Re: [dispatch] Zstandard compression

[Sean Leonard <[e-mail]>]

> On Nov 13, 2017, at 11:22 AM, Murray S. Kucherawy <[e-mail]> wrote:
> 
> On Mon, Nov 13, 2017 at 10:22 AM, Sean Leonard <[e-mail] <mailto:[e-mail]>> wrote:
> Those documents do not specify MIME types, and I do not think this document draft-kucherawy-dispatch-zstd should do so either. The media type registration is weak: there is no way to signal inside the format what is inside other than an arbitrary stream of bytes. It is not a file archive format, or another kind of container format that identifies parts.
> 
> Why does that imply that a media type is not appropriate?  A module receiving this content for processing needs to know what format it's in, and I believe that's what these are for.

Working on this. It strikes me as either…odd…or just not really that useful for the purpose. It is a format, but it’s just a format for a stream of arbitrary bytes. There is no way to label it internally beyond the implicit “application/octet-stream” type.

Content-Encoding is therefore very appropriate, but that is for HTTP, not (as of yet) for mail or other messaging formats.

May I ask, what is the reason why you need a media type?

Let me think about it…maybe it would be good to ask the media-types ML.


> 
> I request that we get an OID registration for this for use with Compressed Data Content Type for CMS, RFC 3274. That way it could be used with S/MIME or other formats that use CMS or S/MIME (seeing as how CMS / S/MIME are general container formats).
> 
> Wilco.
> 
> Surveying other container formats and doing the appropriate registrations with them in this draft would be appropriate, if one can articulate a use case in those respective formats.
> 
> Can you suggest an example?

Not offhand, just thought of those two (Content-Encoding, and RFC 3274 CompressedData media type).

Well there’s TLS v1.2 compression methods. Just joking! :-)

Sean

> 
> -MSK