[Home] [Groups] - Message: [Prev in Group] [Next in Group]
29456: Re: [good] Re: [MUD-Dev] Parser engines
[Full Header] [Plain Text]
From: Travis Casey <efindel@earthlink.net>
Newsgroups: nu.kanga.list.mud-dev
Date: Tue, 30 Mar 2004 22:32:28 -0500
References: [1] [2] [3] <-newest
Organization: Kanga.Nu
On Friday 12 March 2004 03:29, Manuel Lanctot wrote:
> That's the biggest challenge IMO. The English language is absurdly
> simple compared to some others. Sticking with French, this
> language has two possible genders for every noun: "masculin"
> (male) and "feminin" (female). Some nouns, like table, chair,
> river, etc. are "feminine" - which changes the pronoun in front of
> it. I simply can't imagine a dynamic parser recognizing which word
> is which gender. And I'm not even talking about singular/plural -
> there are hundreds
> of exceptions. Basically, you'd need a dictionnary database with
> all the exceptions (I think there are around 10,000 syntax
> exceptions in French). And I guess French is not the worst
> language.
Why worry about it? The genders of inanimate objects are utterly
arbitrary. Simply accept both forms of the pronouns as valid. (For
that matter, trying to accept only the "grammatically correct"
version is likely to cause problems, since people will make
mistakes... and since different dialects may have differences in the
gender they assign to things.)
Now, it is a problem for sentence generation... but for parsing, it
shouldn't be much of one.
--
|\ _,,,---,,_ Travis S. Casey <efindel@earthlink.net>
ZZzz /,`.-'`' -. ;-;;,_ No one agrees with me. Not even me.
|,4- ) )-,_..;\ ( `'-'
'---''(_/--' `-'\_)
_______________________________________________
MUD-Dev mailing list
MUD-Dev@kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev