Re: ASN.1 and XML

[Al Snell <alaric@alaric-snell.com>]

On Tue, 29 May 2001, Rick Jelliffe wrote:

> > But AFAIK nobody has ever produced a national version of ASN.1, probably
> > for the good reason that he/she would have to implement the associated
> > tool! Actually the same story goes with most programming languages,
> doesn't
> > it?

> ASN.1 could not claim to be as useful as XML without it.

It's just a matter of implementations. Given an increase in demand in
non-English speaking countries, there will be enough of a market for an
implementation that uses Unicode and is entirely independent of Roman
script to make it happen.

I see a distinct difference between a well-implemented poorly designed
system and a poorly-implemented well designed system. Many people choose
the former - "Well, at least it's been ported to Windows, and it supports
X, Y, and Z"... but good implementation can only make up for a bad design
so far, and you can't change the design without breaking all those
precious implementations.

The poorly implemented well designed system, on the other hand, can
improve with minimal effort, if only people would stop looking at the
well-implemented system. Sure, you may need to use the well-implemented
one when you need those features in a hurry, but it's in everyone's best
interest to try to focus development on the better *idea*, in order to
move it to the status of a tool that's well designed *and* well
implemented.

I hear lots of people who are really quite anti-software-design. Who want
everything to evolve. This results in Perl, UNIX, Windows, and
Sendmail. Fear! Fear!

Poor implementation can be fixed more easily than broken design...

> 
> Cheers
> Rick
> 

ABS

-- 
                               Alaric B. Snell
 http://www.alaric-snell.com/  http://RFC.net/  http://www.warhead.org.uk/
   Any sufficiently advanced technology can be emulated in software