Page 1 of 1

Is there a ready way to mark NT words in other texts?

Posted: February 10th, 2017, 2:20 pm
by Stephen Hughes
Is there a ready way that I can cut and paste a Greek text into a dialogue box in a website, and it retruns for me New Testament words in one colour and non-New Testament words in another?

For example, if I were to enter this snippet from when Daphnis family recognised their long lost brother:
Longus 4.22.1 wrote:Ἔτι λεγούσης αὐτῆς καὶ τοῦ Διονυσοφάνους τὰ γνωρίσματα φιλοῦντος καὶ ὑπὸ περιττῆς ἡδονῆς δακρύοντος ὁ Ἀστύλος συνεὶς ὡς ἀδελφός ἐστι, ῥίψας θοιμάτιον ἔθει κατὰ τοῦ παραδείσου, πρῶτος τὸν Δάφνιν φιλῆσαι θέλων.

She (Cleariste) was still speaking, and Dionysophanes was kissing the tokens and crying from too much joy, when Astylos, realizing that he was a brother, threw off his cloak and ran down through the garden, wishing to be the first to kiss Daphnis.
It would return:
Longus 4.22.1 (NT words in black, non-NT words in green) wrote:Ἔτι λεγούσης αὐτῆς καὶ τοῦ Διονυσοφάνους τὰ γνωρίσματα φιλοῦντος καὶ ὑπὸ περιττῆς ἡδονῆς δακρύοντος ὁ Ἀστύλος συνεὶς ὡς ἀδελφός ἐστι, ῥίψας θοιμάτιον ἔθει κατὰ τοῦ παραδείσου, πρῶτος τὸν Δάφνιν φιλῆσαι θέλων.
Of course while γνώρισμα "token (as in dog-tag)" is not in the New Testament, but γνωρίζω is. The form ἔθει might be misunderstood as a dative singular of ἔθος "custom", "habit", rather than from θέω "run".

I'm thinking in terms of a client side site like TextMechanic. Other ideas for the site might be text- tagging, reduction to dictionary forms, removal of diacritics, ...

Re: Is there a ready way to mark NT words in other texts?

Posted: February 10th, 2017, 4:28 pm
by Jonathan Robie
I don't know a tool like that.

Re: Is there a ready way to mark NT words in other texts?

Posted: February 10th, 2017, 11:17 pm
by Stephen Hughes
Jonathan Robbie wrote:I don't know a tool like that.
Is there a tool that I can create an intelligent tree in the same format as your NT trees as I read? I imagine that it could have AI capabilities to parse everything according to the possibilities, then for each verse; 1) ask a user to choose between the viable possibilities (such as ἔθει means "he was running", rather than "by habit"), and 2) ask the user to identify the referent of any pronouns (such as αὐτῆς refers to Κλεαρίστη Longus 4.20.2), which I presume would be enough human intelligent input to make the trees to a standard similar to what you have now.

I don't think that a couple of "Clarify this for me..." questions for every 2 dozen words of Greek would be very burdensome to a reader. If the resulting trees were licensed under a CC licence, and could be accessed by others, they could be used and modified by the same people that use NT trees. That would be an easy way to add to the body of texts encoded into trees.

Re: Is there a ready way to mark NT words in other texts?

Posted: February 11th, 2017, 11:25 am
by Stephen Hughes
Does anybody else have suggestions for useful text-tools?

Re: Is there a ready way to mark NT words in other texts?

Posted: February 11th, 2017, 12:24 pm
by Jonathan Robie
Stephen Hughes wrote:
Jonathan Robbie wrote:I don't know a tool like that.
Is there a tool that I can create an intelligent tree in the same format as your NT trees as I read? I imagine that it could have AI capabilities to parse everything according to the possibilities, then for each verse; 1) ask a user to choose between the viable possibilities (such as ἔθει means "he was running", rather than "by habit"), and 2) ask the user to identify the referent of any pronouns (such as αὐτῆς refers to Κλεαρίστη Longus 4.20.2), which I presume would be enough human intelligent input to make the trees to a standard similar to what you have now.

I don't think that a couple of "Clarify this for me..." questions for every 2 dozen words of Greek would be very burdensome to a reader. If the resulting trees were licensed under a CC licence, and could be accessed by others, they could be used and modified by the same people that use NT trees. That would be an easy way to add to the body of texts encoded into trees.
There will be. I'm working my way up through a sequence of steps.
  • A website that makes it easy to see the trees for any chapter, with other reading aids. This should be up fairly soon (weeks?)
  • A query facility on the same website that lets you do queries like the ones I've been doing, with some sample results.
  • A feedback mechanism for discussing and suggesting alternate interpretations. At first, this will probably stop short of a full syntax tree editor.
Right now, my software development is focused on these goals. I also want to write software that presents the kinds of lessons I am currently teaching in person, using a living languages approach that is closely tied to the text. I haven't quite decided how much of this to do as a website vs. app.

Encoding other texts comes after that. I have a pretty good idea how to write the editor for this, and how to combine it with parsing, but it's a chunk of work. I'm taking significant time off from work to do these things, but I probably couldn't do that unless I find some source of funding. But I think the tree display format I'm using is simple enough that people could learn to edit texts with it fairly easily, and the finer-grained syntactic stuff could largely be filled in automatically. If someone else does it before I do, let me know! If someone wants to fund this ...

Re: Is there a ready way to mark NT words in other texts?

Posted: March 12th, 2017, 3:41 pm
by MAubrey
Jonathan Robie wrote: If someone wants to fund this ...
Have you considered a GoFundMe campaign?

And once you have an established platform, Patreon could be a feasible means for *some* means ongoing financial support.