Talk:Toy programming language

Is this copied from a textbook?
I noticed the last sentence of this article says "Whether this means that all purely interpreted or virtual machine languages are toys is left as an exercise for the reader.". I don't get why somebody writing an encyclopedia article would say something's left for the reader to figure out. I'd think this is copied directly from some textbook with no source given.


 * No, that wasn't a cut and paste; I wrote it in jesat, as a nod to the academic aspect of programming languages. Perhaps it was in poor taste; it is obviously misleading.  I would not object to it's removal. -- 66.15.179.18 (talk) 21:20, 13 December 2007 (UTC)

Distinguishing test
This last section is quite silly I'd say ... assuming your "toy language" is Turing complete, then of course you can write a compiler for itself in itself.

The operating system test is equally silly. Any language can be used to write an OS (again, if it is turing complete), as long as there is some way to run the language without one. For almost ALL modern languages, this requires some low level piping work - this would include Python, Java, Haskell, C#, Ruby. Surely at least one of those is "not a toy".

A far better test is whether the language implements facilities useful for humans, or whether it only provides primitive constructs necessary for its desired purpose. The latter would be a "toy" - by the admission of its designer. EatMyShortz (talk) 12:19, 16 April 2008 (UTC)