With all seriousness and respect, I suggest good marketing, good people, good technology in that order. -- SunirShah
Here is a short brainstorming list. Let's see how we can expand it and last not least categorize it.
What makes a Wiki attractive for collaboration?
The good infrastructure is not necessarily represented by the built-in features of the underlying WikiEngine. All you need is a good interworking with other systems. Examples:
Dave, take away graphics, sound and programmability, then we are reduced to stone-age hypertext. :-). If a modern Wiki cannot integrate in itself, what it discusses, then it may lose credibility. -- FridemarPache
One of the problems with using anything more complex than text is the ability of the community to support it. Not many people are capable of drawing (e.g. myself as evidenced by the MeatballLogo) or making good scores. Text is pushing it because many people can't read or write very well, but more people are willing to lie to themselves about this than the others! Anyone can write a novel, right?
But, for the right set of people, multimedia can be really fun and powerful. Indeed, one day, if I ever get to coding, MangoOrange? will be multimedia enhanced. Hence my initial research into LicensingMusic. -- SunirShah
I distrust programmability. All too often it is a way of punting on the hard problem, which is deciding what to program. When someone says, "With this Wiki you can have any content you like!", they are really saying they have no idea what the content should be. -- DaveHarris
Well, that's the ticket. If you want content that requires more media than plaintext, like PhotoWiki?, you'll have to put more in. Otherwise, it's overkill. I too believe it is necessary to place limits in order to get a community going. Total anarchy does not scale and is, frankly, boring. Just like suburbanites figuring out what to do on a Saturday... too much to do, so little interest in doing it. -- SunirShah
I would add "good content" to the list. In a sense, "good people" is the derivative of that - content is what you see in a snapshot and people show up as changes in content.
I would add "good theme". Ward's Wiki works because it had a strong theme of Patterns, and later XP, which people wanted to discuss and learn about. I think a lot of people produce Wikis which die because they don't really have anything to talk about on them. See BarnRaising - there needs to be unifying purpose, a barn to raise.
I am not sure what "good people" means. I hope that in any sufficiently large community "good people" will emerge. A person who is "good" in one community may not be good in another. I think technology and social mores can contribute to making people be "good".
Although I agree technology is the least important item on the list, it's something I am personally interested in so I tend to pay it more attention than it deserves. -- DaveHarris
Pretty much. It's not really important to have all those features on this particular Wiki. It would hardly be good to run experiments on your lab notebook. But, it would be a very good idea to implement the Wiki you want based on discussion here... -- SunirShah
Well, if there's a pressing need, sure. Note that Cliff has mumbled something about adding some tags. Full scripting might be useful in another context (I doubt here, but we'll let necessity tell us what to do). I've been mulling this over, especially re: Smalltalk. Not because I'm a Smalltalk bigot, but because Smalltalk isn't a language' per se, but an fully dynamic environment.'' Perfect for this kind of thing, especially if you can think of each page as a class. Kind of like Zope. -- SunirShah
So far, the experiment of using the ContributorLine? has worked well. Pages start in document mode and pretty much stay in document mode. Participation is recognised without ego intruding. Now, if only we could infect WardsWiki with this meme :-)