Re: Superlanguages: Syntactic and Semantic Supersets of other Languages
Posted: Mar 13, 2008 10:22 PM
> > The XL language (http://xlr.sourceforge.net/) is also
> > about this, and this is called "concept programming"
> Great link, thanks! I followed a link on that site to find
> an old DDJ article about
> [a href="http://www.ddj.com/java/184404696"]Moka[/a] which
> appears to be a Java preprocessor for language oriented
> programming. It is quite interesting.
Right. There are also mentionings of Mozart (http://mozart-dev.sf.net/) (confusingly, there's another project with the same name (http://www.mozart-oz.org/), around), and according to the author, XL is pretty much the successor of Mozart, no longer using Mozart as an intermediate language, having a much simpler internal structure of the programs (a programmer-accessible AST) than Mozart, and is actually written in XL, itself (bootstrapped).
I've not had a chance to look that much into XL, myself, but I intend to (there's also a mailing list there, where one may discuss the language. Unfortunately, at the moment, the list is spammy).
> It seems the term "concept" and "concept oriented
> programming" in one that has multiple meanings attached to
> it over the past few years:
> * http://www.ddj.com/184410968 (1999)
> tml (2007)
> * http://www.research.att.com/~bs/popl06.pdf (2006)
Yes. Just like "Language-Oriented Programming", "Intentional Programming", "Concept Programming", it seems that not only are there many names for this - there are also several meanings for some of the same names... :)
When I heard about XL and "concept programming", it was in particular what is described in the last link above (the Stroustrup/Dos Reis article), I was thinking of, which I found a little confusing. And that, uhm, concept, also has several names, such as "constrained generics". That's one of the most exciting things I think is in the coming C++0x, but it's rather different from LOP, etc.