April 2003

There was nowhere to hide.

I had had a terrible, demoralizing week at my translation job, compounded by a major crash on the computer system where I was writing my dissertation, and by the time I got to Caroline's software meeting in Copenhagen I just wanted to withdraw into a corner and vanish forever.  It was clear that this would not be possible since the other early arrivals were rearranging the chairs in the brick-walled basement canteen into a big circle. It looked like they were expecting a twelve-step group.  As the official start time came and went, the chairs gradually filled with about 60 men, many of whom quickly established barricades of unfolded laptops on their knees, and a few women, one of whom, thank god, was Caroline.

We were finally called to order by the ringing of a pair of miniature Tibetan cymbals suspended from the fingers of the local host, Lars.  Lars was the CEO of a Danish startup called Collaboraid on whose business-incubator premises we were meeting.  There were no introductions, since most people had been there for another meeting the day before.  That meeting was about dotLRN, a platform for computer-supported education.  This one would be about OpenACS, the website/database software underlying dotLRN and analogous products for commercial and nonprofit use.  Lars introduced his friend Alex, a professional meeting facilitator skilled in the use of professional speaker hand gestures.  There was a structural pillar in the middle of the circle and Alex circled it while he spoke, as if avoiding a Peircean squirrel.  

"How many people have participated in an Open Space meeting before and know about Open Space?"  

A few cooperative hands went up.  

"Well, Open Space meetings evolved when a really smart guy noticed that what people really liked about conferences was the coffee breaks."  Inclusive hand gesture.  "Why do you think that was?"  

A few cooperative responses, groping toward "because they could organize themselves."  

"That's right!" said Alex.  Power hand gesture.  "And so we figured out a way to have a meeting that was all coffee break.  There are just a few principles, which are on the poster behind you:"

4 PRINCIPLES

1. WHOEVER COMES ARE THE RIGHT PEOPLE

2. WHATEVER HAPPENS IS THE ONLY THING THAT COULD HAVE

3. WHENEVER IT STARTS IS THE RIGHT TIME

4. WHEN IT'S OVER IT'S OVER

1 LAW

YOU HAVE TWO FEET   

What the one law meant was that if you weren't getting anything out of the group you were in, you should go to another one, with the caveats that the all conveners needed to be present in their groups at the beginning to introduce the topics, and anyone taking minutes should hand off the minute taking before leaving.

The next half hour was spent nominating group topics and arranging them in a schedule.  This was when the introductions happened: "My name is Bill - " "Hi, Bill!" "-and I want to talk about the repository."  Following some merging, splitting, and trading of timeslots, there were five meetings in each of four hours.

Caroline advised me to go to one that she said would be talking about translation, and I did, and it was interesting and even relatively comprehensible (you can read the minutes).  The main problem that was informally discussed was programmers and amateur translators were translating stuff in the user interface as new components got written, which caused conflict with the translations done for finished components by professional translators.  It also caused conflict with the professional translators themselves.  Nobody wanted conflict.  Conflict was bad. 

From a technical point of view, all this had something to do with "pretty names."  At the coffee break (Open Space meetings have coffee breaks from the coffee breaks) I caught up with Caroline and asked for a clarification.  "In my community there are short names and pretty names," she said automatically.  "Short names are what the system sees and pretty names are what the user sees."

In my next session, on marketing, the word "community" came up all the time.  It was like a real estate presentation.  It was like a Thousand Points of Light convention.  It was like a seminar on Habermas.

"If we're trying to reach out to people who are new to the community, maybe we need to rethink the website. If you look at the websites for MySQL and PostgreSQL, one is much better at selling the product than the other."

"Well, that's because one's a company and the other's a community."

"We need to get more of those magazine articles about OpenACS rewritten and onto the website."

Me: "Don't you think magazine articles have more credibility when they're the original versions located on the magazine's site?"

"Well, the articles are aimed at slightly the wrong audience.  And there's another purpose here, which is to get the guy who writes them to write more stuff, get him more involved."

"There are decisionmakers who never open their computers.  Maybe we need printed materials as well."

"..."

"I mean some people might want more information on what OpenACS actually does."

"Well, that's in the documentation."

Community, said Caroline, was the big idea behind dotLRN: instead of treating university students in terms of classes with tasks, it treated them as communities that sometimes happened to be focused around classes.  "Because if you think back, the most important things you did in college weren't part of an organized class."

"Why didn't you tell me that when we were seventeen?" I said.

Still nobody had asked me for a badge or programming credentials.  At length I caught on: Open source people were committed to being open.  Although they clearly had in-groups and side conversations, they had to put up with anyone who showed up, or they would be inconsistent, and being programmers they prized consistency.

The community also, on first look, seemed honest to a fault.  A couple of times I tried to get people to tell me why their system was the best.  "It's not," they'd say.  "This alternative does this thing better. That alternative does that thing better.  We have a lot of work to do."  

If they seemed painfully uninterested in traditional marketing (like, you know, a brochure that people like me could take back to our departments) which would take them beyond beyond the traditional free software bailiwicks of computer science labs, nonprofits, and other free software companies, they also seemed remarkably uninterested in lying or hiding their bugs.  This was what another guy named Alex, who turned out to be a social sciences researcher as well as a programmer, described to me as a "culture of transparency." 

The OpenACS community seemed to have a statistical oversupply of people named Alex.

The final session I went to was about diversity in the OpenACS community and was convened by the third of the four women present, who was a communications specialist from a nonprofit.  She was concerned about getting more people into the community who were not able-bodied, prosperous white male programmers living in Tech Valleys in the First World.  

In a way, this session exemplified the problem she was talking about, because the other participants (nearly all of them able-bodied, prosperous white male programmers living in Tech Valleys in the First World) kept diverting the discussion from the main point to obscure technical minutiae by whatever means possible.  I don't mean that they did it deliberately or maliciously; rather they did it reflexively as if getting away from a potentially uncomfortable subject.  I am an experienced subject-changer myself and can recognize it when I see it.

"What I want to know is why aren't there more women in the community, and how can we get more women involved."  The convener was pursuing her point for the second or third time.

"We have women," said one of the guys.  "There's you, there's Caroline, there's Tammy who posts on the boards."

"And I'm concerned about how we can get more people from developing nations involved," she tried again.

"There are people from developing nations.  There are people in Argentina, in Guatemala, in India..."

"Yeah, IIT students looking for help with their homework."

"What I'm concerned about," the convener went on, "is that I think there are people out there using the software, different kinds of people in different places, people of color, women, nontechnical people as well, but they're not posting on our forums because they're intimidated.  I, for example, have only posted a few times because I'm not technical and I'm intimidated." 

"Well, you should be," said one of the guys from behind his laptop screen.  But then he reversed himself and said something politic about not slapping people down for cluelessness.

And indeed, compared to old school Usenet groups, the OpenACS community boards do seem to be warm and welcoming.  In fact, compared to some of the social boards I'm on, they're warm and welcoming.  This was the post-Usenet, netiquette-enabled age of the sensitive New Age geek.  Some people in the community were less sensitive than others, but they were clearly a quite different breed from the self-righteous, know-it-all unbathed flamers who set the tone at the AI Lab in my youth - the breed described by Joseph Weizenbaum as "compulsive programmers." Sherry Turkle described them as people disconnected from their bodies, seeking in computers a degree of control they felt they lacked in meatspace.

At this meeting, on the other hand, pretty much everyone appeared well rested, healthy, fashionably dressed in earth tones, and anxious to be open and constructive, not hierarchical and points-scoring [1].  There were still a few bugs in the system, but community members generally  listened to other people, or did a pretty good simulation.  What had happened to the hacker world?  How had it turned into something that Carol Gilligan followers would call feminized and I would call androgynized - particularly since it still had so few women?

"It's a cult of personality," Caroline said, "but the personality is absent: Philip Greenspun."  Philip, the founder of Ars Digita Inc. and well-known "guy with the dog" (named Alex [2]), actually sincerely wanted people to learn to do what he did.  He taught free courses in his signal skill of building database-backed websites.  The Greenspun photo.net site was the prototype of the community idea.  Philip himself is now walking the earth, teaching computer science and blogging.  The graduates of his "boot camps," who are spread around the world, are continuing to develop his systems through the OpenACS, for lack of a better word, community.

Many were not born computer geeks, but career changed into the lifestyle.  Before doing the boot camp, Caroline had been an environmental engineer and a semiconductor engineer, a developer named Paul had been a chemical engineer, and one named Frank was still in medical school.  I asked what the most unusual background was and received the reply that it might be the cheese guy - there was a developer (not present at the meeting) who had previously been the head of the cheese department at Dean & Deluca, and was the only person known to be a member of both the Guild de Fromage and the IEEE.

As the "outro" [3] we were called back to the circle and asked to sum up the day in five or six words each.  (A group hug was also proposed, but not taken seriously.)  The discipline of the circle where noone could hide broken down and there were now double rows in a couple of places.  There was a longish silence before someone in the circle was ready to start.  "Good ideas from smart people."  Suddenly everyone else breathed out - that's what it was supposed to sound like! - and a clear and rapid poetics developed:

 "Categories, repository, and code sharing."

"Community will grow in the future."

"Time to become more involved."

"Lots to think about later."

(Mine was "How to run a Quaker business meeting.") 

The conveners were pleased: "A lot of this seems to be about the community and how happy we are to be involved in it." There was much thanking of Lars, who would the next night give a very good party at his flat, with wireless network access and self-organizing cooking, and it was agreed that there would have to be future meetings, in different places to accommodate people who could not travel to all of them, and not always in First World countries either ... 

"I hear Iraq is open," said a voice behind the meeting leaders.  "The palace is nice."

[1] And in this they also came off well compared to, say, high Chomskyan linguists.

[2] Alex the Samoyed, as in Philip and Alex's guide to Web Publishing, is the mascot at the top of the OpenACS pages.  Other free software communities with animal totems include Linux (the penguin), PostgreSQL (the elephant), Mozilla (the dinosaur), the Free Software Foundation (the gnu).  Non-open source communities with mascots include MySQL (the dolphin) and, in a more diffuse and variegated way, the O'Reilly book empire.  Only a few unclaimed species left!  Reserve yours now!

[3] Actually, between the Open Space meeting and the outro there was a "Project Cafe" where people suggested projects and signed up to work on them.  "Most of these people are proposing very ambitious projects that will take years to achieve," Caroline observed.  "My immediate project is I want to get the categories functions working on the OpenACS pages themselves.  Some of these things are going to move forward but it will be because of individual people and groups working together, not necessarily because of the promises that are made here."  Also, the outro wasn't really the end because the next day there was a "governance meeting," which was again open to all and concerned with keeping the structure "light" and inclusive of anyone who made contributions.

The frog photo was taken at the party at Lars's; Caroline brought the marzipan frogs and Danish flags.  The Lego photo was taken in the children's section of the Louisiana art museum where we went the next day.  The last photo was taken at the party when the dessert of bananas in chocolate sauce and flaming vodka were brought in.  Some tastes in the hacker community never change.