IPP Mail Archive: RE: IPP> Re: Implications of introducing new scheme and port for existing HTTP servers

RE: IPP> Re: Implications of introducing new scheme and port for existing HTTP servers

Larry Masinter (masinter@parc.xerox.com)
Sat, 6 Jun 1998 00:04:32 PDT

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Larry Masinter: "RE: RE: IPP> Implications of a new scheme, etc"
Previous message: Robert Herriot: "RE: RE: IPP> Identifying jobs in requests"
Maybe in reply to: Ted Hardie: "IPP> Re: Implications of introducing new scheme and port for existing HTTP servers"

In HTTP/1.1, you don't normally send the URL to the server anyway,
you only send the path. So for the most part, the HTTP servers don't
need to know anything about the URL but the path component of it.
The 'transformation' between ipp://xyz.printer.com/printer1
and http://xyz.printer.com:380/printer1 happens in the client
(since the client knows about the rest of IPP too, and a client that
doesn't know about ipp has nothing to say to the URL anyway), and
possibly in the bowels of the IPP server which is generating real
URLs for things like jobs and other printers. But the HTTP infrastructure
of proxies, HTTP protocol stacks, etc. need not know anything about
the URLs at all, I don't think.

Larry

--
http://www.parc.xerox.com/masinter

Next message: Larry Masinter: "RE: RE: IPP> Implications of a new scheme, etc"
Previous message: Robert Herriot: "RE: RE: IPP> Identifying jobs in requests"
Maybe in reply to: Ted Hardie: "IPP> Re: Implications of introducing new scheme and port for existing HTTP servers"