Slashdot Mirror
Improving Wikipedia Coverage of Computer Science
Pickens writes "MIT computer scientist Scott Aaronson has an interesting post on how to improve Wikipedia's coverage of theoretical computer science. Aaronson writes what while Wikpedia will never be an ideal venue for academics because 'we're used to (1) putting our names on our stuff, (2) editorializing pretty freely, (3) using "original research" as a compliment and not an accusation, and (4) not having our prose rewritten or deleted by people calling themselves Duduyat, Raul654, and Prokonsul Piotrus,' he identifies twenty basic research areas and terms in theoretical computer science that are not defined on Wikipedia, and invites readers to write some articles about them. Article suggestions include property testing, algorithmic game theory, derandomization, sketching algorithms, propositional proof complexity, arithmetic circuit complexity, discrete harmonic analysis, streaming algorithms, and hardness of approximation. One commenter suggests that professors should encourage students to improve the Wikipedia articles about topics they are studying. 'This will help them understand the topic and at the same time improve Wikipedia.'"
Essentially all that you have to do (or should have to do) to avoid the "original research" claims is to cite sources. It's not intended to be treated like some sort of scientific journal, it's intended to be an encyclopedia; everything put in the Wikipedia should have been published elsewhere first. Seems reasonable.
It's good to see that somebody in academics is appreciating the importance and usefulness of Wikipedia, instead of ranting about inaccuracies and trolls.
Now let's resume our program of bashing Wikipedia.
Knuth is a fan of Wikipedia, but he's a bit leery of the concept, saying that he would not want to have to remain forever on guard after making technically complex contributions, lest his comments be badly reedited.[citation needed]
Do you even lift?
These aren't the 'roids you're looking for.
"Prokonsul Piotrus" aka just "Piotrus" is a rather controversial figure. He has been bought up in not one, but TWO arbitration cases, one of which is now in the voting phase.
I stopped trying to add any content to Wikipedia years ago. WP:Notability is, quite possibly, the worst thing to ever happen to that website, and I got sick of deletionism bullshit.
A problem to watch out for is that if you add your own research to Wikipedia (even with all the proper citations), you'll get slapped by some self-important wikipedian because it is apparently wrong and evil to have the person directly responsible for the research itself to be included in the creation of encyclopedia content about said research.
Hexy - a strategy game for iPhone/iPod Touch
He needs to create Wikiresearchia or something.
Or submit the articles to DocForge where original research is allowed. It's focused completely on programming and computer science topics. It hasn't grown large enough yet to breed overzealous editors, either.
Developers: We can use your help.
A while back, about a year ago, I spent my time correcting wikipedia - the corrections I made were accurate, meaningful, and relevant to the topic. However, my additions and changes were mostly removed within two hours of my posting. Perhaps those who run wikipedia do not like my educated improvements. One incident that sticks was when a friend and I added a section dedicated to the problems with genetic algorithms; by the next day it was removed. I had sources, a good and well written arguement, and it was fairly long and not biased (at least my professor thought so).
As for adding new topics, one may try, but seeing as additions are not appreciated, than what would become of new articles (even stubs)?
I'm not sure that Aaronson really gets it regarding original research and putting his name on it.
Surely, it is meant to work this way:
1. Researcher publishes research in reputable peer-grouped journal, and makes this paper available on the Web.
2. Researcher writes nice, easily digestable Wiki page on the topic, citing the peer-reviewed research as a source.
The Wikipedia prohibition on 'original research' is really a polite way of saying: 'don't assert things that could simoly have been pulled out of your butt'. The reliance on peer-reviewed external sources is supposed to get around this problem.
----
Anyone know why my posts recently started appearing with Score 1, despite "excellent" karma? I'd love to know.
LINUCES!!!
We CS people gotta keep our prose before hoes.
He's got a list of complaints which is completely the wrong list. Essentially he seems upset about (1) not getting a byline, (2) neutral point of view, (3) no original research, and (4) having what he writes modified by others. Well, sorry, but those are all basic features of WP. They're not gonna change, and IMO they shouldn't change. WP has problems, but the problems are not on this list.
In my opinion, the biggest problems with WP are (1) the poor quality of the writing, and (2) the tendency of the quality of an article to get worse over time, rather than better. Problem 1 is particularly pronounced in my field, which is physics; most of the physics articles read as if they were written by smart grad students who wanted to show off how smart they were. If there was going to be a #3 on my list, it would have to do with the factors that make me personally feel like working on WP has gotten about as pleasant as a proctological exam. But that's really not a problem with WP, it's just a problem that makes me personally not want to work on WP. Plenty of other people still seem to be happily maintaining it, which I think is great.
Find free books.
what, you mean like how p != np?
Those of us who think they know everything annoy those of us who do.
"we're used to (1) putting our names on our stuff, (2) editorializing pretty freely, (3) using "original research" as a compliment and not an accusation, and (4) not having our prose rewritten or deleted by people calling themselves Duduyat, Raul654, and Prokonsul Piotrus"
That really says more about academics than wikipedia. /.). I don't know what 'duduyat' means, could be a name, 'raul654' *is* a name, followed by numbers probably to be a unique ID, prokonsul piotrus seems to be a rather obvious nickname. Or, as some might call it, a pen name. I guess he wouldn't like anything by Mark Twain, Lewis Carroll, Richard Bachman, George Eliot, Andy McNab, Ibn Warraq and many others. Appeal to authority is a fallacy.
(1) Actually means "have our names displayed prominently", since wikipedia does store one's name (if one desires it).
(2) Means he wants a lower standard. Wikipedia is far from perfect, but he wants it to be *worse*.
(3) Is missing the point, wikipedia is an encyclopedia, not a science journal. If he doesn't know that, he's not much of a scientist, is he?
(4) Names are not important (that's why I have almost always posted as AC here on
Most academic issues are handled worse than computer science.
Most of the CS coverage addressed on wikipedia is the kind of stuff that working computer programmers would be interested. There are a few theory articles, but you can't expect much from them. Writing in CS theory or other areas in mathematics is difficult, and requires more than citations. It requires strong writing and editing skills, and strong understanding of the subject at hand. I wouldn't expect to get more than a rough overview of a field from its wikipedia entry.
If they and their students write a Wikipedia article in exactly the same way as they write an academic "literature review," they will have no problems at all.
Literature reviews presents no original research; provide some interpretation and context but no personal opinion; and cite sources for every fact. Just like a good Wikipedia article.
"How to Do Nothing," kids activities, back in print!
it is apparently wrong and evil to have the person directly responsible for the research itself to be included in the creation of encyclopedia content about said research.
Good thing Wikipedia is a Wiki and not an encyclopedia then.
Mob mentality rules on a social encyclopedia. If 1 experts knows something to be true, 100 idiots who agree with each other can rewrite their own version of that truth.
hate to say this but google's knol service might just be what the doc ordered.
My remarks on the comments:
(1) You're only researching published material
(2) You're only researching published material
(3) You're only researching published material
(4) Fair enough.
The remark that original research is an accusation on Wikipedia is correct but it's not really relevant. I'm sure a science fiction author wouldn't complain about not being able to make stuff up in a science journal, and I'm sure that physicists are quite happy that there are no tolerances specified in cookery books.
Could Aaronson explain why "improving Wikipedia" is a laudable goal?
Of course it might be beneficial for an undergrad to write essays on the given topics, but contrast, say, scientific history articles in Wikipedia vs those written by practicing academics in the Dictionary of Scientific Biology. To suggest that an undergrad essay is a suitable reference - to even suggest that it is a suitable jumping-off point, when the undergrad's knowledge will lean heavily toward the limited knowledge and reading material recommended by his tutors - is damaging to academia and to the interested layman.
What is more, Wikipedia's free-for-all editing model is known - even by Aaronson, it seems - to permit his edits to be treated with the same respect as hAX0RKID3000's. The hard part of Wikipedia is not writing something but making sure it doesn't rattle someone's cage. I can't think of much going for Wikipedia apart from the fact that it's popular and it's big. Taken from this angle, wouldn't students be better joining the American Football team or similar?
I'm glad to have heard either rational criticism or appropriate mockery from faculty on this side of the pond, from Oxford to the local ex-poly.
You know, every time there's a Wikipedia-related thread on Slashdot, there's a massive run of people with anecdotes about how they spent hours and hours improving some article only to have it reverted.
I've never once seen someone post a link to the changes they made.
Please tell us what article it was, and what corrections you made. If you go to the article's history you can post a link to the exact changes that you made, and the subsequent reversion. It'll take two minutes, I swear.
You don't even have to go through all that. Just post your user name and the article title and we can find it ourselves.
It would prove once and for all that Wikipedia is as bad as everyone says it is. I'd love to see it. We'd all love to see it. Then we can fix it and make sure that your corrections actually get implemented properly.
Because otherwise you, like everyone else here, are just posting the equivalent of "my friend's friend died from eating Pop Rocks and Sprite." Baseless accusations that don't help anyone.
I am naming and shaming them here. If Wikimedia wan'ts their 6 Million donations, they better start banning deletionists and breaking up the cabals of rogue admins. They are worse than the vandal known as Willy on Wheels! Willy on Wheels only moved pages on wheels, not delete free content knowledge. Look at their contributions to see what I mean.
TTN, MER-C, OrangeMarlin, Oxymoron83, Antandurus, Luna Santin, Alison, SpaceBirdy, CometStyles, IronHolds, Betacommand, Spellcast, Lucasbfr, JzG, Standstien, and more.
I was a former editor with over 5000 edits before I left due to deletionists, so I know the deletionist gangs first hand.
Not notable. Consider for deletion.
My British Lit professor was always pushing us to use Wikipedia as a source for papers and content. After reviewing the list of contributors to the areas that he wanted us to read I found that he was a regular contributor. The point he knew the entries that he was taking us too had correct information because he made sure of. I think what the article is saying is the same thing. Rather than knock it down academics, or at least their grad students, should be making an effort to update the entries regarding Theoretical Computer Science so that the information is viewed as hear say.
Save Pangaea!! Stop Continental Drift!!
Well-known theoretical computer scientists without Wikipedia pages
No, I'm not going to make that list ... but you can.
If Scott Aaronson was hinting at himself here, which I assume he was, it seems that somebody just took the bait.
This really comes down to the distinction "Encyclopedia" (read: "A book, or set of books, or digital version of such, containing authoritative information about a variety of topics, arranged in alphabetical order") vs. Hitchhiker's Guide to the Galaxy (read: "A book containing the compendium of Life, The Universe and Everything, notable or otherwise, as written by everyone with half an interest in writing it.")
Wikipedia intends to be a new-agey digital Encyclopedia, which includes academic drive, unavoidable deletionism, well-cited sources, and some kind of drive for neutrality (no matter how badly it actually fails at such a thing).
What we need is a real-life implementation of the Hitchhiker's Guide. It should be far less careful than Wikipedia (and likely should be a superset of Wikipedia with all of those fun lists like "Things Gregory House has written on his whiteboard on House M.D.") The two sites really should work in concert (i.e. when something gets "demoted" from Wikipedia, it should slide into the Hitchhiker's Guide).
The third effort of having a even-more verified-and-factual Wikipedia is already underway via several projects. Why hasn't anyone looked into the super-set?
IOW, fork the damn project, and make it INCLUSIVE, rather than their fscktarded "OUR knowledge, or NO knowledge" authoridigm.
Also, their obstructionist habit of deleting obvious-truth whenever they don't have an acceptable-to-them reference...
it should be deletable if it's ( perhaps easily ) provably false, not if they don't/won't accept obvious truth...
Bogons...
NATURE is inclusive, ECOLOGIES are inclusive/open, they want their cathedral, while **pretending** to be a bazaar.
One commenter suggests that professors should encourage students to improve the Wikipedia articles about topics they are studying. 'This will help them understand the topic and at the same time improve Wikipedia.'"
How is bringing thousands of people into the mix who don't know what they're talking about (many of whom think they know everything) supposed to improve anything?
Encouraging your students to go "improve" Wikipedia articles isn't encouraging them to speak up, seek knowledge, or debate.
At Wikiversity, a sister project to Wikipedia, theoretically, most of those:
(1) putting our names on our stuff, (2) editorializing pretty freely, (3) using "original research" as a compliment and not an accusation, and (4) not having our prose rewritten or deleted by people calling themselves Duduyat, Raul654, and Prokonsul Piotrus,'
Are at least somewhat mitigated.
Should not be so much of an issue. 1: You pretty much can. 2: You pretty much can. 3: You pretty much can. 4: Less of a problem. We are trying to make this not be the case. We also offer certain protections Wikipedia does not.
contribute at wikademia
That's probably because every student in my computer science department starts their research at Wikipedia. Honestly, the Wikipedia articles are easy, readable, and accurate. It's usually after reading Wikipedia that I can go back to my Algorithms textbooks and understand what they're saying.
"""
One commenter suggests that professors should encourage students to improve the Wikipedia articles about topics they are studying. 'This will help them understand the topic and at the same time improve Wikipedia.'
"""
This would actually be counter productive. Student, just starting to study something, have the horrid habit of thinking they understand something that they actually don't. Students, in general, should be told to avoid writing about subjects they are studying, not encouraged.
I have submitted a project to googles 10^100 compatition that will solve the 4 problems with wikipedia he lists.
The project name is wikixandria and the idea is to make a p2p wiki library.
If the project wins 10^100 we will soon have an academy-friendly alternative to wikipedia.
Also lets remember that is also possible to store knowledge in a knowledge base. Like they do at true knowledge and the let a computer answare our questions.
Also take a loke at my "p2p" news-site crowdnews.eu
Using nationalistic Polish sources that have been tied to antisemitic statements Piotrus pushes articles like http://en.wikipedia.org/wiki/Massacre_of_Brzostowica_Mala. This what the gigantic ArbCom cases like the one Piotrus is involed in are about. Make sure you read the deletion discussion http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Massacre_of_Brzostowica_Mala to see what the horrible deletionists tried to do. Thankfully, no consensus defaults to keep on Wikipedia, so a handful of nationalistic blokes can keep rewriting history.
"I've never heard about this so I'll just speedy delete it"
Firstly, in the majority of style guides numbers are enclosed in two parenthesis like this (1).
Secondly, don't end a sentence with an exclamation mark unless it is to: (1) indicate strong emotion, which yours isn't, as it is listing reasons or; (2) it is an interjection, which your sentence isn't either, as it leads into a cohesive whole.
Thirdly, a comma is supposed to be found before 'or'. So your first reason should be "learn to use a period, or a comma."
Fourthly, your third point that the poster sucks is ad hominem. An informal fallacy; not a reason.
In finishing it looks like you have no idea about writing, and no idea about reasoning either.
Thanks for playing you illiterate schmuck.
Citizendium looked like a great idea until they decided to dump the Wikipedia content and start from scratch. So now, according to their front pages, Citizendium has 8,700 articles and English Wikipedia has 2.6 million. If you want to look something up, chances are it won't be in Citizendium. So you go to Wikipedia instead. And we all know everybody else goes to Wikipedia as well.
If you have a contribution to make, why bother with Citizendium? Chances are nobody'll read it. Academics like their names on things but they also like those things to be read. If you contribute to Wikipedia, the worst thing that can happen is that it gets reverted, and nobody reads that either.
When a new project forks Wikipedia while fixing its organizational problems, then it might attract the best academic contributors. It has to fulfil the following criteria:
Then, smart people can contribute in the hope that the whole project won't get dumped in favor of Wikipedia's established content. The new project can benefit from enhancements to Wikipedia. And contributors to the new project can hope that even if it does die, their changes will have as much chance of surviving in Wikipedia as if they'd made them directly. All of this won't be easy to get right, but they're similar problems to distributed development, and computer scientists are the best placed to solve them.
For now, Wikipedia may be inefficient in all kinds of ways, but it's also an extremely successful project. It has a lot of good content, a lot of contributors, a lot of readers, and a lot of momentum. A rival can't ignore all that.
What this guy seems to be complaining about is that Wikipedia doesn't have enough coverage of the parts of computer science he's interested in. His field seems to be probabilistic algorithms. This is a relatively new field. The general idea is that there are problems for which an algorithm with some randomness is much faster than a deterministic algorithm, at least for the worst case. Classic examples are the simplex method in linear algebra and the traveling salesman problem.
In the last twenty years, there's been a fair amount of work on generalizing this concept to a wider class of problems and putting a theoretical basis under it. Wikipedia has some articles on this. See BPP. I'm not up on this stuff (my CS degree is too old), but at least there's some coverage.
Because this is an area where there aren't many useful results yet, it's hard to write encyclopedia articles. It may be too early to tell what's useful and what's a dead end.
I'd be surprised if, say, Encarta even mentions this subject.
Aaronson writes what while Wikpedia will never be an ideal venue for academics because... [blah blah blah snipped]
Then don't use goddamned Wikipedia.
"How can we use this medium over here that is set up completely opposite to what we need?"
Start your own repository. I thought you MIT kids were supposed to be thinkin' good.
A lot of the maths and science articles had been written by college professors and the like. How else do you think they would know so many things about 322_(number)? just check the article yourself, and you will see:
It is a triangular number and the sum of a pair of twin primes (149 + 151), as well as the sum of ten consecutive primes (13 + 17 + 19 + 23 + 29 + 31 + 37 + 41 + 43 + 47). It is a Harshad number.
why would the average person bother to spend the time to contribute, if all he wanted to do was attack? There are much better places for that.
sorry, should have typed 300_(number)
Here's the Deletionist smoking gun:
Here is what was deleted from Wikipedia preserved on Deletionpedia:
It should not have been deleted.
CHALLENGE: Track down how it was deleted, why it was deleted, who was involved, and what was the motive for the deletion.
Every once in a while I try to look up a known concept on wikipedia, and I am simply shocked by the lack of any sort of useful information provided on some of wikipedia's articles on econometrics and statistics with the former being usually much much worse or often non-existent. (for example try to look up "GMM". Surely what's written there does not give the subject the justice, even by encyclopedic standards).
Lets hope that they will at least include an introduction that can be read by the layman. Already I can see that pages containing cryptographic protocols are squarely aimed at mathematics. This while most people looking for such pages are users of the algorithms rather than researchers. Having fully correct pages is one thing, being able to read them is another.
I've created a perfectly fine, fully HTML-4 compatible user manual (for end users) once only to see it demolished with technical terms by a manager. It was missing some parts that were part of the contract. Unfortunately nobody of the target audience would understand these terms. A completely correct and completely unreadable manual was the result (in Word-98 "HTML" no less).
In short, I'm a bit afraid of university professors and students filling in articles. Sure, they will be correct. But will the be readable (== useful)? Hopefully they will keep their target audience in mind. Even if that just consists of other professors, there's the question on what notation to use/mention.
It was deleted twice. See http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Ulteo and http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Ulteo_(2nd_nomination) . You can see the deletion log at http://en.wikipedia.org/wiki/Ulteo . I'm not sure that deletion was correct. There was only one source that seemed to be arguably non-trivial which was the review on linux.com The article was then put into userspace to be improved (you linked above to where that was connected to Gigglesworth). A few months went by and no improvement occurred so that draft was deleted. Looking at the reviews posted it may be possible to write an article that meets Wikipedia inclusion standards but it doesn't seem like a very strong attempt was made.
If you're creating and publishing new works in Wiki format, that's what Wikibooks is for. Then that whole problem not citing sources problem is solved. Unlike Wikipedia, Wikibooks is intended for some types of original work.
Still I'm not sure how Wikipedia would handle a wikibook being cited as a reference. (If you must cross cite the two, I'd suggest keeping the details in the book and just a brief mention in the encyclopedia article.) But with a wikibook, there's at least a ready and pubic accessible/revisible medium for distributing the information. Seems like a good place to document emerging technology where new terminology, revisions, etc. happen fairly fast and and aren't likely to make it into printed work in a timely manner.
It's not a freaking conflict of interest to have a specialist in a particular topic insert information. Jesus, that's the kind of BS idea that keeps me away from Wikipedia.
Obviously there is a demand for a wikipedia-like place to post original research, speculation, and hearsay (with a loud disclaimer, of course). If such a site existed, then the battle over citations will mellow out.
Everything2
Wikipedia = Wisdom of Crowds aka Direct Democracy
Democracy != Meritocracy
Hence Wikipedia may consider hiring the services of experts to inject merit in its content.
I'd like to buy homeland for our 10 million people. http://twitter.com/mahadiga