[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
- To: xanadu@xxxxxxxxxxxxx
- Subject: RE: "xurl?
- From: Steve James <sjames@xxxxxxxxxxxxxxx>
- Date: Tue, 14 Nov 1995 22:52:46 -0500
- Encoding: 55 TEXT
- Reply-to: xanadu@xxxxxxxxxxxxx
Sent: Sunday, November 12, 1995 5:15 AM
Cc: marlene@xxxxxxxxxx; reg@xxxxxxxxxxxxxxxx; avatar@xxxxxxxxxxxxxx;
XANABULLETIN, to the new Xanadu community
- - - - -
One of the policy questions right now is:
At what level should there be a protocol difference for
But two of my collaborators, Roger Gregory and Sam Epstein,
think that transcopyright protocol should be recognizably
different at the top level, eg
To put my $0.02 in, This seems reasonable. The nature of transcopyrighted
material is likely to require special handling on the part of both client
and server for such functions as electronic signature, quoting of
transcopyrighted materials through some sort of link, and appropriate
billing. Why not let a designation like xu:// be used to denote a
specialised extension of http?
Since we probably don't want to double the number of MIME types for this
purpose, especially since the billability or need for authentication is not
really intrinsic to the type of material, it makes sense to make the
distinction in the transport layer.
As long as xu:// is the name of an extended http protocol rather than some
sort of re-invented wheel, the designation will not be a roadblock to
shareing the advances of http or, for that matter, for http absorbing some
or all of xu's extensions in later versions.
Hopefully, future Xanadu browser software will be well designed enough to
be a fully capable http browser as well and will accept either sort of URL.
In that case, there is also the advantage that legacy materials and
conventional web servers could include references to transcopyrighted
materials just by using the appropriate transport designation. Presumably,
the browser could be configured to remind the user that there may be a
charge for that link.