On Thu, 4 Sep 2003, Robert Norris wrote: >> A Java implementation for transports would be really cool. I think one of >> the major reasons the transports do not really evolve beyond "it works for >> basic IM and no more" is because of C. > No, I think it doesn't get much further than just barely working is that > writing transports is frustrating, thankless and exceptionally un-sexy > work, and few people have the ability to push through that pain. > I've worked on transports, I've seen the hell first hand .. I know ..
Well I have contributed quite a bit to AIM/ICQ-t and JIT. AIM/ICQ-t is frustrating since libfaim is far from intuitive. JIT is a lot better (since it uses libicq2000 which is coded in C++). Be assured if there was a Java transport I already would have coded *by far* more (in fact I hope to combine JGF and MIU once they both reach a usable state). >From my point of view JIT greatly benefits from the fact that libicq2000 is coded in C++. Also, the fact that JIT is at least a bit commented with Doxygen comments (Doxygen docs for JIT are generated automatically and available from its website) makes understanding the code a great deal easier. I find it a pity that neither jabberd14 nor jabberd2 nor most other projects (transports...) make use of Doxygen or something similar or have docs online. You are the jabberd2 author?- I'd like to contribute by helping with code docs. Contact me by mail if you agree. Regards _______________________________________________ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
