Wiki As An Alife Experiment

In an Alife Experiment you create a sort of playground, e.g. the artificial world.

Next you create an array of a data containers. Every data container is a so-called critter (= artificial being) that runs around in the Artificial world.

The critters will interact with the playground (=eat) and with each other (fight and procreate). This all according to all kind of rules. In a sort of way the datacontainers are genes.

This is all very simple.

I wondered what way a WikiWiki page is also an ALife experiment. What is the playfield, what are the critters. I just didn't see the answer, until I realized a page is not a gene, nor a playground, but a meme (e.g.:http://pespmc1.vub.ac.be/MEMES.html)

So a wikiwiki is an experiment/study in evolvement and change in culture.


In a study that I made a few years ago, I noticed a lot of people stay rather short; they discover wiki, next they are very active during let's say, hundred days, producing 20 or up to 1000 actions, next they quit.

All this actions by people is similar to the actions of people in the real world over generations.

Everybody who is very active thinks he or she knows what a Wiki is all about and what way it should look like, but actually we just know it's a whole lot of pages and a way to create and edit pages.


A good way to study what way the pages change is to upload a list of all the wikipages, http://c2.com/cgi/wikiNames, every week and as soon as you have a lot of pages, you can spend some time writing some code to analysing changes in the list.


Moved here from TimeAgainForWikiMutiny (posed there as an argument against mutiny):

An Experiment

In relation to the just above see: NobleExperiment PositiveDialogueCommunity
I'm a bit more optimistic. One can LetHotPagesCool then RefactorMercilessly without risk of stomping on egos. Perhaps allowing a ThreadMode, heated argument to take place then having a neutral (but interested) party factor out signal is to DoTheSimplestThingThatCouldPossiblyWork in a social wiki. IMO, people that jump into a heated argument and complain about the ThreadMess are just irritating hypocrites, participating in the very problem they are denigrating. DocumentMode can't be easily forged out of nothing at all - perhaps it needs heat, argument, hot air, fire and flame, then eventually ash, for a germ of an idea to grow into something more. In the decades-long view, this is how any wiki gets to be InsanelyGreat.

However, there is one thing (ignoring spam and abuse) that needs to be changed: OnceAndOnlyOnce needs to apply to flaming arguments and ideas, too, lest people continue to enter the WikiWiki and present the same arguments again and again.

To do this, people need to be easily able to find both running arguments and DocumentMode pages related to topics of interest. WikiWiki, or any successor to it, desperately requires considerably more advanced and accessible search features in order to break down the WalledGardens. WikiWiki had the right idea, at least, in resisting namespace hierarchies which would do the opposite. Further, to prevent fruitless arguments and repeat arguments, such as those founded in deep misunderstandings (e.g. those falling back so often to LaynesLaw, or that arise from not grokking some theory or model), a wiki must enable a much greater and more thorough degree of self-education than is provided by the very limited and somewhat magical "read the page and though shalt comprehend". A wiki as an educational medium does not need to be entirely passive; DocumentMode should be supplemented with support for interactive education, tutorials, tests and quizzes that are automatically graded, examples that can be run and modified, et cetera - and a wiki would do well to possess social and cultural engineering (e.g. through 'ideals' like WikiPedia's NPOV, and perhaps by making quiz-grades public per user-name, and even by such things as RealNamesPlease) to pressure those humans possessing unmerited arrogance (and there will always be a few in any population, more so among the people willing to speak up) to humble themselves before impartial tests of knowledge, comprehension, and skill. (Too often I have wished to slap around those people who refuse to self-educate, and who therefore cannot see their own fallacy or misunderstanding even when it is pointed out to them.)

As a bonus, construction of such tests and tutorials and examples - even if it is just tests of comprehension rather than of 'truth' - would greatly increase the educational value of any wiki AND help solidify and (most importantly) formalize the ideas held within, considerably reducing imprecision and ambiguity inherent to natural language.

A couple of points in response: I wish I had your optimism. I've been here for over 10 years and see nothing to make me believe the decline will even slow.
What should be of interest to all, after reading this page, is how that this wiki and its effect on its community, stands as a good model of why open communes don't work very well. This wiki is kind of like the old hippy communes of the past, and the only ones that seem to have really survived are the closed religious ones. However, a cyber commune like this one, is ideal for a continuing experiment in social interaction. It is important to give the community a task, the task itself is not that important, but rather what happens when they try to accomplish that task. I have mainly stayed around all of these many, many, years, to observe and interact in this most intriguing experiment. One can learn much here.

Ye'old TheCathedralAndTheBazaar argument, again? I don't think that way at all; this page is just a few people voicing a few complaints, and doesn't at all represent what has and has not "worked very well".


No signed-edit security: yet ANOTHER reason to mutiny :-)

Yes, I note the smiley, but nonetheless, people here used to take care over ethics, even though not enforced by the technology. Clearly now they don't bother.


"but it's clearly too late" "... and the band played on"
 while the Titanic sank, so it clearly wasn't too late ...


Last Song -- DonaldNoyes.201002180545
CategoryWiki

EditText of this page (last edited March 20, 2011) or FindPage with title or text search