Quoting Jeroen Asselman <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel: :Op 20-sep-04 om 8:17 heeft Raphael Manfredi het volgende geschreven: :> Also perhaps PARQ should be re-implemented from scratch now that we :> have :> some experience with it. Like they say, "throw One away". :> :That sounds almost as if you think my implementation is flawed ;)
Not at all. It's a fact that all first implementation of complex protocols are hard to get right the first time. By "right" I don't refer to correctness (all implementations MUST be correct) but rather to other quality criteria like maintainability, flexible architecture, good interfaces, etc... In particular, the "contract" and the interface between PARQ and the upload and download systems need to be re-thought. Note that I can criticize myself on other parts of GTKG which are also complex: the management of the download mesh for instance is a nightmare and could be revisited. The download refactoring I did two years ago did not go far enough and the scheduling is fragile. The management of the download response from the server is not modular enough. So don't view the critic as something negative. It's an objective assessment I'm doing. Naturally, there are reasons why things stay in that situation. It's not exactly "if it isn't broken, don't fix it", but rather "if it does not need to evolve, don't fix it". I think evolution is a great incentive to refactor code, and the right moment to do it. So if we do PARQ/1.1, it will be good to rethink the whole implementation. Raphael ------------------------------------------------------- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php _______________________________________________ Gtk-gnutella-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/gtk-gnutella-devel
