Following on from the Access2OER discussion, one of the participants asked me whether I had any thoughts on linux and access to OER, that might be relevant to the Linux Foundation 'Collaboration Summit'.
There are of course many linux projects that already are (and could) be of very high value within an international development context. One of the biggest contributions linux and open source can make, and one of the biggest needs, is of course education, particularly for schools. So if I had "bridging the digital divide" proposal to put to the linux community, it would be to make linux more usable for schools. Of course, there already is edubuntu, and I don't know how the edubuntu community efforts relate to the "educational" efforts of the linux community at large. But it may well be that the simplest proposal would be to say: "Support edubuntu: http://www.edubuntu.org/Community".
Moving on to more concrete issues, I'll give a "usability" example: for a rural ICT project in Samfya, we used LTSP, to set up a 15 seat thin-client, low-power lab. As described in Samfya, open source played an important role in the success of the event, and the kind of approach employed could really scale.
However, often the issue isn't "features", but "usability". From a system admin point of view, it was quite tricky to set up: We split 15 clients across two servers, and there was a lot of configuration that had to be sorted out with the network, to set up LTSP, set up raided disks, to get printing to work, etc. And many issues remain: For instance there isn't a simple procedure to replace a disk (and reinstate the RAID config accordingly) if a disk fails, etc. So on the one hand there is huge potential, on the other hand, unless you're a specialist, it's hard to draw on this potential. It was a similar story with some of the packages we tried, such as http://italc.sourceforge.net/. It's a very necessary project, it's very promising, but we didn't quite manage to get it to work as we needed (even for our linux specialists).
I have often entertained the thought of a 'linux usability swat team'. In a joint effort with educators, one would first identify (and prioritise) the needed features and applications (for educational / development context). Then the 'swat team' would examine the highest priority features, to make sure they have high usability, and to make sure they run well with edubuntu / LTSP etc. Where they don't run well, the swat team fixes the relevant features. Once the issues for one feature/app are fixed, the swat team moves on to the next.
So if there was a "linux for development" proposal, that would be it: Identify needed features/applications for (ed)ubuntu, fix usability as needed, make sure they make it into (ed)ubuntu.
Bearing in mind my comments above about edubuntu: I don't know the linux community that well, and I don't know to what extent these thoughts would chime with the Linux Foundation 'Collaboration Summit'. Also the above proposal probably already has been made many times, and may indeed already be happening in the edubuntu community. But perhaps a "usability for development swat team" would fit well with existing efforts (such as edubuntu), and might attract enough developers to be viable.