h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

There really is no standard of measure for the success of an encyclopedia. Popularity is nothing, it can be here today and gone tomorrow. Unfortunately, so can everything on Wikipedia. I expact the site to get largeer and larger and it's articles get longer and longer.

Even now, well-established articles are hostile to new content, and people spend more time arguing over contributions than contributing. Imagine the Peer Review system, only every single article is on Peer Review perminantly and none of them ever get accepted.

And THEN you can start to worry about vandalism.

It's utter chaos, quality is drooping, and people who come for quick facts are starting to go back to searching Google again.

One cannot measure the success of a site in terms of popularity, and one cannot measure Wikipedia in terms of accuracy, because any inaccuracy you find you can instantly change. To go through Wikipedia and count all the inaccuracies is impossible, and it would be much more worthwhile to go through fixing things, and that is what Wikipedians in fact, do.

This is all well and good for keeping track of ficticious Universes but in my opinion, Wikipedia falls short the real one.

Probably it's the socialist economic model Wikipedia is based in. Nothing is owned, everything is public.

Utter chaos.
Nerd42

and furthermore, please add http://en.wikipedia.org/wiki/h2g2 to this article. thanks!
Nerd42

... and people might also be interested in this
http://en.wikipedia.org/wiki/Template:H2G2Researcher
Nerd42

I agree with pretty much all of this post, but would add that Wikipedia's real problem is that none of its policies on civility and general behavior are enforcable. The result is that the most persistent, rather then the most correct, tend to be the editors that hang around. Anyone with a bee in their bonnet allied to stubborness will win out - often at the expense of accuracy.

See the article on Walter Sickert for an illustration: hopelessly disproportionate attention given to the Ripper theory. Then see how comprehensively discussion has failed to correct this, and why.

I've added a link to the Wikipedia article about h2g2.

OH good

I've been thinking about starting an online encyclopedia project that all it basically does is encourage other online encyclopedias to link to each other, and especially to each other's entries on each other. Is there alreasy such a project? Does anybody know? If not, I think I would very much like to start one!
Nerd42

1 million articles on Wikipedia English edition as of yesterday.

Isn't that a measure of success?

Frankie

Traveller in Time writing raw entries
"We are close to A10000000, as in: we also have nearly a million entries.

But we have and a "

Number of articles as a measure of success?

...

Given Wikipedia's goals, I don't think so. If that were the measure of their success, then pages wouldn't get deleted so often.

But anyway, I started a project with that goal I mentioned: http://wiki42.wikispaces.org/
Nerd42

I've got no problems with wikipedia being successful...they more it grows and the drier it gets the more its sounds just like the Encyclopedia Galactica, where as the more h2g2 grows and the better it gets the more it sounds like the Hitchhikers Guide to the Galaxy...

Real life is strangely mirroring fiction, we need to get out and visit the local planning department in alpha centuri before it gets too late!

Yes, I think you ought to just to be safe. When do you plan to go?
Nerd42

> "To go through Wikipedia and count all the inaccuracies is impossible"

You could do some sort of sampling system, and determine its accuracy in a statistical sense.

The problem with that is that you'd have to freeze the servers to do the analysis, and the moment you unfreeze them the analysis is no longer valid.

SoRB

Totally - it fluxuates and changes constantly. Wikipedia may be a certain percentage inaccurate one week, and more accurate the next, (and that is assuming you could write an AI that can check accuracy - that would probably require sentient AI or whatever) and less the next. Of course you could get a moving-average type of number, and then take the average moving average since Wikipedia's conception. It would be somewhat like a stock analysis in that respect ... although an impossible one.

"Wiki" means "to quickly change" doesn't it?
Nerd42

Yeah and the sentient AI would need to know what the the real facts are. So, for example; if an article on cheese contained an error about how it was made the AI would have to know the correct way to make cheese to be able to pick up that error....

Ultimately the AI would have to know EVERY fact about everything in the whole world to stand a chance of picking up the occasional error. That would be one hell of a computer AI system and therefore it would be much more superior programme to wiki in the first place!

God could do it.

But I don't think He's in the habit of helping with encyclopedias.
Nerd42

Nah, you could take a snapshot. I'm pretty sure there are regular backups of Wikipedia, so you'd run the survey on one of those. Obviously, given the shortage of affordable sentient AIs, you'd use subject matter experts to judge the accuracy of your randomly selected sample of wikipedia entries.

yeah only you'd have to keep it a secret as to exactly when you take the snapshots, otherwise vandals would target at those times
Nerd42

Yep. Standard rules apply about not changing the object that you're measuring.

well ... ... any studies done on snapshots wouldn't define how accurate "Wikipedia" itself is ... just how accurate it *was* on average during a given period. so it's kind of like a newspaper in that reguard. Whatever standards we apply to newspapers could perhaps be applied to Wikipedia.
Nerd42