Best practice for interlinearising glossary?

Question

We have a small glossary (currently 1 page), and we are using the interlineariser to provide back translations.

To split it up for the print-out for a recent consultant check, and for ease of navigation, I went through it and added verse numbers, with \vp to replace the number with a letter on publication.
e.g.

\v 2 \vp B\vp*
\k Barabbas\k* ....

We’ve now discovered that (at least in pt9.1 and 9.2) the updated text can no longer be accessed by the interlineariser tool (it recognises the verses as divisions and presents everything up to the first \v, but refuses to move from one “verse” to another).

Interlinearising the entire glossary as if it were a single verse is a complete pain, as it’s not obvious where the text was changed, and scroll-together is lost.
Also it becomes impossible to approve part of the back-translation.

I see some old topics about using chapters to divide the glossary, but this seems to similarly stop the interlineariser dead in its tracks.

Is there any way to segment the glossary that the interlineariser does accept? What’s the best practice for this?

Answer 1

Presently using PT 9.2.

Just noticed that our very large Glossary does not let me interlinearize anything. We have a structure with 34 chapters and many verses in each chapter.

Seems other users have found the same problem.

We need back translation for quality control please and the interlinearizer is the best tool to start that. We have a second stage, where we collect the results from the interlin. And we have a third stage, where we clean up the results from the interlinearizer into a free translation. All this is very clumsy without the interlinearizer and its database.

Good news: Just noticed that we can interlinearize in XXC, which we use in one project for the entire “Jesus Messiah” comic book from the NL. We have created a very custom structure of 12+ dummy chapters and some custom markers (and no verses) to best align with the structures that are provided by the publishers in the NL.

And still PT 9.2 is able(!) to provide the interlinearizer! So if it can be done in XXC, there is technical hope that it can be done (unlocked) in the GLO glossary.

Comments

Here are a couple of observations. It is possible to open the GLO in the interlinearizer. You can add GLO to the back translation by going to Manage Books > Add Books. However, when you interlinearize a book it should get added automatically.
However, I noticed that when I have verse numbers in GLO, Paratext does not recognize these in the interlinearizer. When there are no verse numbers, it seems to work, but makes a huge file.

---

The only way we’ve managed to do a back translation was either:
Temporarily use PT8 (if I remember correctly) or Remove all the chapters / verses (which means that you get just 1 check-box for the whole text!)

Needless to say, neither of these should be considered best practice!

There must be a better way , surely?

---

Indeed, I had the same problem.
I noticed that the GLO book made Paratext very slow, though I have a reasonably fast laptop (i5 10th gen, 16 GB RAM). I wondered whether this was because I had the spell check on and the interlinearizer open, but turning the spell check off and closing the interlinearizer did not help. On the translator’s computer (which is not as fast) the glossary was practically unworkable - edits took minutes of waiting - and in the end I copied the whole text to MS Word, moved my notes into the Word file as well (aligning them in table columns) and we’ll edit it there.
At one point I wondered if it would help to segment the glossary into verses, but as anon848905 noticed, the interlinearizer did not recognize them. Besides, I imagine that versification, even when it’s applied only temporarily, may mess up the linkage to Biblical terms.
Without segmenting the interlinearizer does work, but it is hard to browse through the non-segmented text (esp. after moving to another book and back).
Another disadvantage of a non-segmented book is that, when a note gets dislodged from its original position (because of changes in the text) and ends up at the top of the file, it’s quite hard to retrieve its original location.
Could this be a feature request: an option to segment non-Biblical books in some way? Perhaps by adding non-printable/publishable verse functionality??

---

Not only do notes frequently become dislodged in these no-verse books, but the note files also become huge. As I understand it, the entire text of the glossary becomes attached to each note made, so putting many notes in glossary is a quick way to make your project file huge and slow. We found this out the hard way.

---

Thanks for pointing that out. Perhaps this would account for PT slowing down in the GLO book. With over 100 notes in the glossary, PT has to deal with a huge amount of underlying text.

PJK

---

In 9.3 the XXC book is just as impossible to interlinearize as the GLO.

I’ve put the entire glossary in XXC, created a single chapter, and a huge number of verses - one per paragraph. But the interlinear tool only pulls up XXC 1:0 - it won’t go past.

I agree with the earlier commenter who remarks that the GLO is poorly supported and clunky. My requests would be:

1. Make interlinearizer work “out of the box”, meaning that there’s a way to do “section-by-section” interlinearizing (based on \k tags?) without having to add verse numbers.
2. Make tags useable to speed up work in GLO.

(Update: since we’re a Protestant team, we don’t need Tobit. So I found I can put the glossary in Tobit - or any “supported” book of the Apocrypha, I assume - and the interlinearizer Works As Expected.)

(Update #2: Because our glossary has over 100 paragraphs and Tobit 1 has only 22 verses or so, I had to use custom.vrs to add verses to to the chapter. I decided to give it 300, so I added the following line to custom.vrs in the main project folder and the BT folder:

TOB 1:300

Working great now.)

---

One comment about dividing the glossary into chapters (this is off-topic, but may be relevant): if you link glossary entries to Biblical terms, Paratext will delete anything in the paragraph preceding the entry, including verse numbers.

A workaround is to place the verse numbers at the end of the (preceding) paragraph, i.e. to invert \p and \v.

Of course, another solution is to create the links only after deleting the temporary verse numbers.

Paulus+Kieviet

---

Brilliant, thank you! My solution was to create ad-hoc verses but then delete them after interlinearizing and doing the BT because of the issue you brought up. I’ll try the “reverse” method and see how that works out.

Of course, there’s still this issues: we’re early in our project, so we’re regularly adding new entries, which would throw off the verse numbering…

---

Adding entries It would throw off the verse numbering, but because the verse numbering is just a temporary tool, it’s probably not a problem if there are two or three entries in a single verse.

Paulus+Kieviet

Answer 2

Apparently this does work in PT8. Please use Help > Give feedback to report the issue in PT9.2

Answer 3

I have had the same problem with interlinearizing the glossary. Used to use \v to break up all the glossary entries in previous versions of Pt, but had to remove them when I upgraded to Pt 9.2 (otherwise nothing showed at all in the interlinearizer).

Now have a very unwieldy list of glossary items showing up as ‘one verse’. Each time I enter something, it goes back to the top and I have to scroll back down to try and find where I was last in editing.

I’d be extremely grateful if Pt developers could find a way to segment the glossary so that it can be interlinearized easily, entry by entry.

Answer 4

Please consider this: A good, well structured, vernacular, monolingual glossary is like the “welcome committee” to any new reader.

For many people groups, there is no option “to go to church” or to join some “Bible study group”. A glossary cannot replace these options but can muchly help an interested person.

So far, it looks as if the glossary feature in PT is treated - priority-wise - like an afterthought. It is not well integrated, it behaves clumsily, not well documented and missing many features and structures.

Yes, it is often found with the “back” matter; but it can be so important, depending in context.

I have placed this as a separate “reply” on purpose. So other users can “like” it to show support or desire for a better glossary in future versions. If you can write it up better or nicer or add your arguments to maybe convince the leaders who assign priorities, please do so.

Answer 5

If I recall correctly how it’s supposed to work, the Interlinearizer should work with verse/chapter numbers in the GLO book if you give it verse/chapter numbers in the versification (custom.vrs). Has anyone tried that?

Comments

anon291708:

If I recall correctly how it’s supposed to work, the Interlinearizer should work with verse/chapter numbers in the GLO book if you give it verse/chapter numbers in the versification (custom.vrs). Has anyone tried that?

I have tried this and it still doesn’t work.

---

anon291708:

If I recall correctly how it’s supposed to work, the Interlinearizer should work with verse/chapter numbers in the GLO book if you give it verse/chapter numbers in the versification (custom.vrs). Has anyone tried that?

A more nuanced reply than anon848905’s one: I say Yes and no
I’ve just tried, using Paratext 9.2.102.19.

My custom.vrs file looks like this:

GLO 1:26 2:26 3:26

(along with some comment lines)

Switching to Show all chapters mode and adding \c 1 \cl A .... \c 2 \cl B etc. in front of the relevant chunks, I can now interlinearise 1:0, 2:0, 3:0, 4:0 etc. as distinct entities (even going past my customisations). PROGRESS!

If, however, I sub-divide the ‘chapters’ and put \v 1, \v 2 between entries, then I can still access the parts of the book that it considers 1:0, 2:0 etc, but I cannot interlinarise (or see!) 1:1, 1:2, etc.

So, maybe there’s something incorrect with my custom.vrs file, or is this as far as we can get for now?

---

anon542642:

but I cannot interlinarise (or see!) 1:1, 1:2, etc.

So if you manually enter in GLO 1:2 in the verse selector, it appears blank?

Answer 6

No, the display is behaving correctly, but the selector will not permit anything except GLO 1:0, GLO 2:0, etc.
If I alter a verse number, then it resets to 0. If I start the interlineariser on v.2, it starts off showing 2 but before loading the text it resets to 0. If I click on the ‘increment verse’ button, it ignores me. If I decrement the verse, then the chapter number drops one, and the verse number goes to zero.

An interesting tweak which might provide a clue, maybe?
If the chapter number is past the end of the book, (e.g. chapter 38 when there is no \c 37 or later) then the chapter/verse selector allows me to enter / increment to any non-existent verse in that non-existent chapter.

Comments

Interesting. That definitely sounds like a bug. Could you report it via Help > Give feedback?