RE: [emergency] ASN.1 (Abstract Syntax Notation) and XML

[Eliot Christian <echristi@usgs.gov>]

At 10:19 AM 2/18/2003 -0500, Allen Wyke wrote:
>Question: as I read through this it seems that ASN.1 is more of a
>method to describe various types of data, in a standard way, across
>communication protocols. In other words, it would allow things like
>audio, video, etc. capture devices to describe the actual data it was
>transmitting, which would then allow many other applications the
>capabilities to read, understand, and/or transform the data. Is this
>correct?

ASN.1 is a notation for describing structures of data elements, and
those structures can be as fine-grained as needed. An ASN.1 abstract
specification of data structure is equivalent to an XML Schema.

In XML, on-the-wire encoding of the XML instance is controlled only
by the encoding filter given in the processing instruction
(the default encoding id "UTF-8"). In ASN.1, a much richer set of
on-the-wire encodings is available. These include "Packed Encoding
Rules" (think of ZIP'd files), "Distinguished Encoding Rules" (each
packet contains only differences from previous packet), and XML
Encoding Rules (create XML instance from the ASN.1), among others.

>[...] what role do you see it playing in the EM TC other
>than a point of reference for device implementers who may wish to
>implement one of our future standards and currently have (or need to)
>go through an ASN.1 process first?

I've heard it said that interoperability is about "coming to agreement
on what few things need to be the same so that everything else can be
different". My point is that the specific syntax of data structures is
one of the things that should be allowed to be different, insofar as
ASN.1 and XML are equivalent for most purposes. People who have worked
out interoperability agreements using ASN.1 do NOT need to have new
meetings about how to do interoperability using XML. Conversely, people
who have achieved interoperability with an XML syntax can exploit the
encoding efficiencies of ASN.1 without changing their XML agreements.

In my experience, the really valuable progress toward interoperability
is made when people focus on agreeing the semantics of data elements
that are being interchanged. Once agreed, the data structures are
typically interchanged using a variety of syntaxes depending on the
circumstances.

Eliot



Eliot Christian echristian@usgs.gov 703-648-7245 FAX 703-648-7112
US Geological Survey, 802 National Center, Reston VA 20192