xml-dev - Re: Another Question

[Olivier DUBUISSON <Olivier.Dubuisson@francetelecom.fr>]

tpassin@home.com wrote:
> 
> > Or why is ASN.1 better than XML Schemas? (these two can be compared
> > because they allow the definition of datatypes.)
> >
> Yes, this is well put.  ASN.1 is actually pretty nice for schemas.  But it
> gets very interesting to try to come up with a set of XMl encoding rules,
> especially if you limit yourself to defining the XML schema with DTDs.  I

If I understand you rightly, you don't want to use the XML syntax to
define XML Schemas. If right, why?

> think the proposed XER rules are weak and not very well put together from an
> XML point of view.

Please note that the XER encoding rules are not an official
proposition of the ASN.1 working group.

> I'm involved right now in XML encoding rules for a subset of ASN.1 (for a
> particular problem domain).

It seems that you're not the only one to develop XML encoding rules
for ASN.1 data.
Would it be possible to you to send me a document that presents your
work and could be submitted as an input document to the next meeting
of the ASN.1 working group, so that we could keep trace and be
informed? Thanks.

> Since I have only a slight reading acquaintance
> ASN.1, it's been even more interesting.  One of the interesting issues is
> that ASN.1 has both named data types and named identifiers (an element has
> both a type and an identifier), but in XML you have only element names
> (basically representing complex data types).  So do you use identifiers or
> data type names, and what do you do with the other of the pair?  There are a
> lot of possible solutions.

Let's define the following structured type:
Person ::= SEQUENCE {
        name    UTF8String,
        age     INTEGER }
A value of that type can be:
john Person ::= {
        name "Doe",
        age  45 }

As a structured value, "john" is a value of type "Person" that has 2
fields: the former is named "name" and contains the value "Doe", the
latter is named "age" and contains the value 45.
I don't know if this answers your question, but the point I want to
make is that you need to consider values (better than types) when you're
defining encoding rules.

I propose that we discuss this point privately if you need further
explanations (I don't think that a lot of people are interested in
this thread :-)
-- 
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/

***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************