> XML's long-term contribution isn't that it solved data exchange
> problems. It's that it made it possible for people to believe those
> problems could be addressed on some useful level.
>
> Both of these effects are primarily social rather than computational.
>
I would put it differently: XML has solved data exchange problems, and
yes -- the reason is that those problems are primarily social.
I saw any number of groups struggling in the 1980s and 1990s with
defining data exchange standards, who failed because they got bogged
down with syntax and encoding issues; once XML came on the scene, they
were able to raise the level of debate and discuss the semantics (where
they actually had some useful expertise).
Michael Kay
Saxonica
