HPTS: the return

I'm back from HPTS. From what I've been told, the submitted papers and presentations should be available at the web site soon. Definitely worth a look. Savas had a great first time experience and was even designated the unofficial workshop photographer - until his batteries ran out!

There were some really good presentations, but it's always the talks in and around the events that I find the most interesting. Maybe there are other conferences/workshops like this, but I've never been to anything where the sense of community is so high and people are able to transcend any normal company relationships. The feeling is more akin to a group of co-workers/co-researchers all collaborating on the same problem and getting together to share experiences (often over the odd beer or two).

In answer to my earlier question, there were more new faces than I saw when we arrived and the quality of the work is just as high as in previous years. Probably my only disappointment with this year's workshop was its almost total lack of controversy and heated (though good) discussions: they've always been a characteristic of previous workshops, but not this time. It's a shame, because I always find them stimulating. However, nothing can really spoil these workshops, particularly when you look at the the surroundings.

My presentation went alright, though it is very difficult to summarise 20 years of development into 20 minutes! After talking to Jim it's clear that the work his group is doing on their Lightweight Transaction Manager is very similar to what we did on Arjuna with recoverable and durable+recoverable objects. It encourages developers to use transactions as a fault-tolerant structuting mechanism without having to always suffer the performance penalties. If it's supported in a promotable manner too, this can be a very nice and powerful facility to provide developers.

The trip back was slightly more eventful than the one down. Our original plans were to drive back and to stop somewhere to eat. Unfortunately, my flight out didn't allow us that option, so we had to go straight up 101. Well, that was the plan. Maybe it was Paul's radio selections, including Billy Joel, Madonna or the Bee Gees, or the discussions about work to come that put Savas off, but we ended up taking a more scenic route to the freeway than intended (and I bet we couldn't retrace our steps now if we wanted to). I was slightly late checking in for the flight back and didn't get a chance to eat prior to take-off, but that (and the serious case of jet-lag) was a small price to pay for the trip.

On the road to HPTS

We finally arrived at HPTS and it was a nice journey. The four of us (myself, Paul Watson, Savas and Dene Kuo) hired a car and drove from San Francisco down Highway 1. We were all probably showing our ages, listening to a-ha, Bon Jovi and others I care not to mention, as we drove. The scenery is great, the drive is sedate and I'd recommend it to anyone who has a chance. We stopped at a few places to stretch out legs and just take it easy: Santa Cruz was great for lunch, despite it being overrun with highschool cheerleaders, like something out of Bring It On, and we stopped for coffee in Carmel, with a walk on the beach. We finally arrived at Asilomar and as usual with this place, it's like stepping out of the real world and into the Twilight Zone: beauty, quiet, little of the 21st century comforts you're used to, and no real concept of time to interrupt some good work to come.

Today was mainly checking in and socialising with the rest of the invited attendees. It's been good to catch up with Jim Johnson, Jim Grey, Pat Helland and others I probably only get to see at this workshop. This year there seem to be less new faces than in previous years. I'm not sure what that means: are transactions so "un-cool" that it doesn't attract people as much, have all the really interesting and ground breaking things in transactions been done already, or something entirely different? I'm looking forward to tomorrow when the real work begins; maybe that'll help answer the question.

Update: I've been asked by Paul to point out that although he was controlling the radio in the car, he wasn't responsible for what we listened to. ;-)

WS-Context useage

I'm at HPTS this week. Will blog about that when I get a chance and if I can find internet access at Asilomar (currently at an airport hotel for the night). That's the thing about a retreat like Asilomar: no TV, no phones in rooms and little/no connectivity; I may even have to resort to the good 'ol modem via the pay-phone!

Anyway, a friend from the School of Computing at Newcastle Univeristy pointed me at this, which I think is really interesting. I'm glad to see more groups taking up the WS-Context model and hope it continues.

Update: I should thank Savas for his help on promoting WS-Context; this is one of the results.

HPTS agenda

The HPTS agenda is shaping up. Although the three papers that I/we submitted were accepted, I'm only talking about the evolution of ATS paper. But that's good: I'm having enough trouble finding time to write just that one presentation!

Microsoft and ESBs

Here's an interesting article. So far ESBs have played predominately in the Java market, but there's nothing inherent in the concept to limit it. So, it's interesting to hear Microsoft's take on things.