> This may be. (I'm not sure, myself.) However, if the exposure of
> job ids is so critical, then why don't we also insist on:
>
> IPP printer name == LPD printer name
I have assumed that IPP printer-name == LPD printer name, but
that IPP printer-name differs from IPP printer-URI. That is
IPP printer-name might be "killtree" and IPP printer-URI might be
"http://foobar.bigcorp.com/printer/killtree"
> If this is good enough, then why can't the same be true for job
> ids, eg:
>
> http://somewhere.net.domain/subsys/server3/345 == 345
The translation between printer-name (LPD or IPP) and printer-URI is
easy because of the two IPP attributes and because a name service is
likely to have both.
On the other hand, jobs won't be in the name service because they are
transient, and the syntax of a job-URI is opaque so there is no way
to determine a Job-Id from a Job-URI.