W3C home > Mailing lists > Public > public-exi@w3.org > August 2007

Re: How to distinguish XML and EXI

From: Kenichi Taniuchi <ktaniuchi@tari.toshiba.com>
Date: Tue, 07 Aug 2007 13:31:45 -0400
Message-ID: <46B8AC81.8000603@tari.toshiba.com>
To: "Vogelheim, Daniel" <daniel.vogelheim@siemens.com>
CC: public-exi@w3.org

Hi Daniel
Thanks a lot, actually I am working on IEEE 802.21 which is for
media independent handover. That spec is still draft version.
In the spec XML format is used for network information service using
SPARQL. but the service will be used for link layer means size sensitive.
So, I am interested in EXI to reduce the size of XML.
To distinguish the format of query reply, IEEE 802.21 use MIME type.
I thought different MIME type definition is good to know actual encoding
type.
Because it would be raw XML, EXI or ZIP(?).
Even MIME is not defined, at least XML engine can distinguish between
XML and EXI
based on the first bits.

Regards,
Kenichi

Vogelheim, Daniel wrote:
> Hello Kenichi,
>
>   
>> Another question is came.
>> Do you think new MIME definition will be needed for EXI for
>> some protocols ?
>>     
>
> Probably yes, but that is still an open question. We've actually been
> discussing this very issue at a recent meeting, and we plan on
> publishing a 'Best Practices' document to accompany the spec to deal
> with questions like this.
>
> In an HTTP setting, the current thinking is that EXI would ideally be
> used as an HTTP content encoding, so that it would work transparently
> with the existing HTTP content negotiation features. For other settings
> or protocols (like serving/storing static files), a MIME type might
> indeed be useful or even required.
>
> The EXI WG plans on collecting some more feedback on the issue, and then
> determine whether to approach IANA on a new MIME type or not.
>
>
> Kenichi, if you have some specific use cases or arguments for or against
> an EXI MIME type, we would keep them in mind when making our decision.
>
>
> Sincerely,
> Daniel Vogelheim
>   
Received on Tuesday, 7 August 2007 17:32:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:52:42 UTC