It might be reasonable to do darknet invitations: a file which allows someone to connect with another node over darknet without exchanging noderefs. (I can't think of how this would work, though.) This could also, perhaps more conveniently, come baked into a generated installer. In a similar vein, generated installers could include additional plugins as well. When I helped someone set up Freenet, FRED itself was fast, but all the plugins (WoT, Freetalk, FMS, Sone) took a while to download. Sure, I can carry the .jars/.zips around with me, but that's annoying and not streamlined.
More extensively, which probably falls outside the scope and intent of gun.io, it would be very nice to have code cleanup, review, and documentation. One cleanup task that occurs to me is separating FProxy and the rest of the UI from - I'm not sure what to call it: the core? Such work could involve using FCP to interact with the backend and using an established templating engine like Apache Velocity Engine. Review could catch bugs and inform any possible cleanup and documentation. Documentation would allow others to add Freenet support to their existing products, or write their own Freenet daemons. It'd also be good for FRED to have a protocol and load balancing specification to work from. These projects are probably still be too large, and I'm not sure about putting them on gun.io as I'd like to do them myself, but I'm interested in realtime public chat using WoT for identities, and more chat programs/plugins which are compatible with FLIP. I'm hoping to work on that myself in the form of additions to my chat plugin. There have also been many posts about a filesharing plugin, which I think is a great idea, but I haven't noticed reports of progress on any of them so far. Most recently, f?nfnull's "What Freenet really needs" in eng.freenet on Freetalk mentions this. How about posting subtasks for cleaning up WoT/Freetalk? -Steve On Mon, Jan 9, 2012 at 10:38 AM, Ian Clarke <ian at locut.us> wrote: > > On Thu, Jan 5, 2012 at 8:28 PM, Michael Grube <michael.grube at gmail.com> > wrote: >> >> If I understand correctly, you pick a task that nobody has time to do. You >> then outsource that task. People then compete for your money with their >> solutions. >> >> It's like TopCoder but you get paid to win! Awesome. > > > Right, and I think anyone can offer to pay people to perform specific tasks, > not just the project. > > What are some good candidate problems for people to address? ?We could invite > both donors and programmers for each problem... > > Ian. > > -- > Ian Clarke > Personal blog: http://blog.locut.us/ > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://freenetproject.org/cgi-bin/mailman/listinfo/devl
