[
Lists Home |
Date Index |
Thread Index
]
Joe English wrote:
> I find it more valuable that new programs should be
> able to read old data than that old programs should
> handle (the same old data structures dressed up in)
> new concrete syntax.
ASN.1 works both ways. Given the Encoding Control Notation and
the ability to define new encoder/decoders, you can build an ASN.1
system that will read any old data structure and then have it
converted to any of the other syntaxes (assuming that the target
syntax is rich enough to express the data...)
So, yes, an ASN.1 suite should make it easier for you to
define the precise conversions of old legacy data to XML... Or, BER,
or PER, or etc...
Since ASN.1 makes a clear distinction between abstract syntax
and concrete syntax, the only question is: "Can you translate your
concrete syntax to the abstract form?" If the answer is yes, you get
interchange. And, since ASN.1 has been in the field for so long, it
has already dealt with most of the issues related to these
conversions.
The essential thing here is to have an abstract, intermediary
representation and then have deterministic transformations to and from
concrete syntax. It is very much the same as the well respected rule
of program design that calls for making a distinction between
interfaces and implementations (or, as I used to say back in the 80's
"Separation of Form and Function."... )
bob wyman
|
