Fred,
thank you for your and Philip's work, I and the rest of the community incredibly appreciate that.
Me and another friend are volunteering for a charity that works with kids, hence the need for content filtering. We have discussed about potentially dedicating some time to help with this effort (he's fairly familiar with C++, I unfortunately am not, but willing to give it a shot).
We've also considered a donation, if we had to buy a product we'd end up spending money anyway so we might as well invest this money in E2G, but given it's a charity it would be a small sum most likely (which is also why buying commercial products isn't much of an option). That however made me wondering if we could potentially do a crowd funding campaign and raise some money. There is not a single icap content filtering server out there and I bet that most places using DG/E2G would want transparent SSL filtering. Going to knock on some of those companies's door you mentioned could also make sense.
Organizations
Open source projects apply to be mentor organizations. Once accepted, organizations discuss possible ideas with students and then decide on the proposals they wish to mentor for the summer. They provide mentors to help guide each student through the program.
As an aside, design wise it's unclear to me if ICAP is what's best or eCAP is. Looking at Squid's pages it seems ecap is their design choice and what they are pushing for going forward. And it makes sense to me since basically the proxy has to implement an interface and that could be ecap and with ecap there's less overhead and more importantly less stuff to manage (you don't have another daemon and code in the adapter to maintain all of those things). It basically becomes a "simple" plugin doing a single job with no other code in it. Even if you were to run a large cluster for example, you can cluster squid and scale that horizontally which has already the means to do so instead of having to maintain two services and say scale icap servers.
Best,
Spike