The Forces Be With You

Last week I got back from two consecutive conferences that could hardly have been less similar. The first was Supernova, in Silicon Valley. The second was ISACA, in Boston. Where Supernova's main subject is "decentralization", ISACA's is a cluster of subjects (starting with audit and control) that all be scream "centralization". Reconciling the two is a tough challenge, but it's one we need to tackle.

First, Supernova.

There is a new conference constituency that bridges a variety of parochial subjects: techies who write. Some (like me) write for print and online journals. The rest (also like me) write weblogs, or contribute to other products in the same broad family, such as mailing lists and wikis. These are not "attendees" looking to be educated by speakers, panelists and leaders of tutorial sessions. They're participants. They come to engage in meet/meat space with peers who not only care about the same subjects, but are motivated to move and shake their worlds.

This population is not small. Two days ago, Technorati* (which, not just coincidentally, runs Apache and MySQL on Linux) passed the three million point in the number of weblogs it watches. There's nothing epiphenomenal about blogs any more. They are becoming critical parts of countless market ecosystems.

Supporting the writing needs of participants has become as important for many conferences as the lineup of speakers and subjects. Broadband Wi-Fi is a necessity, not a luxury. All the participant-oriented conferences I attended in the last year -- O'Reilly's eTech and OSCon, Demo, PC Forum, Digital ID World, Apachecon -- all of them featured free Wi-Fi for in the meeting rooms and common areas.

Supernova was no exception in that respect, or in one other: the frequent failure of the system. Wi-Fi was mostly down on Day 1 and mostly up on Day 2. While there do exist a few conferences where downtime is minimal (PC Forum comes to mind), all tend to suffer. As Esther Dyson (who runs PC Forum) told me while we were commiserating about Wi-Fi problems at Supernova, "It's hard." You can't just hang a Linksys access point on a hub off the hotel's T1 line. You need a proven and robust system, dedicated techies to mind it, and a working relationship with techies at the venue as well. Generally speaking, hotels are not in the broadband delivery business. They often sub out that stuff. Big convention centers often have unions as well. I've heard unions blamed for failures at the Javits in New York and Moscone in San Francisco; but I don't know what the real problems were. As a user (and substandard techie) about all I could do was complain.

Still, as an event, Supernova delivered. Tim Bray, the XML guru who recently became Sun Microsystems' Director of Web Technologies, says there are three kinds of conference models: trade show, academic and power conclave. About Supernova, he reports,

Supernova is definitely a Power Conclave. Fortunately, Kevin Werbach knows a lot of people who are very interesting indeed. So it’s pretty good, but there are a couple of problems: the crowds are big enough to make in-session interactivity a bit awkward, and, to be brutally crass, the presenters are generally not that hungry, and a few of them have coasted a bit.

On the other hand, the hallway schmoozing is really superior.

I tend to be forgiving of sessions that don't excite, provided there are others that blow my mind. Two in particular did that at Supernova. And oddly, there seemed to be an unintended relationship between them.

The first was a session titled The Network is People (like Soylent Green!), in which Christopher Allen, a veteran techie/entrepreneur, spoke insightfully about the "Dunbar Number". In his blog, Chris quotes from the source:

... there is a cognitive limit to the number of individuals with whom any one person can maintain stable relationships, that this limit is a direct function of relative neocortex size, and that this in turn limits group size ... the limit imposed by neocortical processing capacity is simply on the number of individuals with whom a stable inter-personal relationship can be maintained.

The number is roughly one hundred fifty. It's referenced in Malcolm Gladwell's The Tipping Point and many other places. Regardless of how it is used and misused (Chris is equally concerned about both), the fact that it suggests an organizing principle seemed especially relevant to the crowd. I heard nothing talked about more in the hallways.

The consensus seemed to be that there are optimal numbers to groups, and also to the ability of any one human being to scale his or her number of friendships, memberships, email correspondents or whatever.

Within that "whatever" was the subject of a session the next day titled "Fighting Distributed Wars". I wasn't the originally scheduled moderator. That was Rafe Needleman, who had a schedule conflict and swapped for my Digital Identities panel instead. No 'fence to Digital Identity, but I got the better end of the deal. In my (long) life I can't recall a more interesting panel than this one.

The main source of that interest (with no disrespect for the other panelists, who were terrific), was John Robb, a former tier one counterterror operations specialist in the Air Force, where he piloted "uninventoried" aircraft on classified missions in support of Navy SEALs and other specialists doing work for which the adjective "dangerous" would be an understatement.

John has a blog (and soon a book) called Global Guerillas, which follows terrorism in dispassionate and nonpolitical terms that are no less scary without the usual rhetorical camouflage. He talked about the high "leverage" obtained by blowing up pipelines and bringing down power towers that achieve half-billion dollar effects for investments of ten thousand dollars or less. And about how the lawless frontiers of Iraq now serve not only as training grounds for Al Qaeda and other enemies of states, but for stateless groups, mostly numbering under Dunbar's 150, whose goals are to turn civilized cultures (e.g. Saudi Arabia) into the equivalent of Tatooine -- the desert city in Star Wars that Obi-Wan called "a den of scum and villiany". Here they can take advantage of intended market effects, among other nasty deeds. This was extremely interesting stuff that had approximately nothing to do with what everybody came to Supernova to talk about, but which ended up being Topic A in the hallways.

Next, ISACA.

According to Technorati, 164 recent blogs mention Supernova, and 41 point directly to the Supernova site. None mentions ISACA, although 16 point to the ISACA.org index page. Herein lies a story.

I had never been to an ISACA event before. In fact, I hadn't even heard of the organization before I was asked to speak at the Boston event. Even now, I have to look up the name to remember how to spell it and what it stands for. I don't think I've ever encountered a less memorable name, for anything.

The event's full title was the 32nd Annual International Conference and Annual General Meeting of the Membership of the Information Systems Audit and Control Association. My talk was an "educational session" on Open Source. It was scheduled for an hour and a half on the second of the event's three days. Other sessions had titles like Implementing and Improving IT Governance, Incident Response Forensics Investigation and IT Impact of Sarbanes-Oxley Section 409. Against those, mine stuck out like a flag pole in a wheat field. Or so I thought.

There was no Wi-Fi in the rooms there. In fact, in each session I attended, mine was also the only laptop, other than the one used to project slides up on screen. When I pulled mine out and put it on the table in front of me, others looked at me like I was about to perform surgery on a salmon.

What I got out of the sessions, however, was a sense that IT (or IS, as in MIS -- the term in vogue when the organization was founded) mattered a great deal, and had enormous and highly conservative flywheels inside its engines.

To my surprise, my session was lightly attended. With about fifty students, the room was about half full. Nobody fell asleep, and I did my best to introduce a subject that was frankly alien to the majority there. While there were two or three who were open source and Linux veterans, everybody I spoke to reported that open source was still regarded with varying (though mostly high) degrees of suspicion by their organizations. For those unfamiliar with open source, it came as news that many of their core values -- resiliency, reliability, security, effectiveness, availability, quality, efficiency -- were shared by the open source community, and embodied in open source development.

When I talked about DIY-IT, some in the room were flummoxed by the very notion that autonomous and original work could bring anything but trouble to an IT organization that needed to audit everything. One insight about which we agreed was that open source thrived as IT concern mostly at an operational level. In other words, it was a matter of practical implementation, rather than "strategic" in the top-down sense. I could name specific exceptions -- such as J.P. Rangaswami, CIO of Dresdner Kleinwort Wasserstein, who was CIO of the Year in 2003 and decided to "go open" back in 1999. (J.P., significantly, was at Supernova.) But, on the whole, open source has a long way to go in many big IT shops.

But I did see that there are contexts where open source can, and will, inevitably, make sense. Meanwhile, we need to draw better Venn diagrams where the overlap between centralization and decentralization is obvious to both sides.

---------

*Disclaimer: I serve on Technorati's advisory board.