Re: [apps-discuss] Review of draft-levine-application-gzip-01.txt

[John C Klensin <[e-mail]>]

--On Friday, April 13, 2012 17:22 -0700 Ned Freed
<[e-mail]> wrote:

>> Hi John,
>> I am sorry I've missed this document earlier.
> 
>> I am generally happy that you wrote this draft and support its
>> publication. Having said that, I think it needs some changes:
> 
>> In Section 1:
> 
>>     Some applications have informally used the media type
>>     application/ x-gzip.  The media types defined in this
>>     document should replace that media type in future
>>     applications.
> 
>> Also I've seen application/x-gzip-compressed being mentioned
>> on the web.
> 
> And application/gzip-compressed, application/gzipped,
> application/x-gunzip, and
> probably a bunch of others.
> 
>> But I think it would be better to wait for
>> draft-ietf-appsawg-media-type-regs-04 to be submitted to IESG
>> (it is in the APPSAWG WGLC now). That document would relax
>> restrictions on registering x-* media types. I think your
>> document should register media types already in use, as
>> changing existing implementations to use application/gzip is
>> hopeless at this point, unless there is some evidence that
>> application/gzip is also in use. In the latter case aliasing
>> mechanism proposed in draft-ietf-appsawg-media-type-regs
>> should be used.
> 
> I'm ambivalent about this in this case. There are simply too
> many names for
> this media type in use; no matter which one we pick is cannot
> possibly conform
> to all existing usage.

And, of course, there is the separate issue that caused us to
reject various "zip" and related types years ago -- it is
basically a content-transfer-encoding (even if it also used
locally) and not a content/media type.

As with other things, the "horse have left the barn" principle
may argue for registering it anyway, but we (and I believe any
spec) should be aware that it breaks the model.

    john