Day 3 of the WikiSym 2007 started with an Keynote for both, OOPSLA and WikiSym attendees. It was given by Peter Turchi who is actually a writer, not a computer person. His talk was about generally about being open minded to be inventive. One sentence I remembered was “To be lost means to want to be somewhere”, so he sees being lost as the first step of being inventive or as he put it: “To refuse to be lost is a failure of imagination”.
After the keynote I went to some more OpenSpace sessions. Evan Prodromou showed a really neat way to find the real contributors of a page, leaving out people who vandalized or just did administrative work. What he basically did was to create a graph from all edits and then removing all loops, just keeping the edits which lead towards the current page revision. Adding something like that to DokuWiki should be relatively simple by extending the page changelog by a content checksum. This might be useful for other things as well so I might add that soon.
Another session was about the usability of wiki user interfaces and how guidelines might help wiki developers. One thing I was particular interested in was the fact of some engines having an “Add a page” button. In my view that made no sense, because it encourages to create Orphans. Louis-Philippe Huberdeau pointed out that many Wikis now feature many other ways to find pages than just links. We agreed that such a function might make sense or even may be desirable in certain cases.
In another session we talked about collaboration in education and how it can be assessed. Some of the present college professors pointed out that collaboration isn't really rewarded in the educational/science/research system. One example was about how you are only allowed to put a book into your list of publications in Italy, when you can clearly say which part of were written by you. On a truly collaborative work this simply isn't feasible. This leads to “stapler collaboration” where each participant writes her part and then all parts are just clipped together.
One thing mentioned about online collaboration was that collaboration in wikis was improved when the group had an additional channel to talk about the work. Instant Messaging and Telephone was mentioned. This reminded me on having a look at the available AJAX chat plugin for DokuWiki.
The Day closed with a talk given by Ward Cunningham about a system he created for the Eclipse foundation. It took me a a while to understand what this was really about. Based on an automated testing system he basically integrated an way to make the business logic behind the foundation's website more transparent to its users. Users can actually see what goes on behind the scenes when they fill out some form. The whole system didn't seem particular useful as a test suite, especially not for the website users, but made a lot of sense to visualize the bureaucratic steps. Ward gave the example of Wikipedia and the Wikimedia Foundation which aims to work transparently as well but so far didn't succeed to much in this.
I probably will give an additional summary of WikiSym in a few days, when I had time to order my thoughts a little bit.