1) There is a distinction between “free” software and “open source” software. By far, the most talked about part of the conference was Richard Stallman’s keynote, “Free Software: Freedom, Privacy, Sovreignty.” Stallman, founder of the Free Software Foundation and the GNU project, spoke at length about the insidious ways that software is designed to control people, strip privacy and freedom, and force people to spend money when they otherwise would not. Free software does not rely on anything that is proprietary, nor does it place restrictions on people’s ability to modify it or share it, provided that they make such modifications accessible to everyone. Stallman drives a hard line between people who are committed to free software and those who stop at open source. I would argue that the distinction between the two means a lot less than the distinction between open source and proprietary or commercial software. This is particularly true in a community (GIS) that is dominated by a multi-billion dollar giant, ESRI, that is proprietary in all of the worst ways. But nevertheless, people were moved, and Stallman was able to auction off a stuffed gnu for $300 (which was donated to the Foundation).

2) Developing software in a community does not often remunerate individuals adequately, but it can generate social capital within that community. This was one of the points brought up by Paul Ramsey in his talk, Why We Code. In general, I am fascinated (and often in awe of) the efforts that people in the geospatial and library discovery community have made to develop software that benefits a very large collective. Blacklight and GeoBlacklight are two examples of this. I have worked in these communities to an extent but haven’t done active developing. Thus, I’m keen to the dynamic that Ramsey was talking about, and I wonder how others who do develop feel about it. Is the social capital real? Is it desirable in any way?

3) The geospatial world needs better, fluid tools to edit and create metadata. We discovered this (and arguably demonstrated it) in the pre-conference workshop we led on creating GeoBlacklight metadata. There were a number of people at the conference from GeoNetwork, an open source metadata editor, but there were a lot of discussions about the ever-developing standards and practices in the GeoBlacklight and OpenGeoPortal communities. We had a very fruitful meeting to talk about some possibilities, including the development of a common Solr core schema (see notes from that meeting here).

4) The “under the hood” process of Carto’s new spatial analysis features is a great project called Crankshaft. I got to see Carto’s Andy Eschbacher give an overview of the tool, which embeds Python modules within PostgreSQL database search commands. I think it’s great that Carto makes this source code available.

5) Magic still happens at Fenway Park. We got a chance to slip away from the conference and take in a game at Fenway. And like the days of old, a stirring rendition of “Sweet Caroline” brought the melancholy crowd back to life and led to a great 9th inning comeback. The Sox won on a walk off.