You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
389 lines
17 KiB
389 lines
17 KiB
<?xml version="1.0"?>
|
|
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
<html xmlns="http://www.w3.org/1999/xhtml">
|
|
<head>
|
|
<title>Dan's trip to KT2001 in Austin, Mar 2001</title>
|
|
<meta http-equiv="Content-Type" content="text/html" />
|
|
</head>
|
|
|
|
<body>
|
|
<p><a href="../../">W3C</a> <a href="/People/Connolly/">Connolly</a></p>
|
|
|
|
<h1>KT2001 in Austin</h1>
|
|
|
|
<p style="text-align: center"><a href="#L647">Fri 2nd</a> · <a
|
|
href="#L655">Sat 3rd</a> · <a href="#L665">Sun 4th</a> · <a href="#L687">Mon
|
|
5th</a> · <a href="#L695">Tue 6th</a> · <a href="#L703">Wed 7th</a> · <a
|
|
href="#L711">Thu 8th</a></p>
|
|
|
|
<p><strong>Note</strong>: I'm making this trip report
|
|
<strong>world-readable</strong>, but some of links go to confidential
|
|
materials.</p>
|
|
|
|
<h2>Prologue</h2>
|
|
|
|
<p><a href="http://www.knowledgetechnologies.net/">Knowledge Technologies
|
|
2001</a> is a new GCA conference; when I first heard about it last August
|
|
during <a href="/Team/connolly-200011#2000,1">my trip to Extreme Markup
|
|
Languages</a>, I was very interested to help get it going because</p>
|
|
<ul>
|
|
<li>its focus is very hear the heart of the Semantic Web, and</li>
|
|
<li>it was in Austin, where I learned most of <a
|
|
href="/Collaboration/knowledge">what I know about knowledge
|
|
representation</a>.</li>
|
|
</ul>
|
|
|
|
<p>Eric Miller and I served on the <a
|
|
href="http://www.gca.org/attend/2001_conferences/kt_2001/welcome.htm"><strong>conference
|
|
board of advisors</strong></a> (see also: <a
|
|
href="http://www.egroups.com/group/kt2001-board">kt2001-board</a> 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 <strong>development
|
|
of RDF tutorial presentations</strong> remains a concern (as does
|
|
<strong>Semantic Web Evangelism</strong> in general).</p>
|
|
|
|
<p>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.</p>
|
|
|
|
<h2><a name="L647" id="L647">Fri, 2 Mar</a></h2>
|
|
|
|
<p>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 <a
|
|
href="http://web.mit.edu/afs/athena.mit.edu/org/c/cao/www/travpol2.htm">MIT
|
|
policy</a>).</p>
|
|
|
|
<p>I got home from <a href="../02dc-bos/Overview.html">my trip to Cambridge
|
|
for the all-wg shin-dig</a> late Thursday night; actually, 1am Friday morning.
|
|
We were planning to leave at 1pm Friday afternoon; I spent about</p>
|
|
<ul>
|
|
<li>2 hours trying to get tools and background materials for putting
|
|
together my presentations onto my new laptop</li>
|
|
<li>5 or 6 hours sleeping</li>
|
|
<li>5 more hours doing a little office work, packing and fiddling with the
|
|
laptop</li>
|
|
</ul>
|
|
|
|
<p>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.</p>
|
|
|
|
<h2><a name="L655" id="L655">Sat, 3 Mar</a></h2>
|
|
|
|
<p>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, <a href="tel:+1-512-478-9611">(512) 478-9611</a> and settled
|
|
down for the night.</p>
|
|
|
|
<h2><a name="L665" id="L665">Sun, 4 Mar</a></h2>
|
|
|
|
<p>Sunday we visited our old church, <a
|
|
href="http://www.northwestfellowship.com/">Northwest Fellowship</a>. 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..."</p>
|
|
|
|
<p>Then we went to a picnic at eastwoods park that was the tail end of the <a
|
|
href="http://www.cs.utexas.edu/users/cline/alumni/AlumniActivities.html">Dean's
|
|
scholars reunion</a>. The Dean's scholars is an organization that helped me
|
|
find my way among the 50,000 students when I was at <a
|
|
href="http://www.utexas.edu/">U.T. Austin</a>. Unfortunately, I got there just
|
|
as Dr. Cline was leaving.</p>
|
|
|
|
<p>Then we paid <a href="http://www.cs.utexas.edu/users/boyer/">Dr. Boyer</a>
|
|
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..."</p>
|
|
|
|
<p>We went to <a href="http://www.schlotzskys.com/">Schlotzky's</a> 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.</p>
|
|
|
|
<p>Even though we didn't make it to the <a
|
|
href="http://www.io.com/house/introkites.html">Annual Zilker Kite
|
|
Festival</a>, it was another wonderful sunny Sunday in Austin. Too much
|
|
traffic, though; we don't miss that.</p>
|
|
|
|
<p>Sunday night I had a convenient distraction from writing my slides: <a
|
|
href="http://lists.w3.org/Archives/Team/w3t-pr/2001Mar/0015.html">formulating
|
|
our response</a> 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.</p>
|
|
|
|
<h2><a name="L687" id="L687">Mon, 5 Mar</a></h2>
|
|
|
|
<p>I got up Monday and put the finishing touches on my <a
|
|
href="../Talks/0305-kt-upd/">W3C Standards Update slides</a>. I used IE5's
|
|
scheduled favorites to cache them locally -- or so I thought -- and rushed off
|
|
to the conference to catch most of...</p>
|
|
|
|
<h4>Doug Lenat's keynote</h4>
|
|
|
|
<p>Some notes I scribbled on my copy of the program:</p>
|
|
<ul>
|
|
<li>OpenCyc became public knowledge at 15:32Z, with a press release to
|
|
follow Tuesday at 10:30am local time. It's a series of quarterly releases,
|
|
with increasingly liberal license terms, increasing depth and breadth of
|
|
the knowledge base, increasing functionality of tools.</li>
|
|
<li>Lenat contrasted Cyc's approach with the traditional "full understanding
|
|
or bust" approach. Harks to our own principle of <a
|
|
href="/DesignIssues/Evolution#PartialUnderstanding">Partial
|
|
Understanding</a>.</li>
|
|
<li>Something he said about giving up global consistency reminded me of the
|
|
way the Web sacrificed consistency for scalability (cf <a
|
|
href="/DesignIssues/Topology.html">Topology</a> in <a
|
|
href="/DesignIssues/">Design Issues</a>)</li>
|
|
<li>In response to a "how many contexts will there be?" question, Lenat
|
|
answered that rather than treating contexts as arbitrary first class
|
|
objects, their approach has shifted to treating them as points in a
|
|
multi-dimensional context space. cf <cite><a
|
|
href="http://www.cyc.com/context-space.txt">The Dimensions of Context
|
|
Space</a></cite> among the <a
|
|
href="http://www.cyc.com/publications.html">cyc publications</a>.</li>
|
|
</ul>
|
|
|
|
<p>Next up was...</p>
|
|
|
|
<h4>Nic Fulton keynote</h4>
|
|
|
|
<p>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:</p>
|
|
<dl>
|
|
<dt>Industrial Revolution</dt>
|
|
<dd>tupperware parties</dd>
|
|
<dt>Information Revolution</dt>
|
|
<dd>online communities</dd>
|
|
<dt>Tidbits:</dt>
|
|
</dl>
|
|
<ul>
|
|
<li>Reuters has 2000 journalists/cameramen</li>
|
|
</ul>
|
|
|
|
<p>He cited Reagle's <cite><a href="/TR/md-policy-design">Eskimo Snow and
|
|
Scottish Rain</a></cite> 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.</p>
|
|
|
|
<h4>Scott Cooper, Lotus</h4>
|
|
|
|
<p>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:</p>
|
|
<ul>
|
|
<li>keep the mission of the forum visible ("in your face")</li>
|
|
<li>identify a facilitator</li>
|
|
</ul>
|
|
|
|
<p>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.</p>
|
|
|
|
<h4>Standards Update, 11am</h4>
|
|
|
|
<p>At the break, I discovered that IE had downloaded only the first 8 or so of
|
|
my 15 or so slides. <strong>Panic!</strong> I found an analog phone, logged
|
|
in, and discovered that I had chacled my slides after producing the first 8,
|
|
but not the rest.</p>
|
|
|
|
<p>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.
|
|
<strong>Plunk</strong>, down I go behind the table, recovering with
|
|
"<strong>tada!</strong>" 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.
|
|
;-)</p>
|
|
|
|
<p>Dr. James Mason gave the ISO/IEC JTC-1 SC-blah-blah update while I
|
|
recovered. I noticed quite a bit of <strong>overlap betwen work in ISO and
|
|
other stuff</strong> (@@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 <a
|
|
href="http://lists.w3.org/Archives/Team/ietf-new-work/">new-work@ietf.org</a>
|
|
thingy.</p>
|
|
|
|
<p>Then I presented the W3C part:</p>
|
|
<ul>
|
|
<li><a href="../Talks/0305-kt-upd/">slides</a></li>
|
|
<li><a href="http://heddley.com/edd/2001/03/07/m_austin-018.jpg">photo</a>
|
|
by Edd D.</li>
|
|
</ul>
|
|
|
|
<p>Notes on questions from the audience:</p>
|
|
<ul>
|
|
<li>something about whether XML Schema evolution; would templates be
|
|
supported? I answered that there were tools to convert from DTDs, UML, and
|
|
other more familiar notations, hoping that's what the guy meant by
|
|
templates. Not sure I was on target at all, though.</li>
|
|
</ul>
|
|
|
|
<h4>Turner on Web Services</h4>
|
|
|
|
<p>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:</p>
|
|
<ul>
|
|
<li>>200 businesses participate in uddi.org</li>
|
|
</ul>
|
|
|
|
<h4><a name="lxdev">Lunch with the xml-dev crowd</a></h4>
|
|
|
|
<blockquote>
|
|
<h3>Lunchtime Insights</h3>
|
|
|
|
<p>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.</p>
|
|
|
|
<p>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.</p>
|
|
<address>
|
|
<p><a href="http://www.xml.com/pub/a/2001/03/07/kt1.html">Knowledge
|
|
Technologies 2001: Conference Diary</a> by Edd Dumbill March 07, 2001</p>
|
|
|
|
</address>
|
|
</blockquote>
|
|
|
|
<p>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.</p>
|
|
|
|
<h4>Implementor Track: Eric van der Vlist</h4>
|
|
|
|
<p>Good stuff: Topicmaps generated from an RDF DB. Also, RSS. All open source.
|
|
"One of the good guys." See <a href="http://xmlfr.org/">xmlfr.org</a> and <a
|
|
href="http://4xt.org/">4xt.org</a></p>
|
|
|
|
<h4>Implementor Track: Uche Ogbuji</h4>
|
|
|
|
<p>Another good guy, showing how <a
|
|
href="http://OpenTechnology.org/">OpenTechnology.org</a> works. EricM probably
|
|
grokked more of the technical details.</p>
|
|
|
|
<p>Based on his <a
|
|
href="http://www-4.ibm.com/software/developer/library/ws-trans/index.html?dwzone=ws">WSDL
|
|
processing with XSLT (First steps for Web service description processing)</a>,
|
|
I encouraged him to come to our Web Services workshop.</p>
|
|
|
|
<p>I spent Monday evening with family and friends.</p>
|
|
|
|
<h2><a name="L695" id="L695">Tue, 6 Mar</a></h2>
|
|
|
|
<p>Tuesday started with a 9:45 am meeting with Freese, Miller, Pepper re
|
|
XTM/Semantic Web coordination; see <a href="xr23">shared notes
|
|
w/Eric</a>).</p>
|
|
|
|
<p>I didn't go to the 10:30am press conference about the <a
|
|
href="http://www.cyc.com/opencycpressrelease03062001.html">Cycorp OpenCyc
|
|
press release</a>. Oops.</p>
|
|
|
|
<p>I had an interesting chat with <strong>Marc Edgar of GE</strong>; 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 <a
|
|
href="http://www.topicmaps.org/xtm/1.0/">topicmap spec</a>. He offered to put
|
|
me in touch with the <strong>EDI folks at GE</strong> (which is about 2/3rds
|
|
of the world's EDI expertise, I gather).</p>
|
|
|
|
<p>Over lunch, Eric M. helped me put together my <a
|
|
href="../Talks/0103daml-kt/">DAML slides</a>, based on a "briefing" by Mike
|
|
Dean. I think it came together well.</p>
|
|
|
|
<p>The audience asked <strong>great questions</strong>! 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.</p>
|
|
|
|
<p>John Sowa was scheduled to speak after me, but alas, the New England
|
|
weather held him captive.</p>
|
|
|
|
<p>Steve Newcomb presented the <strong>topicmap processing model</strong>
|
|
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.</p>
|
|
|
|
<p>I spent Tuesday evening with family and friends again (different friends
|
|
this time).</p>
|
|
|
|
<h2><a name="L703" id="L703">Wed, 7 Mar</a></h2>
|
|
|
|
<p>[@@story-telling peters out at this point.]</p>
|
|
|
|
<p>Ontopia demo; <a href="http://www.techquila.com/">techquila</a> "a shot of
|
|
the hard stuff"; what was the guy's name? khaled(sp?) long black curly hair.
|
|
Is this <a href="http://www.ontopia.net/software/tmproc/">tmproc</a> thingy
|
|
what he was showing me?</p>
|
|
|
|
<p>during the demo, I learned <strong>topicmaps are undirected</strong>; all
|
|
links go both ways (scalability no-no!)</p>
|
|
|
|
<p>2p closing panel</p>
|
|
|
|
<p>3p Edd D's closing (analagous <a
|
|
href="http://www.xml.com/pub/a/2001/03/07/buildingsw.html">article</a>)</p>
|
|
|
|
<p>Friday's with EricM, Edd D, Dave B. (<a
|
|
href="http://heddley.com/edd/2001/03/07/m_austin-019.jpg">photo</a> by the
|
|
waiter)</p>
|
|
|
|
<h2><a name="L711" id="L711">Thu, 8 Mar</a></h2>
|
|
|
|
<p>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.</p>
|
|
|
|
<h2>Colophon: how this page was made</h2>
|
|
|
|
<p>see also: <a href="../02dc-bos/Overview.html">BOS Feb trip</a> (including
|
|
links to earlier trips)</p>
|
|
|
|
<h2>TODO</h2>
|
|
<ul>
|
|
<li>put slides on /Talks/; on homepage?</li>
|
|
<li>expense report</li>
|
|
<li>read the topicmap spec again, trying to remember what I wrote on my lost
|
|
copy (which wasn't too much: definitions out of context considered
|
|
harmful, some character encoding issue, maybe a few other things)</li>
|
|
<li>RDF/n3 model of topicmaps? merging rules look straightfoward</li>
|
|
</ul>
|
|
<hr />
|
|
<address>
|
|
in progress Mar 2001<br />
|
|
<a href="http://www.w3.org/People/Connolly/">Dan Connolly</a><br />
|
|
<small>$Revision: 1.18 $ of $Date: 2001/03/16 04:42:29 $ by $Author:
|
|
connolly $<br />
|
|
<a href=",access">ACL</a></small>
|
|
</address>
|
|
|
|
<p></p>
|
|
</body>
|
|
</html>
|