The nice thing about big conferences like ETech is that there are plenty of people who do a great job of liveblogging. If you're interested in what happened, both Phil Windley and Oren Sreebny took really excellent notes on the sessions that they attended (I linked to each of their first ETech entries). It turned out that having a laptop with 0 battery capability made it really hard to blog, or even note take. Even though I was plugged into a power outlet most of the time, I experienced several power outages, one of which was able to scramble the text file I was using to take notes. Fortunately the scrambling took place at the end of the file and most of the content was preserved. Since I don't have to try and recount the full content of the talks, here's the commentary.

I have a research background, so when I hear "advanced" or "emerging" technology, I set the bar pretty high. I was disappointed with much of the content at ETech this year for this reason -- a lot of it edged too close to emerged technology for my taste. In fairness, part of the problem is that the industry as a whole has an odd notion of what is innovative. Ray Ozzie's keynote on Live Clipboard was an example of this phenomenon. During his keynote Ray asked "why isn't the clipboard for the web the clipboard?". My question is, why are we still having to ask this question?

The 15 minute keynote format doesn't really work for me. When I hear keynote, I expect to hear someone talk about general trends in the industry, big ideas, or the overall theme for the conference. Of course, nowadays, keynotes at big conferences are places for vendors to pitch their products. The 15 minute keynotes mean that you can be subjected to more pitches per unit time.

There was some content on the program that caught an held my "attention". Linda Stone's talk tried to get at the deep human needs behind all the hubbub about attention, and Michael Goldhaber's talk was just starting to get into deeper waters when he ran out of time. Clay Shirky's keynote on moderation patterns was really interesting -- it would have benefitted from more time. Danah Boyd's session was engaging enough that it ran quite a bit into the break time afterwards. Jeff Han's keynote on multi touch interfaces was also good, and a great example of the kind of "emerging" technologies that I'd like to see more of.

There were a lot of perfectly serviceable (which is to say neither outstanding nor abysmal) talks. So many of them had to do with web 2.0 or AJAX that I had to wonder whether I was at the spring edition of the Web 2.0 conference.

There were several logistical problems which veered from O'Reilly's usually excellent conference operations. I've come to expect conference wifi to be poor, but ETech's wireless was definitely a step down from what I've experienced at OSCON (which is a great O'Reilly conference). Also, the conference had to move to a different floor of the hotel for the last day. The rooms and hallways on the new floor were much more crowded. It was already hard to get into some of the sessions, but the move made that problem even worse. I can't imagine that the O'Reilly folks were happy about that.

Oren ran a calendaring BOF which was lightly attended (lightly being around 15 people). Most people there were unaware of the existence of CalDAV, which says to me that some additional evangelism on CalDAV is needed. I did have several people tell me (in hallway conversations) that they had listened to Scott Mace's interview with Lisa Dusseault, which is a good start.

Mimi Yin talked about the principles behind Chandler's UI. The talk was well attended and afterwards there was some good conversations. Several people that I respect a lot had kind things to say about the talk. This is the first time that we've really talked about the big ideas behind Chandler's UI, so it's nice to know that some people think that we are on the right track.

The value of conferences goes well beyond what's on the program. I had plenty of excellent hallway and meal conversations. Since the population that attends ETech is a bit different from most of the conferences I attend, this was a good opportunity to meet people that I don't normally rub elbows with.

Alex Russell had the honor of being the first person to show me a real MacBook Pro. He even survived without me clubbing him and taking the fool contraption. Alex's verdict was generally positive, which only makes the waiting more painful. A number of my friends have MacBooks on order as well, it seems. I even saw a guy walking through the conference with a MacBook Pro still in the original Apple box. I told him that he could get mugged walking through the venue like that.

No visit to an O'Reilly conference can be complete without at least one photography chat with Duncan, and I had several. In December I picked up a flash for my camera, because things in Seattle were dark, and I was looking for ways to increase my shooting opportunities. Talking with Duncan had highly biased me towards natural light shooting, so I was very interested when he popped into Kathy Sierra's (excellent) tutorial packing a flash. We also spent some time talking about Duncan's new workflow using Aperture, and he's posted an essay detailing how helpful Aperture (even in it's currently flawed state) has been.

Normally I don't go into the exhibit halls at conferences for very long. The ETech exhibit hall was hard to avoid, and several of the food breaks were cleverly organized to get you to circulate into the exhibit area. Despite my earlier ranting about product pitches, I did find two noteworthy things in the exhibit hall. The first is that the OpenLaszlo team has been working to make the LZX compiler generate AJAX instead of Flash. Since the primary objection that most people seem to have to Laszlo is the use of Flash, this ought to be a good reason to reconsider. I've known for a while that that the Laszlo folks were working on having the compiler output to alternate runtimes, and as soon as David Temkin told me they were on AJAX, I realized that I should have been reading between the lines on Oliver Steele's blog.

I was also able to see a demo of Krugle, a search engine for code. I was really impressed with what I saw. The base search engine functionality seems to work pretty well, and the engine can also associate related documentation with the code that it finds. In addition to the basic search stuff, they are doing all the web 2.0 stuff, so there is an API to the engine that applications can use, and you can tag code in the engine according to your taste. This brings up the possibility of some really interesting social networking/"reputational" applications on top of the basic search functionality. Krugle is built on top of Nutch, and the Krugle folks are very open source friendly. You can expect to hear more about that in the future.

One last impression is that the folks are starting to get some traction. I saw a number of interesting demos at the microformats session.