The last day of the Semantic Technology Conference has a few morning panels, a closing keynote, and then some afternoon seminars. But the day is really about saying goodbye to everyone, finally introducing yourself to a few of the people you’ve been crossing paths with for the past week, and making that annual trip to Koo-ki Sushi. Well, that’s what it’s about for me, anyway.

In the morning I went to a presentation on Semantic MediaWiki. Several guys from Monitor Group spoke about what they’ve learned in the year since they implemented semantic plug-ins on their internal wiki. I was particularly interested in this because at Avenue A | Razorfish we have an enterprise wiki of our own, and I’ve been trying to push to make it more semantically-enabled. The Monitor folks want to start a Semantic MediaWiki User group and and I told them to sign me up.

Then there was the closing keynote, which was about the business model of semantics and what you should tell your colleagues about the conference when you get back to the office. I really loved this panel, because there was a wide range of perspectives and backgrounds represented and they did a great job of, collectively, taking the measure of the event. Everyone has the sense that this is a discussion that’s on the brink of cracking open some major mysteries about the way we use the web, and it’s very exciting to be a part of it. And at the same time, we’re left with the feeling that next year is going to be even more interesting.

I’ll paraphrase some of the insights shared during the closing keynote:

• This stuff isn’t all at the beginning of the hype curve, some of it has made a couple cycles already. (That comment was for the folks who are new to the scene) (Jim Hendler)
• If you’re describing the value of something as being about semantic technology, you probably haven’t found the real business value yet. (Jeff Pollock)
• The key message is simplify – both in how we talk about it and how we do the work. (Jonathan Mack)
• “What vs. How” – figure out what we are trying to solve, then figure out how semantic technology can help. (Christine Connors)
• There’s a large variety of application areas, so there’s no one business model. There are different communities with different needs. (Ivan Herman)
• Semantics has no business model, but it has the potential to disrupt other business models before it creates new ones. LinkedIn would be dead in the water if everyone had a FOAF profile. (Steve Hall)
• People forget what the “I” in IT stands for. We’ve been doing a lot of technology without paying a lot of attention to the information side. (John Gilman)
• Never say “Ontology”. (John Gilman)

And what to tell your colleagues?

• There’s an embarrassment of riches here. (John Gilman)
• This isn’t the breakout year for this stuff, but it’s coming soon. (Steve Hall)
• Take the right sized step for your project. Just solve one discrete problem at a time, and don’t try to go back and fix everything. (Christine Connors)
• There’s a lot of real valuable services that already work. We need to get started before our competitors. The implications for the collaboration model are big. (Jonathan Mack)
• In a downturn year, this conference is still growing. (Jeff Pollock)
• People have big ideas here. (Jim Hendler)
• There’s no reason now not to do this. (Dave McComb)
  

Then I headed over to Koo-ki Sushi, which is sort of an annual pilgrimage for me when I come to the Semantic Technology Conference. My first year here I presented with Barbara McGlamery and for our talk we made up a fake website called “Confection Connoisseur” (which I used in this year’s presentation a bit too, so it’s actually become a running theme now). In the course of finding photos for our site mockups, we discovered this place Koo-ki Sushi which makes beautiful chocolates in the form of sushi. They’re handmade, and only sold in a little office park store front in San Jose (or you can buy them online). Of course we had to go there, and I make it a point to go back there every year.

I returned in time to go to the second half of a seminar on Entity Extraction tools. This is the part where they were actually describing the tools they evaluated, which is the part I was most interested in. Mostly I’m going to bring this information back to the tech folks and see if they want to investigate further.

Coming Soon: My wrapup of the conference, including other highlights and tidbits not previously mentioned.

• Day One: Tutorials (sioc & Freebase), Garlik & Twitter
• Day Two: Twine, Zepheira, Blue Organizer, Mainstream Markets & exhibits
• Day Three: My presentation, Future Directions, Calais, Rising Stars & my panel