RE: REPORTS

"John Hall" <johnhall@xythos.com> wrote:

> -----Original Message-----
> From: John Hall [mailto:johnhall@evergo.net]
> Sent: Thursday, September 06, 2001 9:58 AM
> To: 'Tim Ellison'; 'DeltaV'
> Subject: RE: REPORTS
>
>
> >
> > There were no objections raised in the meeting or when the
> > minutes were published.  I'd consider that a consensus.
>
>
> It was the equivalent of slipping language in a bill at
> reconciliation time, without debate, after midnight when
> everyone else had gone home, then rushing the bill out for
> signature before anyone noticed.

Hey, if you can run a country that way, I'm sure we can write a spec like
that<g>

> MAY is more appropriate than SHOULD given the late date at
> which the change was made, the manner of the change, and the
> fact that a commercial implementation of the spec will not be
> implementing this report and will recommend that clients who
> wish to be interoperable avoid it as well.
>
> And if there is no difference between MAY and SHOULD, then
> there should be no objection to making it MAY.

Let's agree on what it _ought_ to say rather than on the process that got
us to this point or what you will recommend.

Given that the report will be useful in a number of interesting scenarios I
think it is a useful optimization that servers SHOULD implement.  Note that
clients are free not use use it, and servers are free not to implement it
and both will be within the spec.  But servers should be encouraged to
implement it and clients should be encouraged to look for it and use it.

Regards,
Tim

Received on Thursday, 6 September 2001 16:04:22 UTC