On Sun, Sep 08, 2002 at 12:00:51AM -0400, Dan Merillat wrote: > > Matthew Toseland writes: > > > > Yup. Build 500 had a combinatorial explosion of announcements (within > > the node). Build 501 largely fixed this. Build 502 fixed another nasty. > > Build 503 will be out later today, and will fix the remaining behaviour. > > ... I'm going to agree with Ian on this one: build number is getting abused. > Perhaps change a date on version.java and use the CVS version for 'minor tweaks' > or just announce the time you made the change and what snapshot it's in. > > For .5, can we add a patchlevel to the protocol version that represents > higher-level things like routing/caching decisions? At .5, it's likely that These ARE routing level decisions, or I wouldn't have increased the buildnum. > someone will implement a non-fred version of the daemon, so we don't want to > be bumping our 'compatability' number for fixing memory leaks. > > The 'check for newer version' can check the node-name (FRED or C-freenet or > whatever) for it's own name + build, and protocol rev for anyone's server. We want a different mechanism for 0.5, since the buildver can be spoofed by any dumb node op, sure. > > It's of minor importance, but something to think about for .5 > > --Dan > > > _______________________________________________ > devl mailing list > [EMAIL PROTECTED] > http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl >
-- Matthew Toseland [EMAIL PROTECTED] [EMAIL PROTECTED] Freenet/Coldstore open source hacker. Looking for $coding (I'm cheap)
msg03894/pgp00000.pgp
Description: PGP signature
