Koine Greek Software libs?

[JoshMuller]

Hey guys!

So, I've been throwing around ideas of using the LaTeX typesetting system to put together and print some Greek tools for myself.
One of the ideas I've had is to shoot off of the Reader's Greek New Testament idea of having footnote vocab for less common words (maybe even with parsing). (Seriously, how cool would it be to have a Reader's Ψαλμοι, or LXX or something like that, which you've made yourself. Or A Reader's Apocrypha? That's kinda what I'm thinking )

I've been playing with various ideas, and realized the biggest bottleneck of a potential project like this teaching a script to recognize the lexical form of a word, and have it use that to cross reference a lexicon. I'm still pretty fresh as far as software development and stuff is concerned, so that seems like a fairly significant task to me . It wouldn't be that hard to set a script up recognizing and reacting to basic parsing concepts, but when you get into irregular words, words that don't decline, and all that fun stuff, I expect it to get really messy.


So, I was wondering, does anyone know if there are any software tool out there that would lend itself to a project like this? Like, a software library that would have parsing classes, or vocab classes like that available? Or even any tips for developing such a project?


ευχαριστω!
Χαρις υμιν και ειρηνη,
Josh

[Jonathan Robie]

New Testament with parsing is available from the biblical humanities github, so is the Abbott-Smith lexicon. For the Septuagint, the CATSS text is available on the Internet, copyright is fuzzy, we are trying to get a parsed version of a Septuagint that is freely licensed but do not have one. So this is easier to do for the GNT, you may be able to clarify copyright issues with CATSS or wait for a freely licensed Septuagint to happen.

[jtauber]

A number of projects linked to from http://jktauber.com/ are relevant to this topic. In particular see https://github.com/jtauber/greek-reader for a LaTeX-based system for readers with footnoted vocab.

[Nigel Chapman]

Hi Josh,

I previously used a millon-word database (from the Perseus dataset) to brute-force the recognition of Greek words in user-entered texts (up to about 95% success), but I'll be attempting this year to write a Greek indexer that knows enough parsing and grammar to contruct words from their stems up, and disassemble them as well, along with lists for basics like participles/artcles/etc, and longer fallback lists for irregular forms. There are only 20ish declensions of nouns, for example, and they just have to be worked through. Verbs will be the test of it, obviously. I've found that there are a few useful shortcuts in this process, like taking grammatical hints from context (say, agreement in case or number).

This will be a part of my Koinos PHP library. --> https://github.com/eukras/koinos

Koinos runs a site of mine, http://hexap.la, which works by ripping a library of Markdown-formatted texts into an index of words I'm aiming to include Apostolic Fathers and Koine background texts next, but theres a bit of interface work that needs to happen first. Once Hexapla can parse arbitrary texts, making that parsed data public is easy. That's one of the end-games.

If you know any PHP developers with an interest in this, just point them my way.

Thanks,

Nigel.

[Devenios Doulenios]

Hi Josh and Nigel,

Do you anticipate that this set of tools would be available for use at some point on an iPad? I had used the Aleiphos (sp?) tool earlier with Firefox on my desktop system but since they went to a subscription basis for iOS I no longer get to use Firefox (no budget for it, alas). A shame, because both FF and Aleiphos are great tools.

Just to clarify, I would be a potential end-user. I don't know how to do any programming or scripting myself.

Devenios Doulenios

Δεβένιος Δουλένιος

[Nigel Chapman]

Hi Devenios,

Well... my stuff is web based, so it should run on anything.

Nigel.

[Devenios Doulenios]

Sounds great, Nigel. Please do keep us posted on your progress.

Δεβἐνιος

[caseyperkins]

I've been playing with various ideas, and realized the biggest bottleneck of a potential project like this teaching a script to recognize the lexical form of a word, and have it use that to cross reference a lexicon. I'm still pretty fresh as far as software development and stuff is concerned, so that seems like a fairly significant task to me

Hi Josh,
I can attest that it is indeed a fairly significant task, having done it!

My iPhone app GreekKit does much of what you are talking about. (Some of the other stuff is in the works). It knows the location of every word in the NA28 and can identify the correct lexical form for any word form found in that GNT edition.

I hesitated to make this my first post (after having join two or three weeks ago), but it seems relevant to this thread, and I did build the app to help people with just such tasks as you mentioned. I hope you or someone else here will find it useful.

Casey