Re: What's the word for...

andrewd@apanix.apana.org.au (Andrew Dunstan)
Tue, 22 Feb 1994 13:56:12 GMT

          From comp.compilers

Related articles
[9 earlier articles]
Re: What's the word for... norman@flaubert.bellcore.com (1994-02-19)
Re: What's the word for... tchannon@black.demon.co.uk (1994-02-20)
Re: What's the word for... moreaux@litsun31.epfl.ch (1994-02-20)
Re: What's the word for... sasghm@unx.sas.com (1994-02-21)
Re: What's the word for... weberwu@tfh-berlin.de (1994-02-21)
Re: What's the word for... jan@neuroinformatik.ruhr-uni-bochum.de (1994-02-22)
Re: What's the word for... andrewd@apanix.apana.org.au (1994-02-22)
Re: What's the word for... muysers@capsogeti.fr (1994-02-23)
Re: What's the word for... sasghm@unx.sas.com (1994-02-23)
Re: What's the word for... lloyd@bruce.cs.monash.edu.au (1994-02-24)
| List of all articles for this month |
Newsgroups: comp.compilers
From: andrewd@apanix.apana.org.au (Andrew Dunstan)
Keywords: theory, comment
Organization: APANIX Public Access Unix, Australia, +61-8-373-5485 (3 lines)
References: 94-02-106
Date: Tue, 22 Feb 1994 13:56:12 GMT

CNS-ksf-+Jordan T.J. (tjj@netnews.summit.novell.com) wrote:
: Could someone please tell me what the word is for a language
: which can be written in itself?
: Thanks
: [Introspective? Bootstrappable? Lisp? -John]


What does "written in iself" mean? Many languages can have compilers
"written in" themselves, but are incapable of formally describing their
own grammar, since they are programming languages, not grammar description
languages. Conversely, BNF can formally describe its own grammar, but you
could not write a parser in BNF, since it is not a programming language.
Let's be a little more precise here.


This reminds me of Russell's version of a famous paradox. Consider
adjectives. Call those that can describe themselves "autologous" (e.g.
"short") and thos that cannot describe themselves as "heterologous" (e.g.
"long"). Now consider the adjective "heterologous". If it is heterologous
then it is autologous, but if it is autologous then it is heterologous.


cheers


andrew dunstan
[Darn that Epimenides. -John]
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.