Slaving over HTML

I spent much of today and Tuesday slaving over the HTML on the compendiuminstitute.org website, making mostly low-level changes to the top navigation and various content, moving a few areas around (which had ripple effects to the navigation that I only realized afterwards, necessitating many further changes).

It's kind of nice to do work like that. It requires care and attention, and has to be done right. While the "hypertext" is not the deep and beautiful hypertext that Compendium provides, it is still hypertext, and thinking about the compendiuminstitute site's far-from-perfect information architecture is not dissimilar to the kind of structuring work I love to do within a Compendium database.

I also find that doing HTML coding gets me thinking about things Compendium still doesn't do.

We've actually thought about, and I've at times advocated, generating the whole site from Compendium itself, much as the output from the 2005 Workshop was done (available here). It wouldn't look like a conventional website, and maybe that would not be good. On the other hand, thinking about and getting it to work "right" -- that the navigation would get people to the right places and content -- would also force us to think about improvements to the tool itself. It would also be an exercise in knowledge art because it would involve a host of aesthetic/technical choices about how to structure both the user experience and the technical underpinnings.

That work -- thinking about and manipulating the aesthetic and technical structure -- the interweaving of the two levels, thinking about and working with them as one -- is the essence of knowledge art. Immersed in the experience of making, there is no difference, no separation between the levels. That's the same as all art, really -- art always involves using a technology and knowing its facets and aspects. What makes it knowledge art is the direct engagement with a "knowledge representation" -- something showing and leveraging the relationships between ideas.

Compendium progress

Compendium has now surpassed 6000 downloads, there are almost 300 members of the compendiuminstitute yahoogroup, and we are up to version 1.4.1. A tremendous amount of progress.

Maarten just visited here for a couple of days, and we were talking about how great all this was, but also that it still seems like much of the potential has not been tapped. We were doing things with lashed-together tricks and QuestMap in the 1990s that were in some ways beyond what anyone, including either of us, have done with the current Compendium software (at least that we've seen publicly).

Part of this is because it remains difficult to talk about what Compendium is really good at and good for, that make it different from other tools. Things like the ability to interweave formal and informal representations, to span the diverse areas of group process facilitation, visual modeling, software integration (talking to and from other tools). To do things in a moment that are also of use in the future, and to draw on what was done in the past, in direct and connected ways (connected in the ways that Compendium lets you connect). To make artifacts of many different sorts.

To me these are not at all separate or contradictory domains, but it does seem like they are for most people, or at least that none of us have come up with ways to make them live in a coherent way outside a small core group of obsessives. People that are interested in one or two of these areas are not interested in the others. I am going to try to write more carefully about this and show some examples.

Since in a blog I don't have to meet any particular standards (e.g. academic/research writing, or business/marketing, or user support, or any of the ways I have tried to write about Compendium in other places), I will here try to let go somewhat and write about Compendium as I have experienced it, without worrying about how it fits in to or measures up to other perspectives.