proposed media type: application/exi

Martin Duerst duerst at
Fri Oct 17 08:57:24 CEST 2008

Hello Mark,

It seems you are saying:
1) application/xml (and text/xml) was a bad ide
2) the +xml suffix was a bad idea
3) For EXI, application/exi is not needed, and a +exi suffix is
   not needed, because EXI will not be popular.

As for 1) and 2), can you tell us what YOU think would be best?
As for 3), please note that there is no need for a format to
be (or expected to going to be) popular in order to register
a media subtype.

Regards,    Martin.

At 05:09 08/10/17, Mark Baker wrote:
>On Wed, Oct 15, 2008 at 9:28 PM, Martin Duerst <duerst at> wrote:
>> Hello Carine,
>> At the moment, just a questions:
>> Given that there are many types of the form application/foo+xml,
>> will we see proposals for application/foo+exi in the future?
>That's a great question.
>In hindsight, I believe it was an error to mint application/xml and
>text/xml for XML content.  While there are valid uses cases for them,
>they are rare in practice (similar reasons to why one would need to
>deliver HTML content as text/plain).  But every single time I've seen
>them used in the wild was in error, and should have instead been a
>format-specific type like application/xhtml+xml or
>It naturally follows that I don't think EXI should have its own media
>type.  That's not to say I believe we need a "+exi" suffix though,
>seeing as a) the value of the "+xml" suffix has been negligible, and
>b) it's not clear that EXI is going to be popular.

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#       mailto:duerst at     

More information about the Ietf-types mailing list