I think that there is a difference between meta-level things like charset
and language and real attributes like job-id.
I think we should pull all URIs from the IPP model. The only place they
should appear in in the (non-existant) 'IPP on HTTP implementation rules'.
This is the only way we can make transport independance work
> -----Original Message-----
> From: Robert Herriot [SMTP:robert.herriot at Eng.Sun.COM]
> Sent: Monday, June 08, 1998 1:24 PM
> To: Paul Moore; 'Robert Herriot'; 'Carl Kugler'; ipp at pwg.org> Subject: RE: RE: IPP> Identifying jobs in requests
>> Charset clearly needs to be at the beginning so that a receiving process
> knows how to decode string fields before it encounters any of them. The
> charset field is a string field, but the names are all ASCII so the
> decoding
> need not be known for the class of encodings that IPP support. BTW, XML
> puts
> charset at the beginning for the same reason.
>> Natural language is the next most important value because it specifies the
>> language of any text or name field. Processing is easier if the implicit
> language is known at the time a text or name field is encountered. XML has
>> similar rules for language.
>> The printer-uri/job-uri/job-id should be easy to find if it not in the
> enclosing layer. For HTTP, the position of the target isn't important
> because the target is redundant. The position would be important for a
> transport where the target is specified only in IPP layer.
>> So that's the reasoning. Do you agree?
>> Bob Herriot
>> At 11:42 AM 6/8/98 , Paul Moore wrote:
> >I obviously missed this one. So 'special position' means that literally.
> I
> >thought it mean 'special purpose'.
> >
> >For my interest. Why are we putting things in special positions?
> >
> >> -----Original Message-----
> >> From: Robert Herriot [SMTP:robert.herriot at Eng.Sun.COM]
> >> Sent: Friday, June 05, 1998 8:46 PM
> >> To: Paul Moore; 'Carl Kugler'; ipp at pwg.org> >> Subject: RE: RE: IPP> Identifying jobs in requests
> >>
> >> We agreed recently that the following operation attributes would be
> >> ordered.
> >> attributes-charset (always first for requests and responses)
> >> attributes-natural-language (always second for requests and
> response)
> >> printer-uri or job-uri (always third for requests, though we are
> >> discusses whether it should be present)
> >> job-id (always fourth for requests if present )
> >>
> >> At 05:00 PM 6/5/98 , Paul Moore wrote:
> >> > I think we are approaching group consensus on this. I propose
> >> that
> >> >we remove "printer-uri" and "job-uri" as request Operation attributes,
> >> but
> >> >leave them in their special position in the protocol.
> >> >
> >> > [Paul Moore] What 'special position'?
> >> >
> >>
> >
>