Insofar as it's been a loose series, this post is a continuation of my thoughts on wikis and hypertext. My leading question is "Are wiki's ever completed?" And "If so, how do we know and decide?"

This is a question that I find myself wondering about a fair deal, and I think the answer--which I haven't come to a firm conclusion on--has a to do with the potentials of the wiki medium.

I should jump in and say that, while wikipedia is a great reference, a great tool, and an important project, because it's the example of "what a wiki is" it has shaped how we think of the medium in a way that I'm not sure is particularly useful. The biggest wikia projects are encyclopedic studies of Star Wars and Star Trek, and while their material isn't quite suitable for wikipedia it is certainly in the same vein and tone.

The encyclopedia form has been revitalized by the wiki, by decentralizing the review process, democratizing (more or less) the focus and the creation of articles, but most importantly by removing the "space limitation" on content. Nevertheless I continue to be convinced that wikis as a forum are capable of so much more.

On the one hand, big projects, like the kind that might be recorded in a wiki, are never really completed as much as they are eventually abandoned. That sounds pessimistic, but I think it's ultimately productive: eventually a project has done what it needs to do, and what with perfection being unattainable, the productive thing to do is move on. The decision of when to do that is perhaps one of the most important decisions that a creator/artist can make about a work.

But who makes that kind of decision about a wiki? Is there a point where people just abandon a wiki? While wiki's are collaborative, that's not to say that they don't have leadership (wikipeida's leadership organization is epic, for example,) but who makes these kinds of decisions?

While I'm prone taking an entire wiki as a single document, the fact that a wiki is really a network of tightly connected texts surely has baring on the answer to the question.

Software projects use the concept of "stable releases" and a "release cycle" to ensure that a project can both continue to develop, and exist as finished cycles. The debian project has it's own procedure for encouraging ongoing development of their system/packages and creating rock solid stable systems.

Additionally, while most wiki's have semi sophisticated version control systems, they for the most part don't have a concept of "branches," which might be helpful for implementing a stable wiki/wiki branch system. Even ikiwiki, which can use systems like git to store history, doesn't have a good display system for switching between branches/revisions.

I'm not sure--of course--if there are really good answers to these questions. While I haven't begun to post any of them, there are a number of projects that I've been playing with in my mind (and locally on my own computer) that are wikis, but I've been hesitant to let them go into the wild in part because of issues like the one discussed above. And above all, if the wiki format is going to grow away from and independently of the encyclopedia format, I think we need to begin discussing questions like that.

So there. Onward and Upward!