[salud] Reasonable rendering

worley@ariadne.com (Dale R. Worley) Thu, 17 April 2014 20:58 UTC

Date: Thu, 17 Apr 2014 16:57:55 -0400
Message-Id: <201404172057.s3HKvtvI012125@hobgoblin.ariadne.com>
From: worley@ariadne.com
Sender: worley@ariadne.com
To: salud@ietf.org
Archived-At: http://mailarchive.ietf.org/arch/msg/salud/qtTWLFk0VVRmHtTtTPaNkhus6aw
Subject: [salud] Reasonable rendering
Precedence: list

[as an individual]

Richard Barnes (our new AD) says:

> Section 13 says that the UA MUST provide a "reasonable rendering".   How do
> I tell if a UA meets that criterion?  Might be better to add in Section 11
> a requirement of the form, "if a UA cannot process the set of Alert URNs in
> a message, then it MUST still provide some alert (chosen by local policy).
>  Failure to parse Alert URNs MUST NOT cause a UA not to alert at a
> all."

I was thinking that it might be desirable to add some text to cover
Richard's issue.  On one hand, it seems that "reasonable rendering"
should automatically include "the user can perceive it", but it might
be useful to be explicit in regard to this particular feature of
signals.

One way to do this would be to extend the last paragraph of section
13, which reads

   The User Agent (UA) MUST produce a reasonable rendering regardless of
   the combination of URIs (of any schemes) in the Alert-Info header
   field.

by adding

   In particular, under any circumstances a UA MUST provide some alert
   unless it is explicitly instructed not to (by Alert-Info URIs that
   it understands, local policy, or direction of the user).

Dale

[salud] Reasonable rendering Dale R. Worley