Misc. thoughts about the 2007 Compendium Institute workshop at NASA Ames

Some overdue thoughts on our recent Bay Area workshop, held at the NASA Ames Conference Center May 2-3, 2007. Work has been unrelentingly busy since the workshop, so no time to reflect until now. Today is Memorial Day, so a hiatus.

The workshop was small, about 16-17 people, but very engaged. Most people came from the Bay Area though there was a KMi contingent of five people, including me, that came from the UK and NY. We did something different this time than our previous workshops. In the past, we've generally had our standard tutorial on the morning of the first day. (We've been offering versions of that tutorial since the mid-1990s when Maarten and I developed the first version. It's a subject for another day, but even though I think there is a lot of good material in the tutorial, it has never landed with the impact that I'd like. It is too much "in the head", I think; concepts and exercises that, seemingly, assume you have already internalized the "Compendium Way" as we have (drank the Kool-Aid)).

This time we spent almost the entire first day on an experiential exercise rather than a tutorial. This was excellent -- at least for me since it ties into my research. I had prepared some advance materials that people could download with images about space (in honor of NASA, our host) and sample formats (the idea of providing people a set of evocative images to work with in the construction of a dialogic exercise owes much to the Center for Creative Leadership's Visual Explorer tool).

At the workshop we divided people into small groups and gave them the assignment of preparing an exercise that they would lead the larger group in, with the sole restriction that it had to involve getting people to engage with, and add to, Compendium maps. Each group had about 90 minutes to devise their exercise. Since I was going to use this for part of my research data, I kept myself out of it except to help with logistics when necessary. Mostly I flitted around the cavernous ballroom to each of the small group locations, taking pictures and making sure that each group had their Camtasia screen/audio recording going.

It was interesting to see the different group styles that emerged. Some groups jumped right in creating their exercise; one spent 2/3 of the time on a meta-exercise, mapping how they were going to approach the problem. But all were very engaged. To me the main point of Compendium is the engagement with the representation, having maps be the focus of a collaborative effort to express something that matters. Our previous tutorials never achieved that, they were much more individually oriented. This exercise got people doing that from the start. (Since my research focuses on the choices practitioners make in the course of constructing participatory hypermedia representations, this is very encouraging, but that is again a subject for a different day). Each group worked hard to come up with something they thought would work well for the large group exercise.

After this prep period, we took a break and then began the large group sessions. Each small group set the stage for what they wanted the large group to do, usually with one person acting as the facilitator up in front of the room and another serving as mapper (hands on the keyboard). The idea was that each group, particularly the less experienced members (the "masters" -- Simon, Maarten, Jeff, Eugene -- were explicitly instructed not to act as facilitators or mappers for the large group sessions), would get the experience of trying to get the large group to focus on and add to the map, then get feedback on how well they accomplished this and what they could have done better.

Each of the four groups struggled (to varying degrees) to get the large group to focus on the maps and keep the session from being just a normal discussion. In each case the subject matter and people's desire to talk about it directly to each other without focusing on the map itself seemed to hold the upper hand, though some of the small groups were successful at returning the group's attention to the maps themselves. After each of these 15-minute sessions, we had an additional 15 minutes where the "masters" and the other attendees gave feedback to the small group that had just presented. The nature of this feedback ranged from facilitation styles, to Compendium software tips and tricks (quite a lot of learning and sharing of 'best practices' in those bits), to meta-questions on the purpose of Compendium.

OK, I have to digress from the above narrative to discuss the general phenomenon of getting people to engage with maps. We had much discussion at the workshop on the theme of "discussion capture" (someone mapping out the flow of a discussion as it happens) vs. more directly engaged forms of group Compendium use involving collaborative construction of an artifact of some kind. Having done both myself hundreds of times with groups over the years, my bias is toward the latter. What Compendium is especially "good for" is providing the means (through our host of tools, techniques, and concepts) by which a group can engage with a visual map of something they're concerned with -- and more to the point, that the map is not a single artifact but an interlinked set of artifacts, with ideas and objects related to each other in multiple dimensions -- and even more to the point, that the whole corpus is set up in such a way as to be available to other tools and applications and settings, so that what you collectively or individually do in one form is not limited to that form but can be exchanged with, and used in, other forms in all sorts of ways.

Even yet more to the point, for me, is the idea that non-technical people (that is, non-programmers) can do computationally powerful things just by understanding and manipulating the visual and textual relationships (nodes, links, tags, transclusions), without having to know what is going on with the code and database. These principles have underlay (underlied?) everything we have done architecturally with Compendium over the years. And yet, we still often find ourselves talking about Compendium in the same ways as tools that don't do any of these things (e.g. note-taking in Word, or mind mapping software -- nothing against those sort of tools but they are not about the above).

The problems with this set of points are many. The "discussion" trap is one of them. We -- including me -- like to talk to each other directly. We saw that in the large group sessions -- once the participants got warmed up talking about some issue about the space program, any thought of Compendium flew out the window and people got into highly engaged conversations with each other about the topic. The facilitators had to interrupt this flow to get people to look at their ignored maps. But, here is the thing. My feeling is that this is not a fault of Compendium or of the idea of getting people to engage with each other and their subject matter via the vehicle of hypertext maps. Rather it is our relative inexperience with this medium and mode of communicating. We have seen, many times, that it is possible to have deep and engaged "conversations" while in the midst of working together on the Compendium artifact. In some cases people say that they are able to get to a deeper level of mutual understanding and dialogue than they would have done talking normally.

But to me that (whether talking normally is better or not) is not the point. Rather it is how to construct the hypermedia "event" in such a way as to be inviting, engaging, purposeful, and productive -- so that it becomes a form of collaborative authoring that doesn't feel foreign, "technical", or abstract. It becomes a way for people to make something together, the same way that making a film or video (or collage or whatever) requires people to engage in the medium itself as part of their overall interaction -- it is not an either/or. That is what I strive for -- creating tools and artifacts that invite and extend human interaction, not replace it. And I think we are only at the beginning of that. It *can* be done -- we have seen and experienced it many times -- but it is often difficult to do well.

We spoke many times at the workshop about how discussions, transcriptions, notes etc. can be done conventionally, and using Compendium for any of them is often compared unfavorably to these conventional means. To me this misses the point and is a mark of the medium's immaturity. People don't say, why have a concert rather than a discussion; why make a film rather than have a meeting; why engage an architect rather than just start building a house. We engage in those highly developed media forms, that require expertise and artistry to pull off, because they are accepted ways of having a valued experience or producing a particular kind of outcome or artifact.

Constructing Compendium maps together has that same potential. We have to learn how to do it right. I think events like the first day of the workshop are steps on that path. I have seen before that it is not at all difficult to get groups to tightly engage with artifact creation if they have the right structure and scaffolding, as the small groups did. For me personally, when I finish with my current trajectory of research -- I am on the downhill slope of completing a PhD -- that is what I want to concentrate on; creating engaging Compendium events that get people working together on constructing sophisticated hypermedia artifacts that both express something of importance to the group, and bring the participants and authors together in unique and rewarding ways. The focus should no more be on Compendium itself and whether it is better or worse than other tools, than the focus of a film-making effort is on the brand of camera used. What's important is what is expressed in the unique medium, not the medium itself (see this related post).

Along with generating loads of useful research data for me, the first day exercises seemed to bring the workshop's attendees together. Many of the folks hadn't met each other before but by the end of the first day they seemed to be very friendly and informal with each other (see this photo of our happy first night group dinner at the Grand Indian Buffet restaurant in Sunnyvale). Most of the second day (as usual for our workshops) was more conventional talks and presentations on various Compendium topics, but my feeling (shared by at least some of the other organizers) was that the participants were much more engaged and interactive with the presentations than we have often seen in previous workshops, because they had a basis of shared experience (and reflection on that experience) to draw on.

It's a beautiful spring day here in near-upstate New York, so I will stop here for now to get out on the bike while I have the chance. I'll write more about the second day soon. Before closing, though, I have to give special thanks to Maarten Sierhius and Eugene Eric Kim for all the work they did to get the workshop organized. Along with all the advance deliberations (conducted with unremitting harmony and consensus (not)), Maarten got us our terrific setting at the NASA Conference Center and dealt with all the food and drink logistics (an unrewarding but highly appreciated task), and Eugene managed the flow of events during the workshop itself with aplomb, even improvising the schedule on the second half of the second day. We haven't yet planned our next workshop (East Coast USA? UK?) but I think we learned much from this one that will enrich all our subsequent gatherings.