W3C home > Mailing lists > Public > xml-dist-app@w3.org > December 2001

Re: Media type encoding parameter?

From: Mark Baker <distobj@ACM.ORG>
Date: Fri, 21 Dec 2001 23:53:27 -0500 (EST)
Message-Id: <200112220453.XAA30178@markbaker.ca>
To: jacek@systinet.com (Jacek Kopecky)
Cc: xml-dist-app@w3.org
Oops, this is embarrasing.  I intended this parameter to be for the
envelope, not the encoding.  It may be useful to bring the encoding
out too though.  I've got some questions for you below, Jacek.

>  Mark,
>  I don't think this is a good idea. First, there can be more than
> one encoding in the message on different pieces of data, even
> inside the Body. Which would you indicate?

Good point.

>  Anyway, it seems if we sum everybody's wishes, we'll find out
> that almost everything from the message is important to be known
> without cracking the envelope. Excuse my outburst, but THIS IS
> XML, LET'S GET OVER IT! If we put everything from the envelope to
> the outside, too, why need the envelope? 8-)

Heh. 8-)

>  OK, the namespace of the root element might be useful for fast
> dispatch, in case of SOAP the "intent" of the message can be also
> useful for that purpose, also I'm nervous about that, but the
> encodingStyle is not a thing the server dispatches on.

That isn't a requirement of a media type parameter.  "charset" is also
not used to dispatch.

> Therefore,
> I say it's not a good idea.
>  Best regards,

It is good practice to provide, as metadata separate from the message
itself, any information that impacts a processor's ability to
successfully process the message.  This helps performance for sure
(though not in dispatch, but in preventing messages from being
processed when it is known that there's no point), but it's more a
statement about interoperability.

So re the envelope, I'm suggesting an optional parameter of "envelope"
whose value would be an envelope URI.

And re the encoding, I'd like to know more about its common use.
If it were most commonly used on the envelope/header/body, and
there was normally only one, then I'd be tempted to create the
parameter.  If not, then I'd leave it.

Mark Baker, Chief Science Officer, Planetfred, Inc.
Ottawa, Ontario, CANADA.      mbaker@planetfred.com
http://www.markbaker.ca   http://www.planetfred.com
Received on Sunday, 23 December 2001 08:17:19 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:11:43 UTC