> On 7. Jul 2026, at 16:46, Michael Richardson <[email protected]> wrote:
> 
> The question for me, becomes: would a new implemention survive real customers
> if it didn't have that extension?   When it comes to doing the base-RFCbis,
> would you want to roll them all up into the new base RFC?

Based on this criteria using a previous set-to-zero flag should not be “amends” 
but “extends” as the ignore on receive rule is explicitly design to enable 
future use. The point is such an RFC basically say, whoever, implements this 
RFC MUST set the flag to zero; however, it doesn’t say that it can never be 
used in a different in future. So effectively you don’t ”change” the existing 
RFC but only define something “on top” of it. Again there is something to 
clarify here in the draft but these flags are designed explicitly as extension 
points. I also don’t think such an extensions would e.g be written in an 
OLD/NEW style.

However, there could be a case where the initial protocols doesn’t work because 
an important flag/function is missing. If the flag is changed for that reason, 
then it would be “amends”. 

-- 
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to