Date: Tue, 5 Aug 1997 16:45:41 +0200 (MET DST) From: "Martin J. Duerst" <firstname.lastname@example.org> To: Stephanos Piperoglou <email@example.com> cc: "E. Stephen Mack" <firstname.lastname@example.org>, email@example.com In-Reply-To: <Pine.LNX.3.95.970805173315.171Cfirstname.lastname@example.org> Message-ID: <Pine.SUN.3.96.970805164236.9086O-100000@enoshima> Subject: Re: Extending URL syntax for framesets? On Tue, 5 Aug 1997, Stephanos Piperoglou wrote: > On Tue, 29 Jul 1997, Martin J. Duerst wrote: > > > I think if anything, you better overload "#" than "?". > > The part after "?" is sent to the server, the part after "#" is > > not. In your case, it's the browser that has to disentangle > > the additional information and load the various frames. > > > > An additional advantage is that the syntaxt after "#" is not > > very abused yet, so that it is easier to introduce new conventions. > > This would be very dificult to implement because html files in the various > frames may also include fragment identifiers (i.e. "load `cow.html' in frame > `main' and scroll down to anchor `physiology'") Well, things can of course go wrong. But there are several solutions. The first is to say: If the part after the "#" contains ";" or "=", then it is for a frame, otherwise it is a simple anchor. Another trick is to say: Check if you find an anchor of that name in the target document, and if not, try to use it for frames. It is very unprobable that an author uses the same tag both for a frame target and for a plain anchor. Regards, Martin.