Re: [openpgp] "OpenPGP Simple"

[Christoph Anton Mitterer <[e-mail]>]

On Mon, 2015-03-16 at 23:48 -0700, Jon Callas wrote:
> ASCII armor ends up being a nice way to encode something so you don’t
> have to play "guess the binary format."
But isn't that basically the same case as with the character encoding
thread?! In the sense of:

It's not OpenPGPs task to fix other applications/protocols!

If these need to use OpenPGP messages and they don't support binary
formats (which by itself is fine), then their own native means for
converting blobs into a suitable form should be used (e.g. XML's or
whichever) and if no such form exist, we should at best give a
recommendation about what to use (in the sense of referring to something
else).

Especially I don't like the idea that applications out there may use the
ascii armor's enclosing (i.e. the ---- PGP ... ----) to detect "this is
a PGP message".
There should be other things around in the protocol that uses OpenPGP
which specifies that (e.g. Content-Type: headers).


I think it would be theoretically[0] fine if a future OpenPGP standard
specifies multiple native formats, e.g. a binary one as we have right
now, a XML representation and whatever other ugly things (JSON?! ;-) )
people might want... but the ASCII armor isn't a real native format but
just an encoding, and that shouldn't be OpenPGPs duty.


Cheers,
Chris.

[0] In practise, though, there should be really strong reasons if more
than one such native format would be defined.