Re: [Json] Media types, extensibility in draft-ietf-json-i-json-02

[Mark Baker <[e-mail]>]

On Mon, Jul 7, 2014 at 10:40 AM, Markus Lanthaler
<[e-mail]> wrote:
> On Sunday, July 06, 2014 7:13 AM, Tim Bray wrote:
>> So, question: Is there anyone else who agrees with Mark's original
>> argument that there ought to be a Media type for i-json. I wouldn't
>> be wildly against it, but frankly can't see much of a practical upside.
>
> I don't think it makes much sense to define a separate media type for i-json and even less a media type suffix.

Whether a I-JSON media type is required or not depends entirely on the
use cases for I-JSON's existence. The only use case I've seen, from
Joe, suggests a new media type would have some utility. But then
others reject that use case.

When I hear "a separate media type isn't needed", to me that means
that I-JSON is to be used - and published - as a strict subset of
JSON, aka it will not be part of the protocol spoken by Web clients
and servers. One implication is that an I-JSON-unaware
client/intermediary could consume some I-JSON, and e.g. add content
with a duplicate key, then process and/or republish as non-I-JSON
JSON.

If the group doesn't consider that a problem, then great, you don't
need a new media type. If it does, then you do.

I hope it's clear that I'm not arguing for or against a media type any
more. I'm just trying to tease out of the group the criteria that
would permit a principled decision to be made.

> It *might*, however, make sense to mint a profile URI as defined by RFC6906 [1] for i-json and register it according RFC7284 [2].

Don't go there. Those kinds of profiles have never worked in their
many, many guises.

Mark.