Re: Non-english IF?!


12 Dec 1995 14:57:26 GMT

In article hl4@selene.rz.uni-duesseldorf.de, lerchj@classic11.rz.uni-duesseldorf.de () writes:

>
>Are some languages more suited to computer-IF than others?
>It seems to me that it is more complicated to write a German
>parser that ''understands'' language considered relatively
>natural than an English one. And Esperanto is claimed to be
>easy to process with computers (I think that originally means
>'easy to translate automatically). How about other languages?
>

Esperanto would be easy because there is the one ending for
nouns, another for adjectives, and so on, without exceptions.
There is one gender (like English), and one definite article
(again like English), and no indefinite article (easier than
"a" or "an" for English). There is an accusative case, however,
but it is easily derived from the word stem.

Most other languages have more inflexions. But languages
like Swedish or German wouldn't be so hard to do, I think.
What is needed is more properties for each object, for instance
one property to indicate which form of the definite article to use,
one to indicate how the plural is formed, etc. Highly inflected
languages, say Russian or Finnish, would be harder to do well.
Russian has three genders and six cases, conjugates verbs according
to both person and tense, and has a lot of exceptions to the rules.

>Also, many IF-writing systems seem to contain inbuilt messages
>in English that make them quite useless for writing in other
>languages.

Z-Machine interpreters don't have much English messages in them;
only the prompts for filenames when you do save or restore.
It would be easy to change them.

Björn