Re: "Binary XML" proposals

[Olivier Dubuisson <Olivier.Dubuisson@francetelecom.com>]

[It's probablyt one of the last emails in this thread I'm sending to the
xml-dev mailing-list; I also Cc: to Al's new xml-bin mailing-list.]

Al Snell wrote:
> 
> On Wed, 11 Apr 2001, Anders W. Tell wrote:
> 
> > ASN1 is a good standard but difficult to use and there its is possible to write
> > much "simpler" parsers ANS 1 ones.
> 
> There's ASN.1 - the language for describing data formats, like a DTD - and
> the actual BER or PER or DER data, though. You don't need to worry about
> ASN.1 itself when using the data; just like not all XML parsers need to be
> (or should be) validating.

Precisely!
And to emphasize your point with a straightforward comparison:
- an XML document is equivalent to a, say, BER encoding;
- a DTD or better, an XML Schema, is at the same level as an ASN.1 module
(notice I don't write "equivalent"!).

As a consequence, a binary encoding for XML such as Binary XML adds a
3rd level to Schema + XML document compared to the two levels of ASN.1
module + PER (with a much more compact encoding with PER, which is also
well-suited for embedded devices).
One can say that a Binary XML decoder gives high-level data structures back to 
the application, but the same applies for a PER decoder.
And this PER decoder can (will) be automatically produced from the XML
Schema if one uses what is described at: http://asn1.elibel.tm.fr/xml/
-- 
Olivier DUBUISSON
france telecom R&D
     _                 DTL/MSV - 22307 Lannion Cedex - France
    ( )           tel: +33 2 96 05 38 50 - fax: +33 2 96 05 39 45
    / \/               --------------------------------------
    \_/\               Site ASN.1 : http://asn1.elibel.tm.fr/