[openpgp] Re: Fwd: New Version Notification for draft-gall agher-openpgp-media-types-00.txt

Daniel Kahn Gillmor <[email protected]> Mon, 17 Nov 2025 15:59:11 -0500

Newsgroups	gmane.ietf.openpgp
Message-ID	<[email protected]>

On Wed 2025-11-05 20:05:06 +0100, Simon Josefsson wrote:
> Yes, probably, but I think breaking RFC 3156 media types ought to be
> avoided (if possible),

With no hats on, i think i agree with Simon here -- when we start
parameterizing, we create combinatorial complexity.  Combinatorics are
great for extensibility, but we're not really looking for extensibility.
we want a fixed name for a thing.

The fact that some OpenPGP tooling can ingest an armored object the same
way as a non-armored object is cute: application code that knows its
OpenPGP tooling has that capacity can check the non-parameterized name
and ignore the parameters in that case.  But it isn't much harder for
application code with such a dependency to test against a list of two
variants, either.

And it discourages clever folks from coming along and adding more
parameters that might have surprising combinatorial effects.

> so I'm hoping these media types will be a strict superset of the old
> ones.

I'm not entirely sure what you mean by "strict superset" here.  It could
mean "This new media type X subsumes media type Y -- if you were to use
X in the past, we recommend you use Y; also, include these new kinds of
documents".  Or it could mean "here is a list of media types, and the
list contains old media types (including Y) and new ones (like X); each
document under discussion can be mapped to exactly one of them"

Anyway, keep it simple!

        --dkg

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

signature.asc (application/pgp-signature, 324 B)