Re: talking about LR(n) parsers (Richard L. Goerwitz)
Tue, 15 Oct 1991 04:31:03 GMT

          From comp.compilers

Related articles
LR(n) parsers (Steve Boswell) (1991-10-10)
Re: LR(n) parsers (1991-10-14)
Re: talking about LR(n) parsers (1991-10-15)
| List of all articles for this month |

Newsgroups: comp.compilers
From: (Richard L. Goerwitz)
Keywords: parse, syntax
Organization: University of Chicago
References: 91-10-036 91-10-046
Date: Tue, 15 Oct 1991 04:31:03 GMT (Boris Burshteyn) writes:
> Briefly, [my] algorithm works as follows. First, the LR0 automata
>is built....

I know this is going to sound pedantic, but it's really important for
people involved in automaton theory to use the word 'automaton' in a
standard, accepted fashion. 'Automaton' is a singular (derived from
the Greek neuter singular), and its plural is 'automata.' So you can
talk about automaton theory, and about creating several automata, but
not about 'an automata.' It's kind of like talking about 'parser con-
struction' and about constructing several parsers. You would never
talk about 'parsers theory' or about constructing 'a parsers.' Carry
the analogy over to 'automaton' and you'll get the idea.

Really, dumb points of usage like this should not impinge on the basic
point people have to make, but inevitably people's inner prejudices
get involved. Bottom line: Here no one presumably cares, but if you
are publishing, be sure you've got the grammar straight.


      -Richard L. Goerwitz goer%sophist@uchicago.bitnet rutgers!oddjob!gide!sophist!goer
[Sorry, I should have caught that. I'm not crazy about "the data is",
either. -John]

Post a followup to this message

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