Bug: Info panel -- translation section display

Reuben Helmuth said:

No matter which (RI) English that one uses, the translation section will always display [variant reading] for the Lexham English Septuagint (LES).

Reuben Helmuth said:

When using the LES, EVERY other version, including the LXX's, simply display [variant reading].

It's by design, but the consequences of the design may not be obvious at first glance.

The way this section works is it takes your selection in the surface text and maps it to original language words using either the words that you selected (if you're in an original language text already) or through the mediation of a reverse interlinear. It then opens all your reverse interlinear Bibles that contain that verse reference and checks them for the same words. If the words can't be found, then "[variant reading]" is shown.

(It really doesn't map back to words, but unique word instance identifiers called "word numbers." We add these identifiers to select original language texts, and the reverse interlinears use them to stand in for the surface words, which may have various variations in form or spelling in any given version/edition of an original language text, but still be the same "word".)

The reason you always see "[variant reading]" for the LES is that all the other translations translate from the Hebrew (at least ostensibly), but the LES translates from the Greek as its "original". So if you select something in ESV, then it maps back to Hebrew words, which can be found in every other translation except LES, which reads a different "variant." (We chose "variant reading" is the English equivalent of "varia lectio" which is often what is used in commentaries and apparatuses, abbreviated "v.l." to indicate that the fore-text in question doesn't match what you're looking at because it reads a different fore-text here.)

Now, if you then select something in LES, it maps back to Greek LXX words, which can only be found in LEX (and possibly Greek versions of the LXX, about which below). I don't know why your middle screen shot doesn't show anything for LXX Swete and Logos LXX; which verse you're in may make the difference. Or it may be a bug.

If then you select something in a word-numbered Greek LXX, like LXX Swete, we have two sets of word numbers, because LXX Swete and Logos LXX are themselves reverse interlinears back to the Hebrew! So when you start from this text, it's possible to match up the Greek LXX "original" for LES, and also the Hebrew behind the other translations. (This is a happy but coincidental outcome that had you asked me without presenting a screen shot I might not have correctly predicted.)

We could, but don't, open the LXX reverse interlinear to convert the Hebrew word numbers gathered by selecting text in ESV into LXX word numbers to get the translation from LES. That is, ESV -> Hebrew -> LXX -> LES. And vice versa. But we only do one reverse interlinear hop as it is, not two.

Reuben Helmuth said:

I also consider it a bug that my prioritized bible (ESV) doesn't stay in the first (top) position. When in agreement with other versions, it does lead the list, but it should stay at the top of the window.

Also by design: The translation options are sorted by frequency, so the most common translation is listed first. 

Reuben Helmuth said:

Eli Evans said:

We could, but don't, open the LXX reverse interlinear to convert the Hebrew word numbers gathered by selecting text in ESV into LXX word numbers to get the translation from LES. That is, ESV -> Hebrew -> LXX -> LES. And vice versa. But we only do one reverse interlinear hop as it is, not two.

Would it be a mountain or a molehill to add code to make the second hop?😆 In my mind it would make the feature much more useful. This is especially true when working in LES! The other option would be to simply remove LES from the window entirely and stop distracting.

I'll add it as a suggestion.

Reuben Helmuth said:

BUG: With the info panel open, when selecting any occurrence of 'month' in the LES which is translating the Gk word μηνὸς, Logos crashes immediately and with no error message.

Reported. On PC I don't crash, but I do get this:

Reuben Helmuth said:Question: Why is it that in the books of Tobit, Susanna, and Bel & the Dragon, only NRSV and LES appear (as expected) but, one or the other is always shown as a variant reading, even when identical wording is used. To me it seems like it must me encoded as from two different LXX's(?).
I'm no expert, so I don't know the whole story, but they are definitely translated from different underlying texts: different editions of different manuscripts. Here's Bel and the Dragon, for example. The LES follows the Swete edition of Chisianus, and the NRSV looks like Rahlf's edition of Theodotion. That is, NRSV mentions King Astyages in verse 1, and LES has no verse 1, but a title that says this story from a prophecy by Habakkuk. 

It looks like the texts diverge enough that they were given different word number identifiers under the hood, even where there is some possible alignment. (I wasn't privy to the editorial conversations, so I don't know for certain all the details.)