Re: Using word generators (was Re: Semitic root word list?)
From: | James W. <emindahken@...> |
Date: | Tuesday, January 9, 2007, 16:17 |
On Tue, 9 Jan 2007 08:37:06 -0700, "Elyse M. Grasso"
<emgrasso@...> said:
> When I was actively building the Cherani dictionary, I wrote a script
> that
> would generate lists of random phonemically valid Cherani words,
> excluding
> those already in the dictionary. Then I picked words that felt right for
> the
> meanings. Sometimes I needed to generate several selection lists before
> I
> found the right word, but having the random lists to look at helped
> decrease
> the Human/English bias in the final word lists. (Cherani was a non-human
> language.)
I did a similar thing when I was working on a:seka`eni (it's still a
living project, but just dormant at the moment).
I created a list of 1-3 syllable words using Perl and then put them all
in a spreadsheet (all 199,654 of them!). Then I made the spreadsheet so
it would choose one entry at random. I'd collect several good (that is,
ones I liked...) words into another list. Then when I needed a lexicon
entry, I'd choose from the second list, based on the appropriateness of
the match.
If I only hand-created each word I'd never get anything done... :)
--------
James W.