Interlinearization using NIV 1984 as model text quit working in PT8

Question

We have interlinearized English glosses for the vernacular. It’s very helpful for advisors and consultants. We used the NIV 1984 version as the model text. We were using those for years in PT7, and when we switched to PT8, they were working fine until a few days ago when they stopped working. I’m wondering if somehow the name of the NIV 1984 file has been recently changed so that PT8 no longer finds it the same way it used to find it.

I have tried to reopen the interlinearizer again in PT8, but the glosses which we had been entering for several years are not available to us now. I believe the glosses are associated with the NIV 1984 project and now PT can not find them.

anon758749

Answer 1

I see after the update the short project name is now NIV, no longer NIV84. But in my test project where I had done glossing with the NIV84 as the model text, my glosses were still present after I updated the resource. I don’t think changing the project short name would affect the glossing. I had wondered from your report if the language ID in the NIV resource had changed, but that doesn’t seem to be the case.

When you said the interlinearization quit working after updating the NIV 1984, did you mean that the interlinearizer wouldn’t run, or gave an error message if you select the new NIV (1984) as the model text? Or did you mean the glosses you had entered before no longer showed up?

Comments

sewhite, I’m looking for some help. A project I am working with, LED_UG, says it is using NIV-84 as its model text for interlinearization. When I put the NIV-84 as the model text, my interlinearized data is very incomplete, with many words with a red Paratext guess.
So I think this is an example of a resource that has changed its language ID. The two folders under the language folder are interlinear_en, and interlinear_en-US. Of these, clearly the one in use is interlinear_en. The project team tell me this is being produced by their model text NIV-84.
I suspect they may have an old NIV-84 copy, with a pre Paratext 8 language id.
When I switch in the interlinearizer to the model text being any English resource with the language ID (eng), bingo, the glosses are working.
What is the best way forward, from here, so that no internearized work is lost, and everything works OK for everyone on the team, translators and consultants?

---

I had something like that happen once when I updated the source text. He should reinstall the exact version of NIV84 that he was using when he did the interlinearizing and when he does that, I expect his glosses will return to the way they were before. And in the future, he should never update the NIV84 source text. If he stays with a different version of NIV84, then he will lose most of his glosses.
anon758749 Lindstrom
PNG branch

---

anon758749, the problem with this is that consultants do not have access to his original source text. We only have access to what is in the DBL, which is NIV84, which has eng-US as the language ID. So I think the solution has to be to switch to an English model text, that has the language ID as eng. I think that is what en became, when Paratext 7 migrated to Paratext 8, some 4 years ago. Does that solution sound right to you?

---

I’m not sure. We’ve been advising one group that started using the interlinearizer for vernacular to English glosses back when it first became available. It worked fine in PT7 and PT8. But at one point, I clicked up an update NIV84 in the DBL resource list in PT8 and that messed up all the glosses. I went back into the resource files and inserted a copy of the NIV84 from before the update and that restored all the glosses to their original form.

We have only had a few consultants who have used that interlinear and I have not heard them having problems. But perhaps they are not getting all the glosses properly. Now I am careful never to update the NIV84 resource and it has continued to work in the interlinear on my computer But perhaps when we work with more consultants we’ll find it does not work for them.

If you figure out a way to keep the glosses and use the currently available DBL texts, let me know.

anon758749

---

@anon084052 I believe that applying this solution would solve your NIV84 issue: A manual restore of interlinear data after migration - #4 by phil_leckrone. Have you had a chance to try it?

james_post

---

james_post,

Thanks for the suggestion from far away. We just got back to PNG and are currently on quarantine living behind the tape.

After looking at that link you sent, I don’t think I understand what would need to be done.

Here’s the problem: If I update the NIV84 in the resources, it messes up all the glosses (at least it did that when the update NIV84 version first appeared.) I fixed it when that happened by reinstalling the old NIV84 resource and that fixed all the glosses. And now I never update the NIV84. So what would I need to do to update the resource so it would not mess up the glosses. Doing that would be helpful to consultants who don’t have the old NIV84 resource.

Blessings,
anon758749

---

@anon084052 anon758749, I think the problem relates to the change in ISO codes, that took place during the migration process from Paratext 7 to Paratext 8. NIV changed I think from en to eng-US, and that is what messed up your data. I believe if you used any English version, that has the language ISO of eng, your data would still work. So I don’t believe there is a need to stick to the old NIV version. Here is the ISO code for English:

image720×40 1.06 KB

You can see that a code of eng in your Paratext English versions, is the same as en. What you MUST stay away from, are all the NIV versions, which are now either eng-US, or eng-GB.
Use any English version that has eng as its ISO for your Model Text when interlinearizing, and I think all should be OK.

---

Thanks. That makes sense. And I checked and it does work as long as the model text is eng (at least it worked for the NRSV which is eng). That will help me when I give the interlinear to consultants to use. At this point, it really doesn’t matter what the model text is since almost every Kapin word already has an English gloss. The only concern is keeping the glosses which now I better understand how to do that.
anon758749.

Answer 2

This was definitely caused by an update to the NIV 1984 resource. I know that because I returned to the earlier NIV 1984 resource and now my interlinear works again. So I will avoid updating the NIV1984 text.

Answer 3

Hi, I am responsible for the NIV84 Paratext resource. I have changed the short name back to NIV84.

The short name should have been NIV84. The abbreviation was incorrect in the PT registry. I am not sure why it got changed to NIV, but think they may have updated names from the registry last week.

I have therefore adjusted setting and metadata to ensure that the NIV84 shall remain the NIV84. I apologize if this has caused problems with your Interlinearizations.

Other NIV resources available are NIV11 and NIV11UK.

Comments

CrazyRocky:

… the NIV84. … Other NIV resources available are NIV11 and NIV11UK.

I guess we never had an NIV84UK in PT7, right? It would be nice to have that as an option. I’m told the NIV84 is a popular source text for MTTs, and if that’s right, I’d have thought it’d be worth having both variants.

Would that ever be possible? Or is the 1984 version too much ancient history to do any work on permissions?

---

We actually do not have the NIV84UK in Paratext so obviously we cannot easily put it on the DBL, which is necessary to make it a resource. If I knew there was a lot of demand for the Anglicized 84, then I would make a request to see if we could make it available.

Let me add that the NIV11UK has an updated format that I believe is much superior to the 84, particularly for lists… We will be transitioning the NIV11 to the same format within a few months. Also as soon as PT 8.1 is released, we will be updating the markup with some of the new format markers.

Although there are a lot of people who cling to the 84 for its older language, the NIV Committee on Bible Translation has put a lot of thought into every change that was made and is continuing to be made to the NIV. Although i do not agree with every change that has been made, I have reviewed all the changes in the NIV11 many times and I do consider it a superior translation overall.

---

[Link Removed]

Answer 4

Please look at the following link that will describe what you need to do to fix the problem with the language ID changing.

A manual restore of interlinear data after migration Helps

It seems that Paratext 8 has a problem migrating interlinear glosses if you use the Interlinearizer in 7 to gloss text using the NT Greek as your model text. A user asked for help in this situation, and I saw that after migration, the language code in the lexicon.xml and in the per-book gloss files was “el” while the code for NT greek should be “grc”. (In a test project, I glossed a few words in Greek, then migrated, and in this migrated project, the code for NT Greek became “lbj”, the code for…

Comments

I think the idea of changing the model text to one that uses “en” instead of “en-US” would work. The one thing I can think of is it might change some of the guessed glosses if they have never been approved. A guessed gloss comes from Paratext doing statistics on when does an English word appear in the same verses as a base text word, and so changing the model text might alter the statistics where the new model text uses different English words. But this might be minimal or undetected if most guessed glosses have been either approved or changed by the team members glossing. All the user decisions (approving glosses or changing a guessed gloss to something else) should remain unchanged.

If you thought it necessary to update the NIV84 resource that the team is using to match what the consultants are using, you could manually update the glossing data so it would show up with the current NIV84 coded as en-US.

---

sewhite, I think your analysis is great. We will try switching to a model text that has the language ID of (eng). By the way, do you know what’s the difference between a language ID of en and one of eng? My current understanding is that en is pre Paratext 8 migration process, and eng is attached to all English versions now, -Paratext 8 migration. Exceptions being unusual language IDs, such as those like eng-GB or eng-US, or something else like that.

---

Sorry, I misspelled the code. You are right, current English projects are coded eng in Paratext.

---

anon758749:

By the way, do you know what’s the difference between a language ID of en and one of eng?

See this :

A manual restore of interlinear data after migration Helps

Yesterday I ran into a situation where in PT7 the language had been “Spanish” for the RV60 and in PT8 the language for RVR1960 is “spa”. I followed the instructions of Steven in an earlier to make the appropriate changes, but the glosses still did not appear as approved (as they were in PT7). anon291708 S. pointed out to me that in PT8 there are certain languages that display the three letter code (in this case spa), but internally use the two letter code (in this case es) for matching the interl…