Re: [tcpm] tcpsecure: how strong to recommend?

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mark Allman wrote:
> Hat off ... 
> 
> This is a lot of hang wringing mumbo-jumbo, I think.  It seems to me ...
> 
>   + If a standards track document says "a TCP SHOULD do X" then as a
>     particular TCP evolves it really ought to do X unless the
>     implementers have a very good reason not to do X.
> 
>   + If a standards track document says "a TCP MAY do X" then as a
>     particular TCP evolves the implementers have discretion to do X or
>     not to do X based on their understanding of the issues.
> 
>   + If a standards track document says "a TCP SHOULD do X subject to
>     conditions Y or scope Z" then as a particular TCP implementation
>     evolves it really ought to do X if it falls within the given
>     conditions or scope.
> 
> Why is this harder than that?

The other nuance is:

	if you do A you MUST do B

i.e., dependencies of the choices within the protocol. That's useful to
capture, even using MUST/MAY/SHOULDs, but it should be clear enough from
the textual context.

> Hat on ...
> 
> It seems that currently the first two bullets apply to tcpsecure.  That
> is, the document does not have any scoping or conditions.  Lars (and
> others) has suggested that perhaps we need to add some scoping to the
> document through an applicability statement.  That is certainly an
> option.  If folks think that we should be adding such an applicability
> statement before worrying about SHOULD vs. MAY then please say so.  (It
> seems to me that thus far that has been a minority position.  I am happy
> to change my view of this based WG input, naturally.)

I think that MAY is fine without conditions.

I think that SHOULD should always have conditions - even if to
explicitly state "all TCPs SHOULD" in the most general case.

Joe

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFG+WK6E5f5cImnZrsRAiuvAKD2eS2cYG2WdY18FUzPUWLxq+KMngCdFXqy
OFp/CYbShaFEFJOqiE0c0eI=
=+FfC
-----END PGP SIGNATURE-----

_______________________________________________
tcpm mailing list
tcpm@ietf.org
https://www1.ietf.org/mailman/listinfo/tcpm