Quoting Jeroen Asselman <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel: :On Wed, 2004-02-25 at 18:39, Raphael Manfredi wrote: :> Quoting Jeroen Asselman <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel: :> :How are we going to confince bearshare which allready uses the :> :X-Features header for about 4 month? :> :Of course we could parse both and only emit one. But shouldn't we get :> :the discussion started on the gdf? :> :> You mean they already send X-Features: along in *gnutella* handshakes? : :Yes, I am sorry if I wasn't clear about this yesterday.
Then we have a problem and I guess it'll be difficult to change it. LimeWire already showed their misunderstanding of X- with the X-Ultrapeer and X-Try header family. Now BearShare is doing it again with X-Features. I guess people forget the origin and the meaning of X- and just take it as "my own private header I'm adding here and which is not standard". It's wrong on both counts because the header is public (specs) and may become standard (X-Ultrapeer and such). No, X- stands for "extension header", from a well-defined published spec (like SMTP, HTTP, etc...). Here, since we write the specs as we go along, the use of X- is not justified. It's even counter-productive, because a casual observer might think those headers are not part of the official specifications and are only private ones. Raphael ------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ Gtk-gnutella-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/gtk-gnutella-devel
