You know, Wolfram has an enormous ego. He thinks he revolutionized science. He thinks he invented computer algebra systems. He thinks he has designed the world's best programming language. 3 strikes.
If you want to look for a simple language, you could look at ones aimed at simple people (children). Logo for example. Or you could look at ones with a simple description (Scheme comes to mind). Or even Turing machines or Markov rewriting systems. Or simple to understand (perhaps Scheme again?) Of course some people might think Mathematica is simple. By the same kind of reasoning, solving "world hunger" is simple: Grow enough food for everyone and give it out. Playing a violin is simple: use one hand to hold down the strings and the other to run the bow back and forth. That is, if you think it is simple, it is because you don't understand it. Now about that "natural language" stuff. People have proposed to do this for 50 years or so, on and off. That is, use natural language for programming. COBOL (1961). Lots of other ideas, too. Warren Teitelman's thesis work on DWIM. Linguistics specialists with any number of ideas that didn't pan out. I have tried Wolfram's Alpha a number of times, and its natural language understanding seems to be quite deficient, and occasionally laughable. Compound that with the fallibility of humans to be able to express algorithms unambiguously in natural language anyway, and one might reasonably doubt the claims that a front end to Mathematica would make it "simple". I just scanned the current wikipedia page on "natural language programming". It appears to be a marketing blurb. too bad, wiki pedestrians. RJF On Sunday, March 11, 2012 3:44:22 AM UTC-7, Harald Schilly wrote: > > > > On Saturday, March 10, 2012 2:53:25 PM UTC+1, rjf wrote: >> >> >> It is not a simple language. > > > I'm sure you all know more about this than me. Is there a common way to > "measure" this? What I'm thinking about are those grammar dependency trees. > From my personal experience and looking at those graphs: narrow trees with > not too many nodes are easier to understand, whereas broader trees with > more nodes are harder. > Here are some examples, e.g. Python is easy, as is JavaScript. Harder is > Perl and Ruby; I also found one for Java and C. > Might be interesting to create one for MMAs grammar! > > Python: http://www.flickr.com/photos/nicksieger/281055485/ > Ruby, Java, JavaScript: > http://antigreen.org/vadim/ProgLanguageComparison/grammar-vizualization/visualization-of-rubys-grammar.html > Ansi C: http://c2.com/cgi/wiki?BnfToDot > > H > -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
