Announcement of Potential New Project : EXPRESS and Binary Data Representation

[David Price]

Hi Ed,

All good points. Especially the last one about SC4 migrating to mainstream
technology.

I can't explain all the requirements, but my understanding is that there are
a few applications where file size, randon access and access efficiency are
still important (e.g. FEA/CFD). For those few applications, there's a move
afoot to do something. I expect the result will either be what you
suggested...

> map EXPRESS to one of the half-dozen standard binary protocols that
> already exist -- HDF, ASN.1 BER/PER/VER, DDIF or whatever the telemetry
> folks are using these days.

or use Part 28 and whatever W3C decides to do on binary. I don't expect the
result will be an EXPRESS-to-SC4-binary standard.

On the existing binary protocols, do you have an opinion on which would
handle EXPRESS-defined data most smoothly?

Also, I hope you noticed the EXPRESS/UML/P25 E2/UML Usage meeting at the ISO
as well. My hope is to begin getting SC4 to addressing some of the migration
path away from EXPRESS scenarios seriously.

Cheers,
David

> -----Original Message-----
> From: Ed Barkmeyer [mailto:edbark at nist.gov]
> Sent: 17 February 2004 16:03
> To: David Price; SC4 WG11
> Subject: Re: Announcement of Potential New Project : EXPRESS and Binary
> Data Representation
> 
> Pardon me while I laugh!
> 
> In 1991, SC4 couldn't use ASN.1, a 1980 CCITT standard (and a 1984 ISO
> standard) for specifying message structures and binary exchange formats,
> because humans couldn't read ASN.1 binary files without software tools.
>   And of course, that logic applied to all kinds of users of electronic
> data interchange standards, so ASN.1, though widely implemented, was not
> successful.
> 
> In 1998, W3C developed XML -- a new standard for exchanging all kinds of
> messages and data, using text characters that people could read, with a
> weak structuring language (DTDs) based on the 1979 SGML standard.  And
> XML is reputed to be widely successful, although 5 years after adoption
> "production" exchange of XML messages/files is a rara avis. Then in
> 2000-2002, W3C developed a language that had all the basic features of
> ASN.1 -- data types, local and global names, <all>, <sequence> and
> <choice> structures, an object-identifier data type (URI) -- XML schema.
>   And now that they have a structuring language equivalent to ASN.1, it
> is time for W3C to invent a new binary exchange notation, equivalent to
> the 1984 "ASN.1 basic encoding rules", which people won't be able to
> read, and whose only relationship to XML is that W3C is developing it.
> Oh, wow.
> 
> In our time of 200 GB desktops and 100 MB operating systems, I'm trying
> to understand the motivation for "compact" binary representation of
> anything.  Wireless downloads?
> 
> But in any case, all we SC4 have to do is stop making EXPRESS schemas
> (because they are not "mainstream"), convert our models to XML schema
> (which is mainstream but nearly unreadable by humans), and let W3C
> develop our binary exchange protocol, right?  Or perhaps we SC4 should
> now expend our resources developing a competing binary representation,
> so as to keep SC4-only toolsmiths in business.  Or maybe we just need to
> map EXPRESS to one of the half-dozen standard binary protocols that
> already exist -- HDF, ASN.1 BER/PER/VER, DDIF or whatever the telemetry
> folks are using these days.
> 
> Unfortunately, like the XML (schema) problem, and the C++/Java problem,
> EXPRESS, especially as used by SC4, doesn't quite capture what you need
> to make effective use of these other data architectures.  You need to
> mark up the EXPRESS model for such conversions.  But do we really want
> yet another god-awful configuration file (like Part 28)?  Shouldn't we
> just bite the bullet and standardize the XYZ representation, using the
> XYZ modeling language, in an Annex to the AP (or Module)?
> 
> SC4 has got to get out of the IT standards business -- we don't know
> anything others don't know and don't have better resources to develop,
> or re-develop, as in W3C.  All future WG11 NWIs should standardize
> migration paths to mainstream information technology standards, and over
> time, EXPRESS should take its rightful place in the dustbin of history.
> 
> -Ed
> 
> --
> Edward J. Barkmeyer                        Email: edbark at nist.gov
> National Institute of Standards & Technology
> Manufacturing Systems Integration Division
> 100 Bureau Drive, Stop 8264                Tel: +1 301-975-3528
> Gaithersburg, MD 20899-8264                FAX: +1 301-975-4694
> 
> "The opinions expressed above do not reflect consensus of NIST,
>   and have not been reviewed by any Government authority."
> 
> David Price wrote:
> 
> > FYI, I saw this in the W3C news.
> >
> > XML-Binary Packaging and SOAP Transmission Optimization Working Drafts
> > Published
> >
> >    The XML Protocol Working Group has released the First Public Working
> >    Draft of "XML-binary Optimized Packaging" (XOP). XOP allows efficient
> >    serializing of certain types of XQuery and XPath 2.0 element content.
> >    Based on XOP, the group also published an updated Working Draft of
> >    the "SOAP Message Transmission Optimization Mechanism" for improving
> >    SOAP performance. Visit the Web services home page.
> >
> >    http://www.w3.org/TR/2004/WD-xop10-20040209/
> >
> >    http://www.w3...org/TR/2004/WD-soap12-mtom-20040209/
> > <http://www.w3.org/TR/2004/WD-soap12-mtom-20040209/>
> >
> >    http://www.w3.org/2002/ws/
> >
> >
> > Cheers,
> >
> > David
> >
> >
> >
> >
> >
> > -----Original Message-----
> > From: owner-wg11 at steptools.com [mailto:owner-wg11 at steptools.com] On
> > Behalf Of David Price
> > Sent: 07 February 2004 13:13
> > To: Wg11 Exploder; Step-Imp; Xmlsc4
> > Subject: Announcement of Potential New Project : EXPRESS and Binary Data
> > Representation
> >
> >
> >
> > Meeting Name: EXPRESS and Binary Data Representation
> >
> > Start Date: 2004-03-02 Tuesday
> >
> > Time(s): 1:00pm - 3:00pm 3:30pm - 5:00pm
> >
> >
> >
> > Agenda
> >
> >
> >
> > 1) Discuss initiation of PWI or NWI
> >
> > 2) Initiation of Requirements Document
> >
> > 3) Initiation of Related Activities/Standards/Tools Document
> >
> > 4) Plan next actions
> >
> >
> >
> > The idea is to standardize or use an existing binary representation of
> > data for EXPRESS users whose applications require large volumes of data
> > such as the analysis APs. An initial meeting was held at the Poitiers
> ISO.
> >
> >
> >
> > Note that there are already binary representations of data and that W3C
> > is having a related discussion with respect to a binary representation
> > for XML documents. All these activities are relevant. The expectation is
> > that a PWI or NWI will result but that it may not be necessary to invent
> > a representation ourselves. We may simply standardize the use of
> > something that already exists or is under development.
> >
> >
> >
> > Please feel free to contact me if you have questions or comments,
> >
> > Cheers,
> >
> > David
> >
> >
> > Phone +44 207 704 0499
> > Mobile +44 7788 561308
> > 8 Highbury Place, Flat 5
> > London, UK
> > N5 1QZ
> >
> 
>