RE: IPP> Re: IPP Scheme

"Bennett, Joel H" <Joel.Bennett@usa.xerox.com> Wed, 15 July 1998 12:24 UTC

Delivery-Date: Wed, 15 Jul 1998 08:24:36 -0400
Content-return: allowed
Date: Wed, 15 Jul 1998 05:09:57 PDT
From: "Bennett, Joel H" <Joel.Bennett@usa.xerox.com>
Subject: RE: IPP> Re: IPP Scheme
To: don@lexmark.com
Cc: Ipp@pwg.org
Message-id: <C565EF2D2B51D111B0BD00805F0D7A72727DBF@USA0111MS1>
MIME-version: 1.0
Content-type: text/plain
Sender: owner-ipp@pwg.org

So I suppose, if I follow this argument, that the problem with HTTP: is that
it hides the fact that this is really TCP/IP? And perhaps, the problem with
ftp: and mailto: URL's as well?  why to we even need one at all? why not
just have them all be "/www.xerox.com/mypage.html" and
"/printers.xerox.com/myprinter" and "/ftp.xerox.com/myfilestorage"  ;-)

>-----Original Message-----
>       
>
>> >The problem with using http: for printers is that it hides the fact
>> >that the resource is a printer.
>>
>> ... and the problem with "ipp:" is that it hides the fact that
>> the protocol is really HTTP!!!
>
>I guess I consider it more important for the URL to describe the
>end resource that it's providing, than for it to describe the
>underlying protocol(s).  We don't have tcp: or ip: URLs; we have
>URLs for higher level services.  Some URL schemes don't imply a
>particular protocol.  However, for every URL scheme that uses the
>name of a protocol, the URL names the highest layer protocol
>on the stack, rather than some lower layer.
>
>Keith

As always, in HIS hands,
                       Joel H. Bennett
mail <mailto:Joel@soon.com>
web <http://JBennett.home.ml.org>
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
QOTD:
All charming people, I fancy, are spoiled. It is the secret of their
attraction.,
-- Oscar Wilde: 1854-1900, Irish writer