A big problem with Wikipedia is that the head editors have no clue about the subject they control. So when people come in who know about the subject and try and edit they get shot down. There was someone from the Peirce-L mailing list who did a dissertation on Peirce and was passing all the changes to the list (consisting of some of the top scholars in the world on the topic). When he posted the corrections to Wikipedia they were rejected because they didn't match some textbook the guy had at MIT that mentioned Peirce in passing.

I've seen this in many other subjects as well. (Read the edit comments) It makes me like Wikipedia for an initial look but it's worthless for anything else. I've just found way too many pages in areas of interest I have with errors - often egregious errors.

I've been lucky; not enough people were personally invested enough in the Hittites to run over my Wiki work there.

Although, I did once write a bio of some pol who sent me junk mail, and referred, in passing, to her opponent's "pro-abortion" stance. Sure enough, one of our wisers and betters drove by and changed it to the Appropriate Usage, "pro-choice" - other than THAT, the editor didn't change a thing.

Remember how Orwell, Bierce, White, and other paragons of proper English taught us to keep our written thoughts free of cant? If you are going to contribute to Wikipedia then FORGET ALL THAT.

this is one rather incoherent post. Wikipedia is great as an overview for a topic you are not familiar with at all. Wikipedia is not a standard reference textbook for your sub-sub-sub specialized field of research. Wikipedia is not for scientific research.

When you contribute anything to Wikipedia, you are basically writing for a broad audience.

The reason academics don't contribute as much as you would want it is because they have no incentive to contribute. (1) will not help with tenure. (2) Takes too much time to get community cred.

It seems like only people who really hate Wikipedia are authoritarians who have very little technical (read: computer science) background.

BTW, there is a very direct, useful analog to Wikipedia in another domain -- open source software projects.

It seems like only people who really hate Wikipedia are authoritarians

It wasn't my fault -- Darth Vader commanded me to criticize Wikipedia. He said if I didn't, Bushitler would send black helicopters to my house. O freed slave, help un-plug me from The Matrix!

Agnostic, like I stated, it doesn't seem like you have a tech background at all -- decentralization is like a key meme of the past decade or so. Wikipedia takes advantage of that meme. Linux took advantage of that meme.

The basic model of science does not apply to every single human endeavor. It seems like you got confused along the way and think if its not science, its not useful.

If you want to understand what Shirkey is saying, I suggest you start by reading about the history of Linux and GNU tools. Like I said, you are missing the entire ship.

Anyone who thinks there is a cognitive surplus has never hit a really good topspin forehand.

Jor -- I made a joke that you didn't get. My post is a response to that moron's claim that editing Wikipedia is like a calling in life, rather than some hobby. Putting time into Wikipedia is the smartie version of watching TV -- mostly a waste of time, when you could be producing something real.

I don't care if something has been a "meme" -- that's a fancy word for "fad" or "fashion statement." You're right, though: it is very fashionable among geeks to talk about decentralization.

I will never read about the history of any geek obsession -- Linux, Wikipedia, World of Warcraft, or anything else. I've already started on my summer reading list: Stability, Instability, and Chaos by Glendinning; and Applied Partial Differential Equations by Ockendon et al.

As the post suggests, I prefer being fruitful rather than masturbatory.

And obviously I don't think it's "do science or do nothing" -- Wikipedia nerds could also be painting, composing music, running a business, engineering a safer car, and so on.

But since they don't do those things either, and have a tendency toward not wanting to train hard, I figure the "do it yourself" studies I propose will be easy enough for them to carry out.

If not, then they can waste away. That's fine by me. Other readers -- probably people who are already productive, just in other areas (again, like business or engineering) -- may do them instead.

My post is a response to that moron's claim that editing Wikipedia is like a calling in life, rather than some hobby.

Shirkey is not saying this. That's why I am saying you obviously completely missed his point. Shirkey's point is that information technology allows us to collaboratively turn some of our mental downtime / hobby time into something very, very useful. And not just a personal skill, but something that can be extremely useful for everyone. For instance, it would be very easy for me to write / edit a wiki entry on anything I'm an expert on. Instead of watching another re-run of friends, I might one day decide to spend 20 minutes and do just that.


The reason I say study the history of Linux is not because its an interesting technical story. Linux is an interesting story in human organization. The utility of Linux is also very easy to measure in $$'s (unlike wikipedia), so it makes the case extremely crystal clear. The main points behind the story of Linux is a chapter in several recent business books (e.g. the wisdom of crowds), so its not a large commitment.

I stand by my original points, since arguing with computer geeks is pointless.

I'll only add that Linux can't be what you say or else it would be widespread in use -- ah, that's right, the Devil-god conspires for the world to use Windows. The wisdom of crowds, if it applies here at all, shows that Linux sucks -- that's why any crowd that matters ignores it.

(I'll delete any further geek propaganda of yours)

Wikipedia really doesn't do anything that an encyclopedia doesn't do, except be free. Its ability to be freely edited means that, for every knowledgeable person posting, there are a dozen morons - some high up in the Wikipedia hierarchy - wrecking it.

It has all the problems of democracy as well as all the advantages. Reality, however, is not a democracy.

and so are bloggers like Steve Sailer, Audacious Epigone, Half Sigma, Inductivist, and others who obtain and analyze carefully select data to answer a question or hunch

Agnostic: Linux is not a desktop operating system -- its a server operating system. Linux has been widely adopted by IBM, Google, Oracle, Sun, etc. Another free unix variant, developed on a similiar model to linux is the back-end to Mac OS X. If that isn't success, I'm not sure what is. Further, that success is not limited to the tech sector as many of the top business journalists have repeatedly spoken about the linux model of organization (James Surowiecki, being one who easily comes to mind). This is not "geek propoganda". Surowiecki is not a geek, he's one of the preeminent business journalists in the country.

BTW, My background is in machine learning, genomics, and medicine.

Caledonian: There is an article in Nature about the comparable accuracy of Wikipedia vs. Britannica. Feel free to look it up yourself. Although the article has some flaws, obviously, things are not as bad as you think.

Yes, we've all heard about that study. But I have direct experience of using encyclopedias, and direct experience of Wikipedia, and Wikipedia tends to be grossly wrong far more often than the encyclopedias.

You really, really like the idea of Wikipedia. We get it. Now stop stumping.

Wikipedia has two advantages: free and convenient. I frequently use it. In many of my areas interest (e.g. the Qarakitai) I find it to be up to date and accurate. In a few areas, not so good. To me it's most useful in areas of peripheral interest, where I know enough to read critically but not enough that the article is unnecessary.

It's pretty much up to Wiki critics to do better. A free, monitored, peer-reviewed Wiki by top people would be better than Wiki, but it doesn't exist. Academia seems to like to do the opposite -- put articles of very specialized interest behind a $1 / page pay wall (but without giving the money to the author or using it to sponsor scholarship.

John Emerson: One exception to that trend is the Stanford Encyclopedia of Philosophy which is quite good but somewhat spotty in what topics have articles.

Linux is not a desktop operating system -- its a server operating system. Linux has been widely adopted by IBM, Google, Oracle, Sun, etc. Another free unix variant, developed on a similiar model to linux is the back-end to Mac OS X. If that isn't success, I'm not sure what is.

OK, let's take a look at how well Linux dominates the server market:

http://www.linux-watch.com/news/ ...5369154346.html

Linux servers posted their second consecutive quarter of double-digit growth and now represent 12.7 percent of the overall server market

According to IDC, this was the first quarter since IDC began tracking Linux server spending in 1998 that Windows server revenue has grown faster than Linux server revenue. Windows comprised 38.8 percent of all server revenue in Q1 of 2007.

agnostic: MSFT is the 3rd largest corporation in America with a market cap of 300 billion dollars, and revenues of 60 billion per year.

Linux was created and is mostly maintained by a distributed loose nit group of developers. MSFT's revenues are twice NIH's annual budget. Somehow, a loose nit group of net developers is able to create product that has 1/3 installed base as MSFT. I think your own comments suggest success.

In terms of IDC numbers, Linux doesn't compete in certain server segments, that are locked down by microsoft's proprietary products (Exchange, Active Directory, etc.). If you want to get an accurate assessment of Linux vs. Microsoft as a server os you have to look at sub-segments of the server markets.

Linux is not the only open source project either, c.f. Apache (50% market share) or Firefox (20% market share).

You previously stated: . My post is a response to that moron's claim that editing Wikipedia is like a calling in life, rather than some hobby.. This is COMPLETELY WRONG.

There are tons of stuff written about this mode of collaboration in the business press and on the bookshelf. I can't explain it to you over blog posts. PICK something up and read it.

BTW, in terms of your projects (2) has already been done very thoroughly in medicine (c.f Ionnadis). (6) There are too many physicists already trying to do sociology and get published in nature and science with results of questionable utility. (that was circa 2000-2003, c.f. barabasi)

There are too many physicists already trying to do sociology and get published in nature and science with results of questionable utility.

But most physicists don't get people -- they get things. The readers here are what Steve Sailer calls "people nerds" -- those who have a good intuition about how humans and their societies and culture work, but who have enough quantitative skills to flesh out a good model.

Per the rule, if you see a good fight going on, get in and get yourself some, Linux is a serious operating system. It was actually crap as of maybe ten years ago. but now it works great. The great advantage that Windows has is that it's a lot more user friendly, as in the user need not have any clue how a computer works to use it.

If you follow such things, Sun Microsystems recently changed it's stock ticker from SUNW to JAVA because it isn't going to be in the hardware business all that much longer. I used to use Suns and didn't want to switch because Sun had a large reliability advantage for the same reason Apple has one over Windows, because they made their own hardware and so didn't have any third party drivers written for peripherals screwing things up. Sun hardware wasn't as good as an off the shelf PC though, but if one wanted to run Unix on an off the shelf PC, one had to run Linux. I had to migrate to Linux PC's because one of Sun's new chipsets had a nasty little bug I found in it concerning calling a isnan function that made it slow to a crawl which made it useless for my high powered number cruching apps. So I switched to Linux, and it's been great! I run Linux on a few 8 processor machine with 26 GB of RAM that if they failed, serious money would be lost. Linux works great, if one is doing some sort of real computing, as in computations, it's the best there is (not graphics or eye candy stuff).

Sun is going to be a software company that sells JAVA because Linux drove them out of the hardware business.

I must admit, I have Windows PC's for stuff like email and web surfing to GNXP.com.

I just looked up the market cap for Red Hat. It's currently $4.2 billion. Agnostic must be Bill Gates, since he thinks that's chump change, and the content of his comments on Linux just prove it :).

Wiki's basically a big message board. It's fine for what it is, though obviously if one is unhappy with what has found there, one can get one's money back.

I also want to say that in my experience wikipedia articles, at least on subjects I know something about (physics, math and CS), tend to be excellent resources. Also, find me one print encyclopedia that has articles on topics of broad interest, but also has technical articles covered with equations.

Yes, we've all heard about that study. But I have direct experience of using encyclopedias, and direct experience of Wikipedia, and Wikipedia tends to be grossly wrong far more often than the encyclopedias.

As in, "the study has to be wrong! I have...anecdotal evidence!!!!"?

Something we can all agree on about Wikipedia

jor, did you pick that pic to appeal to assman's preferences?

Seriously, look at that Perso-Indic schnoz! Now if only the slogan were across her butt... but then most Wikipedia nerds are probably sissy boob men.

The cool thing about Wikipedia is the ability to get experts in obscure fields to edit or add articles. In the Shirkey talk (which seemed to me to be a lot of handwavy bullsh-t connected to a couple really nice insights), the idea was that people could take a bit of time that they were currently spending unproductively and spend that time sharing their own obscure but deep knowledge of something. Of course, the background for this pretty much has to involve that same person having spent many, many more hours acquiring the expertise they're now sharing. Editing the Wikipedia entry for cryptographic hash functions or cytotoxic T cells is relatively quick, but learning enough in either field to do a good job of it consumes a lot of that prime sitcom-watching time.

Agnostic,

I'm not sure if I should call your modesty false, as it might be furtive self-promotion ("You think that child's play is worth reading? Do you know what I really write about?"), but your 'leisure' work is fruitful. I suppose the post linked reveals what intuitive thinking should lead one to conclude. But that 'banality' summarizes a large portion of Steve's posts, and he's the most insightful public intellectual/social pundit I know of.

As in, "the study has to be wrong! I have...anecdotal evidence!!!!"?

As in, "it doesn't matter how accurate Wikipedia is on articles that I'll never read, and on the subjects I use it for, it's riddled with elementary factual and contextual errors that no self-respecting encyclopedia would tolerate".

I will further note that most studies are garbage, and that selective quotation of the garbage permits virtually any position to be defended.

The best defense of Wikipedia is its cost, not its content.

"You think that child's play is worth reading? Do you know what I really write about?"

Ah, but you chose an unrepresentative post to highlight -- whereas most of what I've focused on recently is the appeal of teenage and college girls. There are some OK gasbag posts, but I haven't done any good math / data stuff in a long while (other than figuring out the age distribution for porn stars, and hence when females are at their hottest).

I'm afraid Steve Sailer is going to change my label in his blogroll from "quant" to "pervert."

Maybe some reposting some quantitative stuff from the past would be worth it...

agnostic: Perhaps he could merge the two categories, just for you. Mathematically-oriented pervert? Quantitative pervert? Perverted quant?

The best defense of Wikipedia is its cost, not its content.

Wikipedia's content in certain areas has far more entries (and will be more up-to-date) than what any conventional encyclopedia can provide, e.g. pop culture, video games, anime, and anything computer-related. Of course, not everyone finds this stuff useful.

Wikipedia has a lot of interesting hobby posts. For example, a list of the names all the Swedish noble families together with their century of ennoblement.

Maggie Gyllenhaal is the Swedish equivalent of a baroness.