From Dynamic Knowledge Repository on Wikipedia:

The Dynamic Knowledge Repository (DKR) is a concept currently being developed by Douglas C. Engelbart as a primary stategetic focus for allowing humans to address complex problems. Douglas has proposed that a DKR will enable us to develop a collective IQ greater than any individual's IQ. References and discussion of Engelbart's DKR concept are available at the Bootstrap Institute.

Someone has a problem, and through some process they manage to solve it. The next time they or someone else has that problem, do they need to go through the same problem-solving process again? Are we humans resolved to have to repeat history, over and over and over again?

This is a problem we have on this wiki (and on many wikis that also maintain community mailing lists): even if someone does manage to solve the problem *and* post it to the mailing list the solution is usually lost to the sands of time unless someone is able to wade through the mailing list archives to find it. That's difficult, sometimes practically impossible.

Doug Engelbart talks about how an air force contractor built a fighter jet for the US military, and after a few years all the planes were showing rather extreme cracks around the tail section. When the air force went back to the contractor the company didn't have any records that helped the matter, just all the standard plans and documentation, but nothing about the hundreds of thousands of actual decisions that went into the design. Most of the engineers that had been involved in the development of the plane had either moved on to other projects or even left the company. In any case, it seemed impossible to know *why* the tail was designed with this flaw given all the complicated decision and design process that goes into making a fighter jet.

But then they asked around and found one older guy who remembered the decision making process, partly because he was still angry about having lost a battle over the tail design. He remembered arguing about the design not being able to handle the stresses, according to his calculations, but other engineers had overruled him. He remembered in large part because of his anger; otherwise he'd likely have forgotten. All the small (at the time) decisions were certainly forgotten.

Doug's point is that had the design team kept an efficient set of documentation of not only the design but also all the deliberations, arguments, notes, all the knowledge of the group that could be reasonably compiled during the design process, it would comprise what he calls a Dynamic Knowledge Repository (DKR), then in the future people could refer to the DKR to see what the hell was in the heads of the designers when things fail.

A wiki is a pretty effective DKR if the community uses it to keep track of everything. I know Janne had to make the same decision as many other wiki designers in providing a mailing list, because part of the life of the JSPWiki community is spent here and the contents of this mailing list seldom make it back onto the wiki. If you look at some wikis, say CommunityWiki, they try to conduct conversations right on the wiki, which means the wiki has a *ton* of superfluous content. But it also has people who are constantly trying to refactor those conversations back into the page content. It's a lot of work.

I wasn't able to make it to WikiSym 2006 but I understand that Doug was there and presented on his DKR ideas. I'm sure it's very satisfying for him to see the realization of this concept within the wiki community, a nice rounding up of many decades of advocacy on his part.

The solution: when you find it, in addition to posting it to the mailing list, give everyone else a gift by documenting your found solution on the wiki so the next person doesn't have to bang their head against a wall. If it's valuable, make it part of our collective DKR.

-- MurrayAltheim

Add new attachment

Only authorized users are allowed to upload new attachments.
« This page (revision-2) was last changed on 26-Jan-2007 14:08 by MurrayAltheim