Hi, In my case Fproxy is more important. I never found frost that effective.
Ed On Friday 18 November 2005 16:36, Matthew Toseland wrote: > Which of these is more important? > - SSKs -> minimal FCPv2 -> working frost boards, and maybe working frost > filesharing too. (splitfiles work right now) > - Manifests -> SSKs -> Fproxy (with initially command line site insert). > - Opennet support. > - Ability to fetch logs from nodes... > > Current issues: > 1. We need money. Fast. > One way to get this is to get some general enthusiasm amongst the > Freenet userbase. Another is to get some amongst the slashdot crowd. > Either way, we need more people on the network, and we need more apps. > 2. We need more nodes. We need more people on the network. > Problems: > a) To debug properly, I need people to be online when their node > is, so they can send me their logs if I want them. Possible fix: Make > the node automatically send me logs if I ask for it; minimal testnet > support. Then we can have all nodes online more or less permanently. > b) Can just ask for more testers. (Come to #freenet-alphatest on > irc.freenode.net if you want in!) > c) Opennet is a possibility, but we don't want to have opennet support > before we have shown that the darknet can vaguely work, and finished > with the datastore format changes. Opennet is much less tested > theoretically than darknet. > 3. We need apps. > a) Frost. If I implement SSKs, then a minimal FCP, then Frost should > just about work, once it is rewritten a bit. > b) Fproxy. If I implement SSKs, and manifests, then I can implement > Fproxy, with sites initially being inserted through the command line > interface. > > IMHO it is not yet time for opennet support, as we haven't finished > playing with the datastore... > > The current testnet is very small.
