RE: [idn] IDN WG Last Call: draft-ietf-idn-punycode-00.txt

["Hollenbeck, Scott" <shollenbeck@verisign.com>]

> -----Original Message-----
> From: Adam M. Costello [mailto:idn.amc+0@nicemice.net.RemoveThisWord]
> Sent: Tuesday, January 29, 2002 11:31 PM
> To: idn@ops.ietf.org
> Subject: Re: [idn] IDN WG Last Call: draft-ietf-idn-punycode-00.txt
> 
> 
> "Hollenbeck, Scott" <shollenbeck@verisign.com> wrote:
> 
> > An editorial comment: would it be possible to capitalize 
> instances of
> > the RFC 2119 directives "must", "may", "should", etc.?
> 
> It would certainly be *possible*.  :)
> 
> > Doing so will make it far easier for implementers to not miss
> > something that might come back to bite them later.
> 
> I remain skeptical of that claim.  I personally find the all-caps MUST
> SHOULD etc. to be distracting and not at all helpful.
> 
> If the group demands this then I'll write an awk script to take care of
> it, but I'd rather not.

It would be impossible to prove my assertion of implementer helpfulness, so
let me try another angle instead: consistency.  All of the other documents
in last call use the upper case form of the directives, they're used that
way in 2119 (though 2119 admittedly doesn't require upper case), and upper
case appears to be more common in published RFCs.  In my experience it's
what people are used to seeing.

Like I said, though, it's an editorial comment.  No big deal.

-Scott-