Primary vs. Secondary Resources

In this forum, we will work with people who are training OCR engines for Polytonic Greek, helping them improve the quality of their engines by correcting specific texts.
Post Reply
Louis L Sorenson
Posts: 706
Joined: May 5th, 2011, 9:21 pm
Location: Burnsville, MN, USA
Contact:

Primary vs. Secondary Resources

Post by Louis L Sorenson » May 18th, 2014, 12:19 am

In addition to being able to OCR Greek texts (many are publicly and freely available) , there are many Greek resources that can be digitized that will build upon the scholarship of former generations who had a much greater insight and skill than many of us today. I'm thinking of works like Schlenkl's volume on Epictetus; In it, he includes a 200+ page index of Greek words and constructions. It is that sort of scholarship that is hard to ferret out and link to the actual text of Epictetus -- it would take a student a long time to page through a volume and search through Epictetus' constructions. Schenkl has in depth explicitly described Epictetus' phraseology, and having to start over and do a raw search on Epictetus' text to find the patterns that Schenkl has already 'discovered' seems like a lot of extra work. We cannot just assume that modern vocabulary/grammar/syntax searches will give the same results as the life achievements and contributions of a former expert on any given subject/author who meticulously worked through the text of a Greek author and created lists of syntax structures from that author.

So as ancient Greek texts are becoming publicly available, I see the value of Greek polytonic OCR not in reproducing texts of the original authors the most valuable, but rather in being able to access digitally secondary literature.
0 x



Jonathan Robie
Posts: 3364
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Primary vs. Secondary Resources

Post by Jonathan Robie » May 18th, 2014, 7:39 am

Obviously, we need both. Bruce is the guy who has been digitizing a lot of things from our wishlists, which include both primary and secondary literature. We need OCR that can do both well.
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

Jonathan Robie
Posts: 3364
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Primary vs. Secondary Resources

Post by Jonathan Robie » May 19th, 2014, 11:05 am

Suppose we were to pick one resource from secondary literature. What would we pick?

Bruce can already do good OCR on Meyer's commentaries, the Expositor's Greek Testament is rougher. Would that be a good secondary resource to work on? I suggest we stick with one primary and one secondary resource for now.
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

BruceRobertson
Posts: 22
Joined: May 16th, 2014, 9:49 pm

Re: Primary vs. Secondary Resources

Post by BruceRobertson » May 22nd, 2014, 8:40 pm

Did you mean Schenkl's Dissertationes ab Arriano digestae? I ran it here:
http://heml.mta.ca/lace/sidebysideview2/11956040 from a rather poor copy. I'm interested in other desiderata, both within and outside biblical materials, since it's better that we focus on what people need, rather than specifically what is easy (though there has to be a balance).
0 x

Louis L Sorenson
Posts: 706
Joined: May 5th, 2011, 9:21 pm
Location: Burnsville, MN, USA
Contact:

Re: Primary vs. Secondary Resources

Post by Louis L Sorenson » May 24th, 2014, 5:57 am

Yes, that is the work. It's the appendices that are valuable. Boter's 2007 (Teubner) edition supercedes the textual notes of Schenkl.

The pdf at LetsReadGreek (11 meg) is a much better scan that the archive.org version of Schenkl's 1916 Appendixes http://www.letsreadgreek.com/epictetus/ ... ndixes.pdf. The appendices are very useful for finding words used in phrases, word constructions, etc. The appendixes run from pp. 497-740. The appendixes contain an index of 1) names (Nominum), 2) words (Verborum – 212 pages worth!), 3) references to other authors (Locorum) and 4) scholia. The non-Greek text is written in Latin (the page count and appendix comes from the 1st 1894 edition).

Epictetus (transcribed by Arrian) is written in literary Koine - the vernacular of the common folk, though literary, not the mimicking Attic of the 2nd Sophistic. Arrian also wrote works using Attic (Anabasis Alexandri (Greek: Ἀλεξάνδρου ἀνάβασις Alexándrou anábasis), the Campaigns of Alexander by Arrian, which is the most important source on Alexander the Great). He is one of the few authors (if not the only one) who wrote books in both 'dialects' if that is the right term.

I would think that Stephanus' lexicon (links are available at http://charlesasullivan.com/2179/ancien ... tionaries/ would be a very good project to digitize. The link talks about the value of that lexicon.
0 x

BruceRobertson
Posts: 22
Joined: May 16th, 2014, 9:49 pm

Re: Primary vs. Secondary Resources

Post by BruceRobertson » May 24th, 2014, 8:36 am

Thanks for these suggestions!
0 x

Jonathan Robie
Posts: 3364
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Primary vs. Secondary Resources

Post by Jonathan Robie » May 24th, 2014, 9:02 am

I think we're basically fishing around for one secondary resource in addition to Migne? Let's pick one that at least several people are enthusiastic about, for which there is a really good scan.

Stephanus seems plausible, so does Sophocles.
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

BruceRobertson
Posts: 22
Joined: May 16th, 2014, 9:49 pm

Re: Primary vs. Secondary Resources

Post by BruceRobertson » May 25th, 2014, 5:16 pm

I'm happy to recognize anything anyone wants, using existing classifiers, so requests on that level are quite welcomed. We do, however, want to focus our training labour here on just a couple of texts.

Sophocles we have already run from an archive.org copy. Here is the OCR output that we have:
http://heml.mta.ca/lace/render_page?hocrtype_id=6835

I think, though, that the copy provided in pdf at http://charlesasullivan.com/2179/ancien ... tionaries/ is clearer.

The weakness in all of these mixed-text (Greek and Latin-script) is in the Latin-script italic font. Because there are fewer examples of this in a random sample, it tends not to get as well trained, and thus it gets confused with Greek.

Stephanus is very much like Migne: I think those who are working on training for Migne can rest assured that they are also contributing significantly toward Stephanus. The problem is that the Stephanus scans from Google books are pretty low-rez. and blotchy. Perhaps the OCR fairy will visit and tuck a USB stick under my pillow :-)
0 x

Post Reply