Last year we got 4 SoC students. This year it is unlikely we will get that many. So here are my current top 4:
1. More NAT traversal code. Universal Plug and Play, NAT-PMP and all sorts of related stuff. If student gets through it really quickly I'm sure we will be able to give him more related work to do. 2. Sorting out the plugins system. I won't have time to do this until after opennet is implemented and deployed and load balancing is sorted out. This could be a while. However client/plugin authors depend on a usable plugin system, and by making a self-contained and secure plugin system we could greatly encourage more innovative apps. I am now of the opinion that a secure plugin system - with various limits - is entirely feasible. But even separating the API from the internals of the node would be a great step forward. All that takes is completing my work so far on the new plugins API, and implementing it. 3. Many different ways of introducing people to the darknet. One-time invite references, passwords etc. Would need to be concretised into a priority list of mechanisms. 4. Better searching: Make Librarian use the new format, extend the new format to support splitting properly, bundle a user friendly spider, improve the spider's performance, make it recognise when a file isn't going to be HTML/text based on the MIME type of the file, make it pick up Frost posts and provide links to Worst etc etc. This is assuming that nextgens, me or somebody else gets around to implementing at least STS before summer; if not then STS would be a high priority item. The page: http://wiki.freenetproject.org/SummerOfCode2007 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20070310/5e568188/attachment.pgp>
