What will be the warning signs of the coming collapse? Options

[Unrepentant Vandal]

A comment I made recently made me realise that Wikipedia will collapse when wikipedians realise their folly and begin to leave it in droves.

What will be the warning signs? How will we recognise that the end is nigh? I've been thinking and I believe that it will be worth documenting and carefully analysing when it does happen, as the dying days and collapse will be of considerable value to several fields.

This post has been edited by Unrepentant Vandal:

[Wolfe]

One problem with articles that lack consensus is the idea that, if everyone were an expert, consensus would follow. Unfortunately, that is not quite true. In part, this is due to the design of academia, where you get rewarded for publishing new and contrarian material, rather than backing up a consensus. Of course, there aren't the wild conspiracies like the Moon landing hoax, but there is plenty of dispute, and much of it venomous, over many of the same articles that suffer on Wikipedia.

Most problems with Wikipedia could be solved by tweaking its structure. The project really needs professional adminstrators who are prevented from making substantive edits, and whose job is merely to arbitrate and discipline. None of these people ought to be involved in making policy: they just enforce it. That's a pretty obvious separation of powers that most countries follow.

What's killing Wikipedia is collusion. It's well known that the wisdom of crowds works best when each person approaches the subject as an individual. Once people start forming organized groups, you lose the benefit of decentralized authorship because there is cross pollution of information and groupthink. The ideal Wikipedia editor is someone who doesn't edit all the time, and who doesn't have the time or inclination to get involved in Wikipolitics. Editors should live in a private bubble, so to say.

I guess what I'm saying is that the people that are now thought to be the most valuable contributors to the encyclopedia – those who make thousands of substantive edits – are actually the ones who are least valuable to the project. The encyclopedia really needs to be rid of them. They are the ones who get involved in politics, cabals and groupthink, and thus they are the enemies of the wisdom of crowds.

If the goal of Wikipedia is to produce a great encyclopedia for readers, then a principle that has to be strictly adhered to is that editors have no rights. I mean that, no rights at all. The ban hammer needs to swing much more frequently than it does. Anyone suspected of collusion should be banned, and anyone who causes too much strife or edits too much should be banned as well. That is what is required to rescue the project: it needs to be made clear that it is not a community or an online role playing game, but a project that requires a little from a lot of people, rather than a lot from a few.

The truth is that every individual editor is dispensable. There are always more fish in the sea. If you are spending hours every day editing Wikipedia, then the chances are that you are, unwittingly, damaging the project more than you are helping it.

Larry Sanger is missing the point with Citizendium. He's trying to make people take more of a stake in the project by getting rid of anonymity, when he should be aiming to try to get every individual to take less of a personal stake in the project. Once people have a stake in the project. competition and hierarchy naturally follow, and the scum often float to the top. (IMG:smilys0b23ax56/default/wink.gif)

[Firsfron of Ronchester]

Thanks for your interesting and thought-provoking analysis, Wolfe.

Most problems with Wikipedia could be solved by tweaking its structure. The project really needs professional adminstrators who are prevented from making substantive edits, and whose job is merely to arbitrate and discipline. None of these people ought to be involved in making policy: they just enforce it. That's a pretty obvious separation of powers that most countries follow.

There clearly isn't a seperation of powers on Wikipedia. That alone has caused a lot of the abuse on WP. But how would preventing a class of adminstrators from making substantive edits prevent abuse of priveledges? I am not poo-pooing your idea (it's the first time I've heard this proposal), I'm just asking for further clarification.

What's killing Wikipedia is collusion. It's well known that the wisdom of crowds works best when each person approaches the subject as an individual. Once people start forming organized groups, you lose the benefit of decentralized authorship because there is cross pollution of information and groupthink. The ideal Wikipedia editor is someone who doesn't edit all the time, and who doesn't have the time or inclination to get involved in Wikipolitics. Editors should live in a private bubble, so to say.

This is sort of opposite to the way I've been editing Wikipedia; I've worked individually, but I've usually asked for back-up or advice on how to write many articles, either because I lack the expertise, or because I start thinking that my writing may be unclear, and want a second opinion. The idea that I was colluding, killing Wikipedia, hadn't occurred to me. I see your point about avoiding groupthink. But wouldn't one way to avoid groupthink be to ask for outside opinion, outside the organized group? "Peer review" on Wikipedia is mostly a joke, with untrained individuals making recommendations on how to "improve" an article. That's not an academic's definition of peer review at all. But Wikipedia's "peer review" system, I think, is good for one thing (and one thing only): improving clarity. Someone who hasn't read the article before is much more likely to spot areas where the article needs to be clarified.

I guess what I'm saying is that the people that are now thought to be the most valuable contributors to the encyclopedia – those who make thousands of substantive edits – are actually the ones who are least valuable to the project. The encyclopedia really needs to be rid of them. They are the ones who get involved in politics, cabals and groupthink, and thus they are the enemies of the wisdom of crowds.

I don't agree with this point. I think Wikipedia needs substantive edits, if only to clean up the mess that a collaborative editing system creates: articles which look like they were written by someone with Multiple Personality Disorder. Articles which are written in various styles, formats, and spelling conventions. Articles which go much too far in detail or are written too technically. There are a million of these articles, and they need substantial adjustments. Minor edits won't fix them.

If the goal of Wikipedia is to produce a great encyclopedia for readers, then a principle that has to be strictly adhered to is that editors have no rights. I mean that, no rights at all. The ban hammer needs to swing much more frequently than it does.

I'm really surprised this part of your post hasn't received much attention yet. You really believe people should be banned more frequently?


(edited for spelling)

This post has been edited by Firsfron of Ronchester:

[Wolfe]

Thanks for your interesting and thought-provoking analysis, Wolfe.

YW. I'm glad you found some value in it.

This is sort of opposite to the way I've been editing Wikipedia; I've worked individually, but I've usually asked for back-up or advice on how to write many articles, either because I lack the expertise, or because I start thinking that my writing may be unclear, and want a second opinion.

Asking for advice on how to write an article is fine. Wikipedia should really employ professional writers to give advice to editors on subjects like this. It's colluding on content that is counter to the principle of the wisdom of crowds.

The idea that I was colluding, killing Wikipedia, hadn't occurred to me. I see your point about avoiding groupthink. But wouldn't one way to avoid groupthink be to ask for outside opinion, outside the organized group? "Peer review" on Wikipedia is mostly a joke, with untrained individuals making recommendations on how to "improve" an article. That's not an academic's definition of peer review at all. But Wikipedia's "peer review" system, I think, is good for one thing (and one thing only): improving clarity. Someone who hasn't read the article before is much more likely to spot areas where the article needs to be clarified.

As I understand it, the wisdom of crowds dispenses with the need for peer review. The whole point of it is that everyone is a peer. WOC won't work if that isn't the case. The problem occurs when people edit too much and seek others to help them support their edits (which is natural).

The proper attitude one should take to Wikipedia is this: "I've seen something I think is wrong, so I am going to change it. But I am not going to come back every day to revert it if someone else changes it. Rather I am simply going to leave it. If, over time, it survives the scrutiny of thousands of individuals, then it was a good edit. If it doesn't, then that's just too bad. What I can't do is keep insisting I am right and attempting to "own" that part of the article".

I don't agree with this point. I think Wikipedia needs substantive edits, if only to clean up the mess that a collaborative editing system creates: articles which look like they were written by someone with Multiple Personality Disorder. Articles which are written in various styles, formats, and spelling conventions. Articles which go much too far in detail or are written too technically. There are a million of these articles, and they need substantial adjustments. Minor edits won't fix them.

And Wikipedia should employ full time writers to clean up articles. I don't believe that there is such a thing as too much detail. What that usually means is that the article should be spun off into a number of sub articles.

It's not like Wikipedia couldn't afford this. It's one of the most visited sites on the internet, and wisely placed Google ads would enable it to fund a large number of professional administrators and clean up crew. These people would not be contributing content, but just helping to organize the contributions of the amateur editors more efficiently.

I'm really surprised this part of your post hasn't received much attention yet. You really believe people should be banned more frequently?

Yes. Every individual editor is dispensable.

As I understand it, the WOC works like this: you have a jar with a certain number of jellybeans in it, and you then get everyone in a room to make one individual guess as to how many jellybeans are in the jar. Contrary to expectation, it turns out that the average guess is startlingly accurate.

What messes this up is a situation where people are allowed multiple guesses and are allowed to start discussing the issue amongst themselves. In the first case, a very few people are likely to care a lot more than others and keep making more and more guesses, while everyone else gives up, and in the second case collusion will produce distortion via groupthink.

Wikipedia would operate in an optimal fashion if everyone edited occasionally, but no-one individual or group of individuals edited excessively. That is, of course, an ideal situation. You can't force all readers to edit, and you can't prevent individuals from editing more than others. What you can do is prevent excessive editing of individual topics by simply banning people who do that for a month or so. The worst POV articles on Wikipedia content wise are not caused by thousands of editors making small edits to them, but by a small number of editors making thousands of edits to them in the context of POV wars. That's not the wisdom of crowds, but the folly of a few.