So FOSS4G 2006 is now over. A lot of cool things have come out of the conference – way too many to talk about in one entry. Perhaps the most striking thing about the conference has been the number of open source tool kits that are out there – Mapserver, GeoServer, PostGIS, KaMap, MapBender, GDAL – have all featured in workshops or talks over the last week. The presence of kits like this is a great resource, meaning that we can get on with designing better applications without worrying about the details of creating our own infrastructure from the ground up.
The OSGeo are quite a new organisation that is trying to represent the open source geo community – both software and geo-data. They’ve had a big presence at the conference and I’ve had the chance to talk with quite a few of their representatives. They have a special geodata committee that is really worth looking at. I think that grass roots organisations like OpenStreetMap can really benifit from the advice, experience and leverage of larger organisations like OSGeo, who are debating and dealing with a lot of the issues that OSM deals with everyday. Legal issues like licensing and copyright can potentially cause a lot of problems to OSM. Forming working relationships with people like OSGeo can only help to strengthen the position of OpenStreetMap.
Another popular thread at FOSS4G has been the adoption of open standards. Companies who do not adopt open standards really anoy me -I’m thinking Sony, with their ATRAC rubbish, Microsoft with… all of their formats, ESRI (has anyone tried importing XML into Arc?). Open source projects are usually a lot more supportive of open standards, an issue that was raised by Jo Walsh following my talk on OpenStreetMap. Jo asked why OSM doesnt support open standards like WMS, a standard that facilitates the sharing of Geo-data in the form of raster image files (jpg, png etc). I was quite hard pushed to answer. Is it because no one has fixed to broken script that should generate WMS? Or are we too inward looking at OSM? In the first instance, the solution is that I should do it myself (I also dislike people who always complain but never contribute!), but in the second, perhaps we need to start to think of OSM as a drop in the larger ocean of open source, open licensed geo data?