tlhIngan-Hol Archive: Fri Sep 26 09:39:30 2014
[Date Prev][Date Next][Thread Prev][Thread Next]
Re: [Tlhingan-hol] Sorting
Very clever use of Excel's character functions. Impressive and effective.
I considered the issue you bring up about using "z" instead of a number, but I realized that if I ALWAYS notate EVERY "n" as "n1" and EVERY "ng" as "n2", then there is never an instance in which it matters whether anything else sorts before or after numerals, so long as "1" sorts before "2", since "n" is never followed by anything other than a "1" or a "2" in this field.
Basically, so long as we notate things properly, your system works and my system works. My system has the weakness that I could have a Romanized entry that is correct, but I could fail to notate the sort correctly. Meanwhile, I can tell you that after I've done over 2,500 of these records, I've gotten pretty good at it. Whenever typing a "c", "g", "n", "t" or apostrophe into the sorting field, a little voice in my head sounds, "ah-OOOOO-gah! ah-OOOO-gah!" and the character and those following it get my full attention.
My sort of your example would be [biq2 gaywIz paz].
I don't maintain separate K-E and E-K entries. It's less maintenance and yields fewer opportunities for error to have every Klingon word have only one entry. Of course, that brings up arbitrary decisions about what qualifies as a word. I can find either an entry by word or definition with the search function. I haven't used paper in a long time, not out of any disrespect for it. It just is harder to carry around than an iPad or an iPod Touch, and I can't write as fast as I can type.
Using a database app, the search simply finds what I type in the search field in any field of the database. I find it useful to search by the sort field contents when I'm looking for Klingon entries because q1 and q2 don't pull up as many false positives as "q" and "Q", since the search ignores case.
I do notate derivative words with a checkmark so that if someone asks how many words there are in Klingon, I can give a more accurate (though still arbitrary) number by counting entries that don't have that check mark.
Words that reappear with slightly different definitions from different sources take some notation tricks to keep the various definitions and sources separated. I tend to put the definitions in order of discovery, separated by slashes, and I do the same in the Source field, so I can tell which definition comes from which source.
Animals have a check box indicating that they are an animal (so I can pull up all the animal words without relying on consistency in the wording of the definition), and the rest goes into the definition.
I have a lot of respect for the work you've put into this and the usefulness of the tool you've developed. We likely both deserve medals for our dedication to our geeky task far above and beyond the bounds of normal human behavior.
We have boldly gone where no one (except perhaps a few other people on this list) have gone before. If we were stonemasons, these dictionaries would be cathedrals.
On Sep 26, 2014, at 10:45 AM, SuStel <[email protected]> wrote:
> On 9/26/2014 10:02 AM, [email protected] wrote:
>
>> The sort field replaces some characters with others in order to
>> force the database to be sorted according to the rules that Okrand
>> uses in TKD:
>
> I did mine in Excel (which let me port it to Word for formatting and printing). I created a field that automatically transformed the Klingon word to make it English-sortable. If A2 contains the Klingon word, then this is the formula for the Sort Order column on row 2:
>
> =SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(A2,"ch","cz"),"gh","gz"),"ng","nz"),"Q","Qz"),"tlh","tz"),"'","z")
>
> I used z instead of a number or symbol because some alphabetizing systems might swap whether those come before or after letters. Likewise, I transformed the glottal stop to a z.
>
> So a term like {bIQ ghaywI' pa'} has a Sort Order value of "bIQz gzaywIz paz." Sort the spreadsheet by the Sort Order field, and you're done!
>
> For best use of space in a digest-sized printout, my dictionary's pages have two columns. This is here illustrated using the words {bep} and {boq}, using HTML-style markup to show how it's formatted (wrapped lines have a 1-em indentation starting with the second line):
>
> K-E
> <b>bep</b> <i>n.</i> agony<br>
> <b>bep</b> <i>v.</i> complain, object, gripe<br>
> <b>boq</b> <i>n.</i> alliance, bloc, coalition<br>
> <b>boq</b> <i>v.</i> ally with, form an alliance with; also used in math for addition<br>
>
> E-K
> agony <i>n.</i> <b>bep</b><br>
> alliance <i>n.</i> <b>boq</b><br>
> ally with, form an alliance with <i>v.</i> <b>boq</b><br>
> complain <i>v.</i> <b>bep</b><br>
> gripe <i>v.</i> <b>bep</b><br>
> object <i>v.</i> <b>bep</b><br>
>
> Derivative words that appear in lexical form by Okrand, e.g., {boqHa'}, {boq'egh}, have their own entries.
>
> Most of this formatting was done simply by importing from Excel to Word, with a bit of finding and replacing to abbreviate parts of speech and changing styles (italics, bold). In Word I can therefore also have word guides at the tops of the pages, which are very helpful.
>
> The whole, completely updated dictionary, with both E-K and K-E, is still small enough to print as a digest-size booklet and staple in the middle.
>
> Most of the effort went into typing all the data into Excel, including various synonyms like "complain," "gripe," and "object." Certain categories of words required special treatment. For instance, there are a ton of animals, some of which aren't described. Here's how I did those, using "animal," "bird," {Qogh}, and {cha'bIp} as examples:
>
> K-E
> <b>bo'Degh</b> <i>n.</i> bird (general term)<br>
> <b>cha'bIp</b> <i>n.</i> type of animal, a bird noted for its speed<br>
> <b>Ha'DIbaH</b> <i>n.</i> animal, meat<br>
> <b>Qogh</b> <i>n.</i> krog, type of animal<br>
>
> E-K
> animal <i>n.</i> <b>Ha'DIbaH</b><br>
> animal, list of (see also bird, list of) <i>n.</i> <b>boqrat, chemvaH, jajlo' Qa', 'er, 'IghvaH, Hun, Qa'Hom, Qaj, qa'raj, QIncha', Qogh, qa'vam, lIngta', lung, la'SIv, pIpyuS, mIl'oD, Sargh, Suy', tangqa', teghbat, tI'qa' vIghro', tIqnagh, to'baj, topvIr lung, toppa', vIghro'</b><br>
> bird (general term) <i>n.</i> <b>bo'Degh</b><br>
> bird, list of <i>n.</i> <b>borghel, cha'bIp, cha'Do', cha'qu', cha'naS, cha'par, Da'nal, Da'vI', jentu'</b><br>
> chabip, type of bird <i>n.</i> <b>cha'bIp</b><br>
> krog, type of animal <i>n.</i> <b>Qogh</b><br>
>
> --
> SuStel
> http://www.trimboli.name/
>
> _______________________________________________
> Tlhingan-hol mailing list
> [email protected]
> http://mail.kli.org/mailman/listinfo/tlhingan-hol
_______________________________________________
Tlhingan-hol mailing list
[email protected]
http://mail.kli.org/mailman/listinfo/tlhingan-hol