Re: Lexicon storage methodology (was: Lexicon counting...)

Iain E. Davis wrote:
> Embarassingly, I mis-keyed, and inadvertantly sent this before I was
> finished. Ignore the previous version, or laugh at it, as you like. :)
>
    I did it a few times myself, too. I can't blame you ; ).
> -----Original Message-----
> From: Iain E. Davis [mailto:feaelin@kemenel.org]
> Sent: Wednesday, 2006 September 06 8:38 PM
> To: 'Constructed Languages List'
> Subject: RE: [CONLANG] Lexicon counting (was: Weekly Vocab #1.1.1...)
>
> Edgard wrote:
>
>> I used Excel as well, but I'm using the Openoffice equivalent for the
>> Unicode compatibility. Then I save it as a CSV file,
>>
> Hmm. To what degree did Excel not do what you needed unicode wise?
> I was able to come up with what I needed for IPA symbols, which was my main
> concern. I was happy from there. :)
>
> Hm. Was it sorting related? I had to rig some things up to get sorting the
> way I wanted. :)
>
> Not that I think you should change. If OpenOffice works successfully,
> awesome!
>
Excel opens the file more easily. But when I save an Unicode table as
.CSV, and the I open it again, every accent is messed up. What I'm
thinking to do is to edit it in transcription... easier than adding
macros... and writing another function for converting it back to
Unicode. But I'm playing with an online editor for the dictionary, then
I could forget other programs. But they are useful for mass editing and
that sort of thing.
>
>> and open it with PHP. That part is quite easy. Then I show it through
>> PHP, and I'm now fighting against inflection ; ).
>> Here is the result so far:
>> http://ausonia.parnassum.org/dicionario.tudo.php
>>
>
> Pretty. I've not anything so well constructed, web-wise. :)
>
    Thanks! It's not that well constructed, I just know how to use the
plaster, ehehehe. The code is chaos, I almost see the primeval deities
while editing it.
>
>
>> But I just realized that I can't reverse the entries. I guess I should
>> have a English dictionary, for instance, and link my Ausonian entries
>> to the English entries? Ouch.
>>
>
> The method I used for achieve that effect doesn't produce a
> English->Taraitola dictionary, or in your case an English->Ausonian one.
>
> What I did is one of my columns (in addition to the _definition_ column) is
> "English Word" column. This is the closest equivalent english word, if any.
> Then for the English->Taraitola "Cross-Index" a entries looks like:
>
> I: äm
> You: säm
>
> And my expectation is that you look up äm and säm in the Taraitola->English
> dictionary, to be sure that you're using a word with the correct flavor.
>
> One thing I've never gotten around to fixing is output where two Taraitola
> words map to the same english word...this generates two entries on the
> English->Taraitola side.
>
> I also appear to be repeating myself. I think its my week for redundancy.
>
You mean, even if two Taraitola words go to the same English one, each
output is counted as a separated entry? Hmm... you could do first an
Taraitola > English table/matrix, and then collapse the entries with the
same fields:

Taraitola - English
blabla1 - blabla2
blabla3 - blabla2

to

English - Taraitola
blabla2 - blabla1, blabla3
>
>> Could you not be more specific in the entries? Or it is a word-to-word
>> association?
>>
>
> Perhaps. In at least the case of a trio of words that all mean 'love', there
> really aren't any other words that match up at all. Since my main intent is
> to give an english speaker a way to find the right words, it serves its
> purpose. :)
>
I thought using more words to describe it. But I understood now.

[cut]

>
>> Someday, hopefully, I will pass the thousand frontier ; ).
>>
>
> Depending on how you count it, I'm either looming close, or still have three
> hundred or so to go to hit that mark. :). After I tackle Swadesh's list....?
>
Well, if I count every inflected form, I'm very rich already, heheehe ; ).
>
>
>> What about giving for each word/entry an example phrase?
>> Specially for verbs, it would be useful for showing the
>> prepositions... the case of the object, and for exercising the
>> fluency, too...
>>
>
> Like many dictionaries do, in fact. Sadly, I've not done so, but it would
> make sense to do so. Much like the 2nd weekly vocab has 1. Mushroom and then
> "I found mushrooms in the forest". As you say, this is especially useful for
> verbs, practice, and the specificity of prepositions.
>
> I currently don't have such an animal, but maybe I should put it on my
> horizon. :)
>
Such an animal?
> Hm. My columns are: Taraitola Word, Pronounciation, Part of Speech,
> definition, English Word, A or C (regular word, or proper name), and the
> ever-arbitrary "Tag" column. And then some other, "calculated" columns,
> primarily needed for the 'dictionary generator' VBA script.
>
Mine are: form, class, derivation, conjugation, aspect (of the verbal
root), genre, declension type, portuguese meaning, english meaning,
sureness ; ), revision, Pokorny (as it is an indo-european language),
laryngeals, latin, greek, and sanskrit cognates, comments, etc.

My most recent problem is that some verbal roots take another roots for
the lacking aspects... I can add another column for that, or add this
comment on each entry, or give each stem aspect for each verb. Or I can
forget that details and create myself, but I am so happy when my verbs
look like the vedic ones ; ). Anohow, that will hurt, I can feel already ; )
> Iain.
>
>
Edgard.

Reply