rbbb@rice.edu (David Chase) (10/11/86)
> I find that the Three Holy Documents of C (K&R, H&S, the draft ANSI > standard) are a little muddy on this point, though I agree with Fabbian. If the "standard language definition" is ambiguous, then it isn't a definition. Have them clean up the definition. In this case, having an answer is much more important than which answer you have. David
throopw@dg_rtp.UUCP@ndmce.uucp (Wayne Throop) (10/17/86)
> rbbb@rice.edu (David Chase) >> throopw@dg_rtp.UUCP (Wayne Throop) >> I find that the Three Holy Documents of C (K&R, H&S, the draft ANSI >> standard) are a little muddy on this point, > If the "standard language definition" is ambiguous, then it isn't a > definition. Have them clean up the definition. In this case, having an > answer is much more important than which answer you have. If you can find me a definition of *anything* (in English, of comparable complexity to a general purpose computer programming language) which is unambiguous, then I'll eat the manual this description printed in. Applying this overly rigid restriction on the meaning of "definition", I doubt you can find a "definition" of any programming language at all. However, in a suitable diluted form, I agree with David's point. On the other hand, I do think that progress is being made. H&S is clearer than K&R on this point, and the draft ANSI C standard is clearer still. -- Perfection must be reached by degrees; she requires the slow hand of time. --- Voltaire -- Wayne Throop <the-known-world>!mcnc!rti-sel!dg_rtp!throopw
anw@nott-cs.UUCP (10/31/86)
In article <286@ndmce.uucp> in@pollux.UUCP writes: >If you can find me a definition of *anything* (in English, of comparable >complexity to a general purpose computer programming language) which is >unambiguous, then I'll eat the manual this description printed in. The Algol 68 definition is known to be unambiguous. Sadly, it is also known that all the printed versions contain bugs, some doubtless not yet discovered. Oh yes, :-). Seriously, W-grammars are *very good* for providing careful definitions of languages, and it is a shame that they are not more widely used -- many of the silly arguments one gets into over C and Pascal could easily be resolved by reference to the standard grammar, if one existed. -- Andy Walker Maths Dept, Nottm Univ.
henry@utzoo.UUCP (Henry Spencer) (11/06/86)
> >If you can find me a definition of *anything* (in English, of comparable > >complexity to a general purpose computer programming language) which is > >unambiguous, then I'll eat the manual this description printed in. > > The Algol 68 definition is known to be unambiguous. ... It's not written in English. > ...it is a shame that [W-grammars] are not more widely used -- many of > the silly arguments one gets into over C and Pascal could easily be resolved > by reference to the standard grammar, if one existed. I believe that X3J11 specifically rejected use of a formal mathematical definition of C (which is what a W-grammar would be) on the grounds that while it would improve the precision of the definition, it would greatly reduce the size of the audience that could understand the definition. Joe Random User, or even Joe Random Compiler Writer, cannot be assumed to be fluent in W-grammars. Although there are complicating factors, it would appear that languages defined in English have been rather more successful of late than languages defined in formalisms. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,decvax,pyramid}!utzoo!henry