RE: RFC 2534 list of ua-media and paper-size values

Are you suggesting that the "print-media" attribute is too narrow and that
it implies some more general concepts, such as "media-size".

Perhaps, as Larry suggested, there should be a "media-size" which applies to
some of the media, e.g. "printer" and "screen", but not to "aural".  For
printers there could be additional "media" attributes, such as "color" and
"opacity". If we adopt this solution, the IPP "media" values are names for
particular "media-size", "color", "opacity", etc.

-----Original Message-----
From: Graham Klyne [mailto:[email protected]]
Sent: Wednesday, October 20, 1999 2:59 AM
To: Herriot, Robert
Cc: Conneg WG; Manros, Carl-Uno B; Johan Hjelm; [email protected];
[email protected]
Subject: RE: RFC 2534 list of ua-media and paper-size values


At 18:31 19/10/99 -0700, Herriot, Robert wrote:
>It doesn't seem reasonable to merge the values of HTML "media" and IPP
>"media". The former describes media in a large sense. The latter desribes
>media in the context of a printer.
>
>I agree that all of the print related values of RFC 2534 should be
>deprecated and that the word "print" should be used instead as a value of
>"media"  It would be reasonable to add a new attribute "print-media" which
>contains the values of the IPP "media" attribute.

Robert,

This would seem a reasonable approach if one takes a view that it is always
known a priori in what kind of environment a resource will be used.  But I
do not share that view.

The original CONNEG work was motivated in part by a desire to have a way to
express common media features that could used across a range of transfer
protocol environments.  This in turn, I belive, creates an environment in
which it becomes easier to build integrated services that use a variety of
protocols;  e.g. web servers for e-mail collection, IPP for quality
document transfer, e-mail for fax transmission, etc.

The model that we have evolved for Internet fax, and which I believe could
usefully be employed in other scenarios, is to use common vocabulary for
features that are common across a variety of service environments (e.g. per
RFC 2534), and to develop a more specialized vocabulary for features that
more more specific to the fax environment.  This approach is realized by
having the generic feature tags be very broadly defined, and using
specialized feature tags to refine them (in the sense of constraining
rather than revising).

In summary, I would suggest resisting a divergence of vocabulary for
features that are common across a range of services.

#g

------------
Graham Klyne
([email protected])