I attended a meeting last night of the
ACCU
(Association of C/C++ Users); there was an
interesting talk by Nile Geisinger of
"dLoo".
He presented an interesting theory of what causes
programmers to move to new programming paradigms,
and their move beyond Object-Oriented Programming
to what they call "Symbol-Oriented Programming".
Our discussion about it included mentions of
Aspect-Oriented Programming, Web Services,
Functional Programming, Security, etc.
dLoo's product, "SpringBox", uses the combination of
Symbol-Oriented Programming with web-addressable
code units ("Symbols") to build a community-based way
of growing domain-specific languages.
Current version of SpringBox is written in Python; plans
are to move it to being wholly written "in its own Symbols".
It isn't really related, but it reminds me in some ways of the
gzigzag project (name
change due soon) that implements some of Ted Nelson's
original ideas about how information might usefully be
structured.
Whether either of these projects yield useful programming
language insights, I think they both relate to Lambda's
charter in that they show a couple of directions for programming
languages to step away from the single-view, linear syntax
that they've had for a half-century. I know that such steps
have been attempted many times before, and have mostly failed
to influence programming languages very much. Don't know
why I think that would change now -- wishful thinking?
|