I've just returned from one of the best conferences I've ever been to, well if not best then most different.

It was the first 'open source' conference that I have been to (humm, unless you count 'SwarmFest...') and it made a refreshing change from academic conferences. It was wired being somewhere where 90% of the attendees where hackers/geeks. (Thats hackers in the old fashond nice way, not the media adopted criminal meaning).

In particular it was nice to finaly meet some of the developers who I have been collaborating with on GeoTools for so long and never met. I spent the night before the conference started talking well into the night with Dave Blasby, the GeoServer leed when I should have been finishing of my talk for the EoGEO session the next day, so not much sleep, but the presentation seemd to go well.

The conference included a set of lighning talkes that were only 5mins each, they were fantastic, just enough info to get you interested in a project but not enough to send you to sleep.

1) Javascript is back, technologies like AJAX mean that interactive javascript pages (where the 'refresh button' is a thing of the past) are becoming an attractive way to develop pages. That and the javascript console in mozila browsers of course. I feel that google are in part responsible for this resergence, showing just what can be done with google maps, google suggest and even the gmail interface.

2) The clients have arrived! For a while now projects like GeoServer and MapServer have been serving standards based content using the WMS and WFS spects to fairly dumb clients, but projects like MapBuilder and MapBlender open up a LOT of exciting oportunities.

3) GeoSpatial is everywhere and it is becoming big business (look at the upcomming Where 2.0 conference for example). It was interesting to see that ESRI had sent a rep to take a look round and a good few other organizations were seeing if Open Source was right for them.

4) The apache model is not for all. The closing keynote from apache presidnet Dirk-Willem van Gulik was very interesting. He gave an overview of the Apache Foundation, how it worked and what it did and what we, in the geospatial community could learn from it without copying it wholesale. In particular a lot of the Apache foundations life revolves around legal issues to prevent people from suing each other! In a nutshell, we should all know where our code comes from and we should try to be as modular as possible so that we are isolated from the effects of loosing a developer or from experimental code.

There are about a dozen ideas I want to follow up on, I'll write about some of them in upcomming blogs. For another perspective on the conference from a non-map-hacker take a look at O'Reilly Radar > Nathan Torkington's blog