Definitely with you if it looks like being worth setting something up. One up-front question though. If we go down this route should it be a non-profit or for-profit? I'm agnostic about this. On one level not-for-profits can re-distribute money into development and/or supporting other activities (e.g. biogang). On the other hand profit motives can bring in more interested supporters. Just want to raise the question. As I said elsewhere, would be willing to put in cash into a pool to support someone with the expertise to look at the existing options in terms of build requirements and scalability. Maybe NPG/PLoS/Seed/Google/Amazon/Microsoft... might be interested in making small contributions? - Cameron Neylon

@Cameron: as long as it is Open Source, it does not matter... commercial is not evil, proprietary and vendor lock-in is... - Egon Willighagen

Egon, absolutely agree but I'm talking less about the software and more about the organization we might want to build around it. The nature of the organization can have a strong influence on how it behaves. A for-profit might move away from our needs (for perfectly rational reasons) but at the same time might bring in more resources. A not-for-profit could be more focussed but might not have the same resources - Cameron Neylon

It is intrinsically difficult to have one organization (company, foundation) to serve a community... traditionally, many such things are merely loose collection of servers at universities around... I would start that way... it's cheaper, and more community driven, ensuring tight integration of the service and the community... - Egon Willighagen

Its cheaper but is it realistic to build something sensible with no resources? A lot of boring development work is likely to be involved surely and it also has to happen quite fast to take advantage of the current opening. Can this be done without actually paying a few people to do it? Or at least offering them the opportunity of a pay out further down the track? - Cameron Neylon

@Cameron... sounds like a postdoc position to me :) - Egon Willighagen

If our blog3 grant comes through (which is still completely up in the air), we'd have three full-time developers sitting here working on related projects... We'll know more early 2010. - Björn Brembs

Federation would help things along quite drastically: http://friendfeed.com/clonefe... - Björn Brembs

Neil, I agree but hopefully at some point we reach a stage where we could design something useful including building in the options for serendipity. Worth having the discussion as to whether we are there yet, and if not (probably the case) what more do we need to know? - Cameron Neylon

It's not a trivial decision. Distributed projects work, but only under very special circumstances, especially when not done full time. Can it be done, yes. Lots of open source projects started that way. If we do it, it has to be Apache licensed. I don't see any other way. Having said that, just in the last few days, FF has added some subtle, but really nice features. That continuous iteration cycle (which Ian points out in a comment to this post on the blog) is one of FF's strengths, and quite honestly, I'd like to see where it goes, cause I still really like it :) - Deepak Singh