Re: [Uta] smtp-sts-04 JSON

[Daniel Margolis <[e-mail]>]

In the interests of not looking like I'm ignoring your mail, just to
reiterate, I have no personal feeling about this. My read at the meeting in
Chicago was that consensus was not totally obvious, but I'm really happy to
defer to the chairs here on what the consensus seems to be or how to
unblock this.

As I have said before, while this is certainly an issue people have
legitimately strong feelings about, it's not really a *fundamental* issue,
and, speaking personally (and I think for the other authors), I would be
happy with any resolution in any direction--the main point here is to
provide some additional security to email-in-transit, not to provide
security to email-in-transit by leveraging JSON. ;)

Dan

On Sun, Apr 23, 2017 at 4:58 AM, Viktor Dukhovni <[e-mail]>
wrote:

>
> > On Apr 22, 2017, at 1:35 PM, Daniel Margolis <[e-mail]>
> wrote:
> >
> > That does not jive with relatively easy to parse...
> >
> > JSON supports comments, elements that are integers, strings,
> > arrays or associative arrays (nested JSON objects).  JSON
> > strings are UTF-8 and allow embedded NUL octets.
> >
> > Your JSON reference is to the obsolete RFC4627, the non-obsolete
> > reference is RFC7159.
>
> So any compelling arguments for JSON from MTA implementors?
>
> A much simpler format would be a block of lines (CRLF terminated),
> with each line either a boolean propery name, or a name value pair.
>
>         version STSv1
>         mode report
>         max_age 8640000
>         mx mx1.example.com
>         mx mx2.example.com
>         ...
>
> Line length is limited to the usual 1000 bytes including the
> terminating CRLF.
>
> This simplifies parsing, because a simple line-orienter parser
> can now process the entire policy.  It is no longer possible
> to have multiple policies, as with JSON, because there is no
> way to express multiple objects.  The JSON format is much
> too general for the needs at hand.
>
> --
>         Viktor.
>
> _______________________________________________
> Uta mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/uta
>