RE: Comments on latest packaging spec

[Sam Odom <samodom@webmethods.com>]

Dick --

I'm not sure if this question has yet been raised or not, but I was hoping
that someone could provide some insight on how packaging is supposed to work
with FTP.  According to the latest version of the spec: "...in the case of
FTP no transport envelope is needed."  I can understand how no transport
envelope would be needed since the actual 'document' could contain the
multipart MIME message, but in the case of other transports, the headers
actually provide information specific to the multipart message.  In
particular, the content-type header specifies the boundary and (possibly)
charset of the message.  If FTP is used without the document containing an
enveloped message, how is one to determine the boundary and charset of the
message?  It's easy enough to determine the appropriate content handler for
a message received via FTP if you associate file extensions (or URIs?) with
particular content-types, but how do you digest it without knowing some of
the information the headers give you?

Thanks --
Sam Odom
webMethods

> -----Original Message-----
> From: Dick Brooks [mailto:dick@8760.com]
> Sent: Sunday, July 23, 2000 7:10 PM
> To: Ebxml
> Subject: Comments on latest packaging spec
>
>
> I've received the following feedback from ebXML's "weekend warriors":
>
> There are incorrect values ("ebxml") for the type attribute on pages
> 8,11&14. The correct value is "application/vnd.eb+xml". I've made these
> changes and will issue an updated spec with these corrections.
>
> There has been some discussion over the choice of iso-8859-1 as
> the default
> character set for the charset attribute in the Message Envelope
> Content-type
> header. It was pointed out that several XML related
> specifications recommend
> using the UTF-8 or UTF-16 character sets.
>
> Note: Implementers are free to use and support ANY character set that's
> appropriate to their needs. In other words, the ebXML packaging
> spec doesn't
> limit the use of charset. The use to iso-8859-1 as a "default" is intended
> to provide a common "baseline" level of character set support
> expected from
> ebXML software implementers.
>
> Given our close alignment with XML, should the ebXML packaging spec be
> changed to use one of the UTF-* character sets as the default, instead of
> the iso-8859-1 character set?
>
> Thanks to all who provided feedback.
>
>
> Dick Brooks
> Group 8760
> 110 12th Street North
> Birmingham, AL 35203
> dick@8760.com
> 205-250-8053
> Fax: 205-250-8057
> http://www.8760.com/
>
> InsideAgent - Empowering e-commerce solutions