Fri 2nd · Sat 3rd · Sun 4th · Mon 5th · Tue 6th · Wed 7th · Thu 8th
Note: I'm making this trip report world-readable, but some of links go to confidential materials.
Knowledge Technologies 2001 is a new GCA conference; when I first heard about it last August during my trip to Extreme Markup Languages, I was very interested to help get it going because
Eric Miller and I served on the conference board of advisors (see also: kt2001-board egroup) along with Steve Pepper et. al. At first, I expected the board of advisors to be involved with all aspects of planning the conference, and I was surprised that, for example, the CFP went out without consulting the board. Topicmaps were much more heavily represented in the conference submissions; later we realized that the CFP had gone to all the topicmap fora and none of the RDF fora, which at least partly explains the difference. But development of RDF tutorial presentations remains a concern (as does Semantic Web Evangelism in general).
In the end, though my contribution to the board was not very substantial, but I benefitted considerably: I got on the program twice; once to give the traditional W3C part of the standards update, and once to present DAML.
Mary and the boys came a long for the trip; we travelled by car (expense report note: $0.34/mi up to air fare price, per MIT policy).
I got home from my trip to Cambridge for the all-wg shin-dig late Thursday night; actually, 1am Friday morning. We were planning to leave at 1pm Friday afternoon; I spent about
before I decided it was time to cut bait and go. We drove to Oklahoma City, where Mary's sister and her family hosted us for the night.
Saturday, we bid farewell to the cousins and headed for Austin, arriving just after 7pm; 737 miles from KC. We checked in to the Radisson Hotel & Suites Austin, (512) 478-9611 and settled down for the night.
Sunday we visited our old church, Northwest Fellowship. We caught up a bit with our friends there, parted with an over-optimistic "we're in town for most of the week; maybe we can get together for dinner one evening..."
Then we went to a picnic at eastwoods park that was the tail end of the Dean's scholars reunion. The Dean's scholars is an organization that helped me find my way among the 50,000 students when I was at U.T. Austin. Unfortunately, I got there just as Dr. Cline was leaving.
Then we paid Dr. Boyer a social call, introducing Mary to him and his home, which is one of the oldest in Austin. I left him with yet another overly-optimistic "we're in town for most of the week; maybe I can come by your office and chat about this web logic stuff..."
We went to Schlotzky's for dinner: "Funny Name. Serious Sandwich.(tm)". Yumm. They had iMac's set up for net access, but it was mostly worthless: not only could I not ssh in to get my mail, I couldn't send mail with my yahoo account; they had it blocked, presumably to avoid privacy risks.
Even though we didn't make it to the Annual Zilker Kite Festival, it was another wonderful sunny Sunday in Austin. Too much traffic, though; we don't miss that.
Sunday night I had a convenient distraction from writing my slides: formulating our response to an urgent request from Lenat to contribute a quote for their OpenCyc release. Turned out to be something of a tempest in a teapot, in retrospect.
I got up Monday and put the finishing touches on my W3C Standards Update slides. I used IE5's scheduled favorites to cache them locally -- or so I thought -- and rushed off to the conference to catch most of...
Some notes I scribbled on my copy of the program:
Next up was...
Nic claimed that Reuters had been doing knowledge management since 1851 and went on to present lots of evidence of such in a wonderfully insightful manner. Analogies like:
He cited Reagle's Eskimo Snow and Scottish Rain as background [er... to what? I forget what, exactly]. And he told a story about G. W. Bush's campaign and google where the punch-line was: links in the web need labels (aka properties ;-) in addition to start and end addresses. And he praised standardization, explaining how ISO-xyz and ISO-pdq container standards had revolutionized the shipping industry in the 1970's, eliminating the need to unpack trucks and repack the cargo at the dock.
Another nifty keynote; this time with lots of practical experience deploying groupware. For example, they studied collaboration tools and made some profoundly mundane observations about what makes a discussion forum effective:
With this insight, for a negligible investment in software development, they were able to increase the effectiveness of their groupware software a zillion times over.
At the break, I discovered that IE had downloaded only the first 8 or so of my 15 or so slides. Panic! I found an analog phone, logged in, and discovered that I had chacled my slides after producing the first 8, but not the rest.
That hurdle crossed, I went on stage to discover that I'd need to reboot to get my machine to produce a video signal. I sat down to do so, only to discover that Ron Daniel had moved my chair out of my way. Plunk, down I go behind the table, recovering with "tada!" clown dance to delight the audience a moment later. I don't think they'll soon forget me. And yes, I was wearing my W3C shirt. ;-)
Dr. James Mason gave the ISO/IEC JTC-1 SC-blah-blah update while I recovered. I noticed quite a bit of overlap betwen work in ISO and other stuff (@@TODO: take another look at his slides, which should be on the conference website RSN). I made a note to ask them to participate in the new-work@ietf.org thingy.
Then I presented the W3C part:
Notes on questions from the audience:
I went to Turner's talk on Web Services: WSDL, UDDI, etc. It was aimed at managers, so there wasn't much technical beef, save one reminder to look at pi calculus. A tidbit:
Lunchtime Insights
At conferences such as these, it's often who you sit next to at lunch that leads to some of the most interesting conversations and insights into the work of various developers and organizations. Today was no exception, and I found myself seated with XML.com author and O'Reilly editor Simon St.Laurent going head-to-head with Dan Connolly from the W3C. Anyone who thinks that the rages of XML-DEV can't be reproduced in the flesh should have been there.
Beyond the flying sparks, the lunch gave some interesting insights into the workings of the W3C and the people behind the organization. One encouraging sign was that both Connolly, the W3C's XML lead, and Eric Miller, the new lead for the Semantic Web activity, were strongly in favor of as much openness as possible -- an encouraging sign for future development of at least Semantic Web related technology at the W3C.
Knowledge Technologies 2001: Conference Diary by Edd Dumbill March 07, 2001
Some details... Simon St.L. mentioned his strong disagreement to the spread of XML Schema types into XPath, but when I asked him to send his comments on the XPath-revision requirements to the address indicated in that document, he said it's more cost-effective to be an outside agitator that to participate directly in the W3C process. I tried to explain that we're very responsive, at least in the www-xml-schema-comments and xmlschema-dev forums, but I had to admit that W3C as a whole W3C has a spotty track record with regards to responsiveness.
Good stuff: Topicmaps generated from an RDF DB. Also, RSS. All open source. "One of the good guys." See xmlfr.org and 4xt.org
Another good guy, showing how OpenTechnology.org works. EricM probably grokked more of the technical details.
Based on his WSDL processing with XSLT (First steps for Web service description processing), I encouraged him to come to our Web Services workshop.
I spent Monday evening with family and friends.
Tuesday started with a 9:45 am meeting with Freese, Miller, Pepper re XTM/Semantic Web coordination; see shared notes w/Eric).
I didn't go to the 10:30am press conference about the Cycorp OpenCyc press release. Oops.
I had an interesting chat with Marc Edgar of GE; I explained RDF to him and got a gratifying "yeah, that's pretty simple and powerful" response. We commiserated over the over-hypedness of JINI and confessed our shallow understanding of UML to each other, trying to puzzle over the UML diagrams at the back of the topicmap spec. He offered to put me in touch with the EDI folks at GE (which is about 2/3rds of the world's EDI expertise, I gather).
Over lunch, Eric M. helped me put together my DAML slides, based on a "briefing" by Mike Dean. I think it came together well.
The audience asked great questions! Almost all of them raised their hands when I asked if anybody had heard of OIL and SHOE. They asked about the limits of formal systems, trust and context, and all the good stuff.
John Sowa was scheduled to speak after me, but alas, the New England weather held him captive.
Steve Newcomb presented the topicmap processing model instead. Steve said he put the talk together in a hurry, but I think it was just about my speed; I think the lightbulb finally went off. This is really their data model, but the discussion of it in terms of processing obscured it from my view for the longest time; their discussion of "topic merging" treats it as a computational effect, rather than a logical inference; as if the inference that 2+2=4 were a consequence of some physical computational process necessarily happening over some amount of real time.
I spent Tuesday evening with family and friends again (different friends this time).
[@@story-telling peters out at this point.]
Ontopia demo; techquila "a shot of the hard stuff"; what was the guy's name? khaled(sp?) long black curly hair. Is this tmproc thingy what he was showing me?
during the demo, I learned topicmaps are undirected; all links go both ways (scalability no-no!)
2p closing panel
3p Edd D's closing (analagous article)
Friday's with EricM, Edd D, Dave B. (photo by the waiter)
drive back to KC. Long drive. left around 11am; couldn't afford to stop and see the cousins on the return leg; Justin was bummed. Got home around 3am.
see also: BOS Feb trip (including links to earlier trips)
