Why Use Mod Wiki, And Wikis In General

A bit of background: Wiki wiki is Hawaiian for "quick." Ward Cunningham invented the first Wiki, a web site built for easy collaborative authorship, which has inspired many similar engines. For those who prefer paper, there is a new Wiki book out: The Wiki Way: Quick Collaboration on the Web, by Bo Leuf and Ward Cunningham.

A WIKI ENGINE I LIKE

It turns out I like Clifford Adams' Use Mod Wiki. It's a pretty faithful Wiki Clone with some very good additions, such as (simple) versioning of each page and straightforward user accounts. Pages are stored as individual text files instead of in a db, which is good and bad, but simple and straightforward, at least. It does Inter Wiki (a start at interlinking Wikis), has other cool tweaks, and something that's very convenient: a single Perl script (with a couple simple, optional configuration files) drives the whole thing.

OH WHAT A TANGLED WEB...

As it happens, I had previously found Wikipedia and thence Nupedia. I liked the Wiki Clone driving Wikipedia, but didn't have time to find out which it was.

Later, to answer the question about what WIKI meant, from Ward Cunningham's Wiki Wiki Web I found a link to Sunir Shah's Meatball Wiki, for which Adams wrought Use Mod Wiki, which also runs Wiki Pedia -- serendipitous circle complete! So I'm glad the subject of Wiki came up.

INTERESTING WIKI STARTING PLACES

Nupedia -- an open content encyclopedia project, with a small amount of funding and a lot of drive. Also supplanted the GNU encyclopedia effort, with a link from GNU to that effect, so it's got a bit of a push there, too.

Wikipedia (parallels Nupedia) -- because a Wiki is is a fun way to build an information space like an encyclopedia collaboratively.

Meatball Wiki -- looks like a good example of a Wiki with a forceful editoral presence, which helps grow a good online community, and because its theme is intercommunity and online communities.

Use Mod Wiki -- to download the Use Mod script itself. It's in Perl, and will run easily under Apache in either *nix or Windows. I've set it up in Linux, and it was very easy.

Wikki Tikki Tavi -- cool features like Use Mod has (or perhaps it's vice versa), but built on My SQL and PHP.

Wiki Wiki Web -- it's the oldest Wiki and has collected a lot of wisdom on what's good and bad about Wikis and online communities and their infrastructure.

SOME OBSERVATIONS ON WIKI COMMUNITY

Wikis in general lend themselves best to collaborative writing that ends up being hard to attribute to one particular author. This makes for robust community shared knowledge, but also brings challenges of attribution and information property rights. Who holds what rights as this knowledge develops? What happens when an individual is destructive, or wants to withdraw his contributions?

Knowledge on a Wiki grows best from having an community-accepted editorial focus, and from continual preening, pruning and rearranging the information to make it better. This reminds me of what Neil Larsen (Max Think) used to call "knowledge annealing" or PARC's Scatter/Gather. On some Wikis, it's come to be called "factoring" or "re-factoring," in analogy to a similar (but different) activity which is a part of Extreme Programming.

LINKS TO MORE INFORMATION ABOUT WIKI COMMUNITY

Starting points for some good reading about these and other community knowledge issues from Wiki Wiki Web, Meatball and Wikki Tikki Tavi:


 * http://tavi.sourceforge.net/WhyWikiWorks [highly recommended!]
 * http://c2.com/cgi/wiki?WikiPrinciples
 * http://c2.com/cgi/wiki?AvoidSignatures
 * http://www.usemod.com/cgi-bin/mb.pl?StyleGuide
 * http://www.usemod.com/cgi-bin/mb.pl?WikiLifeCycle
 * http://www.usemod.com/cgi-bin/mb.pl?GreatChallengesToWikis
 * http://www.usemod.com/cgi-bin/mb.pl?MeatballWikiCopyrightDiscussion
 * http://c2.com/ppr/wiki/WikiPagesAboutRefactoring/WikiPagesAboutRefactoring.html
 * http://www.c2.com/cgi/wiki?WikiWikiClones

CAN'T WE ALL JUST GET ALONG?

Now, my conundrum for the day:

Given that different people like different ways and different places for expressing and collecting knowledge, how do we help that knowledge coalesce across venues? How do wikis and weblogs and things like Slashdot and Everything2 and mailing lists cross-pollinate and synergize instead of duplicating effort? It is perhaps a rhetorical question; I know people are working on things like OHS/DKR, RSS, XTM, KIF, etc. -- there's just a lot more work to be done.

-- Peter Kaminski, 20010507


 * They cross-polinate with, as you say the Meat Ball:Semantic Web. The technical problem is just that all the protocols are incompatible. The social problem is politics. i.e. It's my project and I'm not going to cooperate with you! But that's ok. On the Internet, there is no cost (to a potential user, not a community or host) to create a new site and no cost to switch and no cost to link. I say roll with it. -- Meat Ball:Sunir Shah


 * There's talk about Wiki:Xml Rpc (Meat Ball:Xml Rpc) sometimes, but i don't know of any implementations. The closest sort of thing might be Meat Ball:Open Wiki, which displays Rss Feeds with a simple command. On a related note, a distributed TWiki will likely appear soon: http://twiki.sourceforge.net/cgi-bin/view/Main/TWikiStoryOfInktomi. --Wiki:John Abbe

Hi Pete! I'm trying to see if I can snag changes by gunning my script on your Wiki, in particular the Recent Changes page. It's starting to seem quite impractical, although I've made a good start. I keep a copy of the page, then look for lines in the new version that don't exist in the old version. This still picks up awkward things like timestamps. So I throw away all numbers in making my comparison. But even so, there's inconsistencies across lines in working with http://ms.memes.net (yours seems more regular). I think Lucid Fried Eggs has a "random link" that I pick up. It might be possible to just ignore changes of a certain size. But then I wouldn't pick up changes in your Recent Changes. So it's quite touchy. I suppose ultimately the solution is to do this on a page-by-page basis for the one's that work (your seems to). We can create special solutions for the cases that don't work, all the special cases. Andrius 20020910

My feed is picking up your changes pretty well, see: http://www.ms.lt/zz/output.html I could improve it by converting relative links to absolute links. That's a good example of the kind of trickery needed on the receiving end. I told you how I'm thinking that customization on the receiving end trumps customization on the sending end because, even though it's harder, the necessity is on the receiving end, and the reuse is on the receiving end. So priming data will happen only as a consequence of the desire to have information received, that is, only as an aid to customizing on the receiving end. That bodes for a very intense idiosyncratic coding tricks of all kinds on the receiving end. Andrius

I noted some specific techniques for increasing relations across wiki-type sites at http://webseitz.fluxent.com/wiki/WikiWeb (--Bill Seitz)