Syntax Parsing, BNF Grammar, Republicans, etc.

I thought it was fascinating to read about the sort of syntax parsing used for the analysis of programming languages. The language recognizers and generators briefly described in the chapter sound like they might also be effective tools in the realm of natural language processing. They'd just have to be incredibly complicated, is all.  I'm mostly excited about this because of its potential relevance to AI systems that can understand and communicate through natural language, which is, of course, one more important step toward my conquering the galaxy through the military superiority of my cyborg armies and one day establishing supreme hegemonic rule over all of known space making the world a better place for everyone!

The BNF notation strikes me as a cool method for describing languages and pointing out differences between them in papers and such.  It's particularly neat to see this in documentation for languages, example cases of which you can find right here.  And also here! (What we're learning is helping us to learn other stuff more effectively O.o !?) I do wish people would listen to ISO every once in a while and go along with a standardized notation, though.  Everyone using their own bastardized version of the same notation system just isn't helping anyone...  

Not much else to say at the moment.  Nice catch on the stuff about the various versions of ALGOL and their mysteriously and counter-intuitively differing levels of understandable metalanguage, Mary.  I doubt I would have noticed that.  It's a funny thing about humans, that we tend to see things as needing attention only when their failures are glaring us in the face.  The metalanguage used to describe ALGOL 60 was perfectly lovely (warning: making assumption here), so less attention was paid to that aspect of language description when the folks were working on the new version of ALGOL, resulting in prohibitively abstruse documentation for ALGOL 68.  On the other hand, if the documentation for ALGOL 60 had been inscrutable, the ALGOL 68 stuff would probably have been a lot clearer.  When problems are current, we see them as needing to be fixed, so we fix them!  When the problem has been fixed for a while, it's easy to forget that there was ever a problem in the first place and to fall back into making the same mistakes all over again.  Incidentally, this phenomenon also serves to explain why Republican presidents are sometimes elected.