Re: [clue] Ben Campbell's Discuss on draft-ietf-clue-data-model-schema-15: (with DISCUSS and COMMENT)

["Ben Campbell" <[e-mail]>]

On 1 Jun 2016, at 14:43, Alexey Melnikov wrote:

> Hi Ben,
>
>> On 1 Jun 2016, at 18:46, Ben Campbell <[e-mail]> wrote:
>>
>>> On 1 Jun 2016, at 13:37, Alexey Melnikov wrote:
>>>
>>> Hi Ben,
>>>
>>>> On Wed, Jun 1, 2016, at 06:24 PM, Ben Campbell wrote:
>>>> Also inline,
>>>>
>>>>> On 1 Jun 2016, at 12:41, Roni Even wrote:
>>>>>
>>>>> Inline
>>>>> Roni
>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Ben Campbell [mailto:[e-mail]]
>>>>>> Sent: Wednesday, June 01, 2016 5:23 PM
>>>>>> To: The IESG
>>>>>> Cc: [e-mail]; 
>>>>>> [e-mail];
>>>>>> [e-mail]; [e-mail]
>>>>>> Subject: Ben Campbell's Discuss on
>>>>>> draft-ietf-clue-data-model-schema-15:
>>>>>> (with DISCUSS and COMMENT)
>>>
>>> [snip]
>>>
>>>>>> ----------------------------------------------------------------------
>>>>>> DISCUSS:
>>>>>> ----------------------------------------------------------------------
>>>>>>
>>>>>> - 11.2: I would like to discuss whether it's a good idea to allow
>>>>>> arbitrary
>>>>>> values for mediaType, beyond those types registered in IANA.
>>>>>> The text seem to encourage proprietary values. Did the working 
>>>>>> group
>>>>>> consider requiring IANA registration of some sort for new values?
>>>>>
>>>>>
>>>>> [Roni Even] ] I think that the meaning was to allow the media 
>>>>> types
>>>>> from the IANA registry
>>>>> http://www.iana.org/assignments/media-types/media-types.xhtml . 
>>>>> The
>>>>> issue will be about the usage of other media types in the CLUE
>>>>> environment and not about the values themselves.
>>>>> Alissa also suggested that we clarify it
>>>>
>>>> I'm not sure I understand your response. When you say "other media
>>>> types", do you mean other types that are registered in the 
>>>> media-type
>>>> registry, but not described in RFC6838?
>>>
>>> In case it helps: top level media types like "model", "application",
>>> "message", and not full media types like "application/pdf". These 
>>> should
>>> be very rare. JUSTFONT WG is working on another one, but this is the
>>> first time something was added in more than 10 years. I honestly 
>>> doubt
>>> anything other than "video", "audio" or "text" would be usable for 
>>> CLUE
>>> anyway.
>>
>> That was my assumption. That's part of why I wondered about the 
>> ability to ability to invent arbitrary values. (But note that I've 
>> already cleared my DISCUSS.)
>
> My understanding was that each new media type would have to be 
> registered in the media type registry first, which is a rather high 
> bar anyway. But maybe that is not what the WG had in mind.

I didn't get that from the text in version 14 or 15. But I might have 
read it wrong. (It happens.)

>
> I think it is more likely that "application" would be used, rather 
> than an entirely new media type.

I think you are likely correct.

>
>>>> Or other arbitary media types
>>>> that are not registered or otherwise publicly documented?
>>>
>>> I would be impressed if any undocumented or unregistered top level 
>>> media
>>> type would be interoperable, so I don't think any such media types
>>> matter.
>>
>> Does that mean you don't think people will use undocumented 
>> media-types due to the lack of interoperability, or that anyone who 
>> does will be effectively proprietary and we should assume they don't 
>> want to interoperate in the first place?
>
> Both :-).
>
> Experience with media types in other places shows that new media types 
> are hard to deploy.
>
>> Thanks!
>>
>> Ben.
>>