical, nor do I foresee that it ever will be.
Let's look at a few statistics, shall we?
Wikipedia's Wikistats show that for November 2004, there were over three-quarters of a million edits. That's an average of about 25,000 edits every day.
There are just over 15,000 registered "Wikipedians." Of these, approximately 1,000 have performed at least 100 edits. Let's call these people "active Wikipedians" and assume that these people all have time to moderate on a daily basis and, more importantly, are willing to moderate on a daily basis. That leaves each active Wikipedian with 25 edits each and every day that must be moderated.
Now, let's look at Wikipedia's growth during 2004. Since January, the number of monthly edits has increased by a factor of just over four. The number of active Wikipedians has increased by a factor of just over three. In one year's time, if these rates hold steady, the daily moderation burden of each active Wikipedian will increase to about 33 edits.
The number of edits is increasing faster than new Wikipedians are joining, which means this problem is only going to get worse.
In order for a moderation system to work -- I'm trying to be optimistic here -- Wikipedia would have to implement something that judged the "degree" of each edit. Edits that make large-scale changes -- where, say, more than one percent of the page changes -- would be a top priority for moderation, because it's these edits that have the most potential for destruction. Edits that simply change a character or two, copyediting stuff, wikifying, etc., would be less likely to be specifically harmful, and perhaps could be moderated at random.
Moderation, like meta-moderation here at Slashdot, could then be used to drive a karma system. The more useful edits a user makes, the higher his/her karma. After a certain point, perhaps that user's edits could be flagged as "low priority" for the moderators, because it's very likely that a user who has made many useful contributions in the past is continuing to do so.
In short, moderating every edit will never be practical, but moderation could probably be put to good use all the same. Implementation would be a nightmare, though.
p
In Korea, long hair is for old people!
The current wikipedia state...Is sad. When anybody can change the entire entry without anybody noticing.. the "Douche" entry was insulting some girl with first and last name for about a week or two before it was changed.
Without a serious review system, I can see it becoming a nest of crap that no one will be able to use.
I just went through the entire history of the Wikipedia article on Douche. I learned more about douching than I ever wanted to know. (Still, the review is much easier with the new Mediawiki v1.4, implemented in beta just this week. You can go directly from any version of the article to its immediate predecessor or successor, or you can do the same in the "diffs" that display the changed sections and highlight what was changed.) When I review the article, I don't find anything like what you describe.
The article seems to be a favorite place for the kiddies to insert people's names, but this vandalism gets reverted quickly. The first one ("Oh, and Eric's a douche") lasted all of one minute back in March before it was reverted.
Here are subsequent corrections reverting such edits, with their lag times showing how long the vandalism stayed up before it was caught:
one minute
three minutes
two minutes
seven minutes
one minute
nine minutes
one minute
Now, I'll admit, they got us this week. The vandalism that added someone's name at 2:02 on December 21 wasn't reverted for thirteen hours. I guess we were all at our Winter Solstice rituals. But there is nothing remotely close to "insulting some girl with first and last name for about a week or two before it was changed."
So, if you had added such a claim to a Wikipedia article, I'd just delete the misinformation, while giving my reasons (as above) in the edit summary or on the article's talk page. If you could back up your assertion, you could restore the passage. If you and I couldn't reach agreement, we'd get other participants involved. Here on Slashdot, with its "serious review system", however, all I can do is post this response.
Isn't Wikipedia just a subset of THE encyclopedia (the internet)?
Google's pagerank still rules the day. If Wikipedia's article on some subject is indeed the best web-wide it will be pulled to the top in search results. But that rarely happens in my experience.
So what the fuss is all about?
P.S. I wish i could exclude Wikipedia-related articles via /. preferences.
I used to be a massive fan of Wikipedia and a regular contributor in the fields of computer science, programming and military history.
However as Wikipedia has became more popular it has also became completely overwhelmed with pop-opinion, poor rigour and fact checking. It has become completely bogged down blatant bias and revisionist history, and simply trying to keep on top of this became exhausting.
At first I assumed this was simple ignorance, and tried to work withing the wikipedia process for resolution, but it was pointless, over time I came to understand that the trouble causers seemed to exhibit the same personality traits as usenet trolls and MOG griefer. The ignore facts, build straw men and resort to personal assaults. However the usual tactic of ignoring them doesnt work because they carry on changing the articles anyway, use revert bots to change articles on mass. Some examples.
- One contributor who tried to suggest that encapsulation was not a fundemental feature of OO.
- Another contributor kept removing the word riot from the blood Sunday article.
- Another contributor kept removing the evidence of JP Jones war crimes.
These are just some of the many problems I experienced at the hands of revert bots.
In the end I gave up and left them to their ignorance.
Half of the last funding drive target was "pushed through" by me, when I suggested raising it from $25,000 to $50,000.
My motivations are very simple: I estimate what I think reasonable growth based on past performance will require and project roughly what it will cost to buy the equipment to keep up, then suggest a sufficient target to cover those needs.
For the quarter now ending that estimate was three database slaves and 15 Apache web servers as the reasonable maximum we'd need based on past growth, with 2/10 more likely. 2/10 was just about sufficient and we've been discussing and I'm preparing the last of the three anticipated orders for the quarter now. Performance suffered for a while because of equipment failures (more than 5 still out of service), delays getting those computers (compatibility issues the vendor sorted out, bits of bureaucracy and timing issues largely). So we're preparing to handle a larger number of failures as well...:)
For the next quarter I'm looking at something higher. I'm expecting to be in the top 100 sites on the net during the spring quarter, with a fair probability of the top 50. Not at all bad for a place funded solely by donations from well-meaning people who want and like the resource.:)
The "big" item coming soon is ordering a new master database server to handle the English and Japanese encyclopedias, so we'll have it in test service for two months before switching to it. Followed soon by similar very capable database slaves for them. If anyone knows a place willing to donate 12-40 15K SCSI drives...?:) Or, for that matter, any fairly fast drives, including drive maker refurbs, since everything is RAID. Or anything in the way of quite high end disk systems or high capacity RAM modules, for that matter. It's a fine opportunity for high profile public good PR.
Japanese is paired with English because Japanese load is falling while English is rising and vice-versa.
Instead of having one ULTIMATE explanation, which the original author 'restores' continuously, we could have 'alternative' pages for each topic.
Readers would be able to rate these (like on Amazon 'was this review useful to you?'). When you search for an item, only the top three or so would be shown, with a link to see all of them.
Imho this would NOT lead to an abundancy of pages, because for non-controversial topics no-one would be urged to give an alternative explation for e.g. 'DNA base pairs'. For controversial topics, alternative viewpoints would exist next to each other, instead of intertwining and damaging each other. I can imagine people love their 'wikibaby' so much, and try to restore it every time, but hopefully no-one would go so far as to intentionally destroy others work for the sake of it (e.g. to decrase its rating). Besides, destroying others' work is also possible today.
Z
A validation system is in progress. Ideally, it would have a sort of advogato-style trust metric, so that community consensus would ensure all validated articles were of a particular quality; more trusted users would have a bigger vote towards validation.
Thus, there would be the 'live' version and the 'validated' version, trailing a short interval behind the live one.
Check out test.wikipedia.org for a really shitty implementation of validation. (It's vulnerable to all the same problems that editing is, thus providing no additional benefit, and a kludgy interface to boot. But validation could do what you say, in a scalable and extensible fashion.)
--grendel drago
Laws do not persuade just because they threaten. --Seneca