spf-discuss
[Top] [All Lists]

Re: BTFOOM

2005-05-20 01:56:27
wayne wrote:

I-D issues sent to the council for rulings are ones that I
think reasonable people can disagree on and often there is
*no* good choice.

Sure, Julian's idea makes sense, but we have no time travel to
get it into draft-mengwong-spf-00

The domain literal "FAIL malformed domain" problem and the now
proverbial "SPF is no receiver policy" still ring into my ears.

council members basing their votes on their technical
understanding is more important than basing votes on their
understanding of the SPF community.

I'm almost sure that you all did this, in different ways, Chuck
with the good instinct of "if it ain't broken" + "never change
a winning team".  Mark went in the old "FAIL malformed domain"
"receiver policy" trap.

PermError => policy FUBAR is also not exactly new, it's one of
the corner stones of your "validating implementation" concept.

After too many broken policies, a somewhat suboptimal wizard,
and many hard fights with James, MarkL, and Meng about it here.

[23:44]
<grumpy> "sign", however, is a bad name

Still the best idea after "prefix" was sent to lala-land.
 
And what was that idea?

What you said 23:44, Julian's original proposal "sign".
 
Maybe "indicator"?

Better than "mode".  As you say it's not really important, I'm
much more interested _where_ Julian found this general design
rule "name follows function, not form".  I'd love to use this
as an argument for "id-domain" instead of "id-right" in USEFOR.

                               Bye, Frank



<Prev in Thread] Current Thread [Next in Thread>