Re: [Dime] [SPAM] RE: rfc3588bis version number

["Glen Zorn" <gwz at net-zen.net>]

Sebastien Decugis [mailto:sdecugis at nict.go.jp] writes:

> > Are you saying that "SHOULD" and "MAY" are equivalent, then?
> >
> 
> No, I mean that the implementation can do A or B and claim to be
> conformant; i.e. the specification does not restrict the possibilities.
> In our case, a new implementation would do B since it's a SHOULD, but
> an
> existing implementation that already do A will not need to be modified
> to claim conformance.

I'm still not seeing the problem: that seems to me to be pretty much the
definition of "backward compatibility" ;-).  As long as both ways work, it
shouldn't matter WRT interoperability & hopefully, legacy implementations
would upgrade over time; in any case, v2 could deprecate the old behavior.

> 
> Hence my conclusion...
> 
> Best regards,
> Sebastien.
> 
> --
> Sebastien Decugis
> Research fellow
> Network Architecture Group
> NICT (nict.go.jp)
>