Attributing textual layers of the Tanakh

[Aharon Varady]

Efraim and friends,

Using JLPTEI in app.opensiddur.org -- how might one attribute textual layers in the Tanakh?

I've been reading through Tzemaḥ Yoreh's chumash in which portions of the text are attributed to different unknown authors according to his supplementary hypothesis. There are seven of these layers, the earliest dating from the mid- to late eighth century BCE (the "Elohistic" text), and the last dating sometime after the end of the Babylonian exile, post 5th century BCE (the "Bridger" text). Yoreh helpfully highlights all the texts in the chumash indicating its attribution according to his theory.

I was wondering how one might use app.opensiddur.org to indicate possible attribution of unidentified material in the database.

While Yoreh's color-coding represents a kind of primitive visual metadata in the manner of R' Jacob Freedman's Polychrome Historical Haggadah and Prayerbook, once the chumash is tagged with this data it would be interesting to show different ways the text could be displayed. I'm attaching a PDF with a few pages from his work as an example. (Tzemaḥ is open about sharing his work with an Open Content license.) The example shows Numbers 21 from Yoreh's Canon v.13, a text with 4 different textual layers.

Aharon

--

Aharon Varady
the Open Siddur Project
http://opensiddur.org

[Efraim Feinstein]

Hi,

Interesting question, and we may need to do some experimentation on what the best way to handle this is:

1. Treat it like a regular annotation: make 1 (textual) annotation for each kind of metadata. Annotate each section of text as normal. Use the same annotation link for each source (so, you have 1 annotation file, say TzemahYorehsSources with 7 annotations, one for each layer, and you keep linking to the same 7 annotations). This is totally doable in app.opensiddur.org as it stands today.

2. Treat it like a setting: Basically, you make a conditional feature structure with 7 features, one for each layer (This part is only possible in the XML editor, though it is scheduled for 2 releases from now). You then either (a) "set" the text to be associated with a particular layer or (b) "condition" the text on a layer as if it were a conditional. This will be possible in the next version of the app, due to be released very soon (coding is done, needs some more testing). If you do #2b, you could then generate a text that contains only one of the layers using the conditional processing mechanism.

In either case, the coloring *may* be possible with some fancy custom CSS (also possible in the app, though it will require you to go out of the visual editor to actually write the CSS). 

Two notes: 

1. The granularity of the settings or annotations is the segment.

2. All the structure of Tanakh is not currently supported in the visual app. As of now, the visual app just supports two elements of structure: paragraphs and segments.

-Efraim

--
You received this message because you are subscribed to the Google Groups "opensiddur-tech" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opensiddur-te...@googlegroups.com.
To post to this group, send email to opensid...@googlegroups.com.
Visit this group at https://groups.google.com/group/opensiddur-tech.
For more options, visit https://groups.google.com/d/optout.

[David Troidl]

Hi,

I'm not sure how this would fit in with your Tanach, but the WLC on tanach.us has color coding for the 'documentary hypothesis'.  I'm also not sure how the sources would correspond.  If you go to: http://www.tanach.us/Tanach.xml
and enter Numbers 21 in the search box, then choose DH: All, in the upper right corner, you can see the example.  (Unfortunately, the site doesn't allow copying the URL for the display I'm seeing.)

Peace,

David

Virus-free. www.avast.com

[Aharon Varady]

From the data shared by Tzemah Yorah, I made a spreadsheet indicating the source to attribute for each "kernel" -- each textual unit comprising Genesis, and for each range of verses attributable to a particular source.

By attribution, I mean both authorship and altering/supplementing a verse.

So for example:

The textual unit "In the Beginning 1," the kernel of which is "J" the verse range "chapter 1 verse 1 through chapter 2 verse 4" is attributed to "P."

also

The textual unit "In the Beginning 1," the kernel of which is "J" the verse range "chapter 2 verse 4 through 9" is attributed to "J."

and so on. Take a look.

I think this should suffice for automatically generating annotation, except in cases as with Chapter 2, verse 2 where there is more than one author attributed to portions of the verse. Such cases might have to be tackled manually.

Am I on the right track, Efraim? What more would I need to do next in order to generate the proper annotation needed for integration into our DB?

Aharon

P.S. I added a column for counting the number of verses attributed to a particular source, for some fun quantitative analysis, e.g. percentage of verses attributed to any particular source in Genesis.

Aharon Varady, M.C.P., M.A.J.Ed.
Community Planner, Educator

[Efraim Feinstein]

Hi,

Since you have a spreadsheet, you can do this with code, so that means you're on the right track.

1. You need to decide which of the methods from the prior reply you want to use (annotations or conditionals). That will determine which kind of layers you want to add.

2. You need to decide whether you'll be (a) adding layers to the existing Tanach text (in which case, you definitely need to use code because it is auto-generated and re-generated every time there's a server upgrade) or (b) work in a separate set of texts that incorporates the existing Tanach text by reference.

3. You need to write the annotation or conditional documents for what you want to record (I guess that's "kernel" and "touched by").

4. You need to write the layers for #2a or #2b -- since you have a spreadsheet, either way, code is the best way to do it.

Having more than one annotation overlapping is handled perfectly fine, though if you use the conditionals method, it will require some special case handling.

Once you make decisions, we can go into more detail.

-Efraim