Re: [Extra] IMAP RFC 6855 (IMAP EAI) implementation experience: input to IMAP4rev2

[Arnt Gulbrandsen <[e-mail]>]

Chris Newman wrote:
> It's interesting that my implementer hat and standards hat 
> result in diametrically opposed positions on this issue, but so 
> be it.

I think this is because RFCs are obliged to respect other RFCs more than 
implementers are. You, as implementer, may reasonably look at what hurts 
most in practice in your part of the world; the set of RFCs ought to be 
consistent if at all possible.

BTW, this is why RFC 6858 exists in the first place. Enabling reading using 
6857 is IMO too much work compared to just-send-8, which is IMO tantamount 
to encouraging violating 3501's rules. The simplest possible conformant 
implementation shouldn't be hard.

You should find 6858 easy to implement well enough that mail can be read 
without violating any rule in 3501. If not, I failed. Almost as an aside, 
doing so should minimise the risk of provoking a bug related to sending 
mail, by making replying and composing fail in a safe/predictable manner. 
But not implementing it is IMO also entirely reasonable, even if an RFC 
should not say so.

Arnt