Re: Hyperlinking a dictionary to a corpus

Gary Shannon wrote:
> Elomi, which, for the record, is the invention of
> Larry Sulky, (I just play with it for fun because it
> sounds neat and is easy to learn) does have that
> advantage. I'm not sure how I'd do it for more complex
> languages. My own Latin-like Tazhu has all kinds of
> case endings and verb conjugations that would make a
> project like that a nightmare. I imagine the program
> would need some kind of cross reference tables to
> relate one form of the word to all the other forms so
> that (to use a Latin example) the table would have to
> equate "cogitare", "cogito", "cogitas", "cogitat",
> "cogitamus", etc., etc. to each other so that they
> would all generate the same HTML anchor name, probably
> "#cogitare".
Probably, yes. And shouldn't things like "morior, loquor" have not only
their own links, but also a link to a general discussion of deponent verbs??
It do get complicated.
>
> For an agglutiating language it would be a real
> nightmare!  I don't een want to think about that one
> ;-)
I was looking at Kash today, to see how/what I might exemplify, and decided
that _every_ word probably doesn't need to be. Further, what to do about
sandhi changes (e.g. most suffixes have -C(vl)V, -(nasal)C(vd)V and -C(vl)rV
allomorphs, which actually is discussed on the Morphology page-- so perhaps
just a link to that).  Otherwise I think I'll do a large bunch of
"organized" example sentences for some of the knottier words (like _mesa_
'one' which can be used in various constructions and meanings-- 'one of...',
'single, sole', 'first (ordinal)', 'first of all (sentential adv.)' etc.
Other matters, like compound verbs, double accusatives etc. are or ought to
be discussed in the Syntax portion, which I haven't worked on for a long
time-- so links to that too. Ditto the as yet unwritten "Colloquial"
section.

More work for the winter season............:-))