Re: [Json] Proposed rechartering for the JSON WG

On Sun, Feb 9, 2014 at 8:32 AM, Carsten Bormann <cabo@tzi.org> wrote:

> Having a minimal parseable description technique for JSON-based protocols
> would save us from tedium such as in RFC 7071 section 6.2.2.
> (Not needing something like that ever again*) in the future is my personal
> benchmark for what we need to achieve here.)
>

Hm, I just went and looked at
http://tools.ietf.org/html/rfc7071#section-6.2.2 - it reads like a
perfectly sensible piece of specification-ware to me. Looks to me like it
leaves implementors very little uncertainty, and it’s easy to understand.
 It all fits on a couple of screens on my laptop. The dorky ALL_CAPS makes
it typographically ugly, but it’s hard to imagine a formalism that would
make it dramatically shorter, or more useful.

I actually thought what we were talking about was something to replace
6.2.1 in that spec, so people don’t ever need to write an “Imported JSON
terms” section again.

>
> The important thing**) here is the 80-20 rule — we don’t need to make this
> thing powerful or perfect, because we can always fall back to English.
> But we shouldn’t always have to, even for the most basic purposes.
>
> (draft-newton looks like an 85 % solution to me; maybe this can be even
> simpler.)
>
> Grüße, Carsten
>
> *) Yes, I know why it was done, and I agree that it had to be done at the
> time.
>
> **) The other important thing is that a JSON description technique doesn’t
> have to be parse as JSON, just as an XML schema does not have to be parse
> as XML.
> Why are people falling into this trap again and again...
>
> _______________________________________________
> json mailing list
> json@ietf.org
> https://www.ietf.org/mailman/listinfo/json
>