Parallel passages problem

Question

We started a Synoptic check and ran into a problem in Luke 3:4. When we select All parallels, it gives us one OT reference crossed out in red and 3 NT references. Since most of the OT has not yet been translated, we are checking NT parallel passages. When we select NT/NT both the OT reference and the 3 NT references disappear. If we select Synoptic Gospels the references to Matt, Mrk and Jhn disappear in addition to the OT reference. This is not the behaviour I would expect.

I noticed that when doing a NT/NT parallels in Romans, there are two lines, one with the OT reference, one without. This means that we get the references no matter whether we select NT/NT or OT/NT, except that the last option has an additional OT reference and a duplication of the NT references.

It seems to me that some more work needs to be done on the Parallel Passages tool.

Iver+Larsen

Answer 1

The design of the NT/NT selection is that it will show parallels that only go from NT to NT. It does not act like a filter to filter out the OT references. What you are seeing is by-design behavior. The same applies to the Synoptic Gospels selection.

Iver+Larsen:

It seems to me that some more work needs to be done on the Parallel Passages tool.

Yes, more work is planned to clean up some of the UI and make it easier to use/understand.

Comments

As I mentioned, the design in Romans is different from the design in Luke (Gospels). Both cannot be correct. The list of passages needs to be adjusted for the Gospels in order not to be misleading.

---

Let me explain the current problem from the example of Rom 13:9. If you choose All Passages, you see four lines. The first line has two OT references and 9 NT references. One OT portion is reflected in 10 NT passages. These NT passages needs to be checked against each other for a NT project without the OT.
The second line has 2 OT references and one other ref. in Romans.
The third line has one OT reference. This text is reflected in 9 NT passages. These 9 passages needs to be checked against each other in a NT project.
The fourth line has only one other reference in Romans, a duplication of Line 2 without the OT references.
If we select the NT/OT filter, only the first 3 lines with OT texts are shown. That is correct.
If we select the NT/NT filter, only line 4 is shown. This is wrong, because we then lose the 10 passages from line 1 and the 9 passages from line 3. The NT/NT filter should in fact show the first 3 lines because there are NT references to this verse in Romans in other books (or in another place in the same book).
In short, the NT/OT filter should include all lines with one or more OT references. The NT/NT filter should show all lines with one or more NT passages.

---

I was just talking to the developer regarding how these filters work and basically they do not work like you are expecting.

When you select NT/NT it will “only” show those parallels that exist “only” in the NT. So if a parallel also has an OT verse then it will not show up in that list. In the Romans 13:9 example the passages that have OT verses will not show up under the NT/NT filter since they have OT verses. This is the design of the filters.

When you select Synoptics it will only show those parallels that exist only in the Synoptic Gospels. So for instance if All Parallels shows a parallel in the Gospels and Acts, when you choose Synoptics it will not be listed since it has a parallel outside the Synoptics.

Bottom line, you are correct that you should probably be working from All Parallels most of the time. You can make the list simpler by looking at Current Verse or Chapter.

As has been stated earlier in the thread, there is work being done to the UI so beaware that things will probably change.

---

Thanks. My next question is whether the developer is willing to listen to users. What we users need is not the current design, but the design I suggested. The NT/NT should show NT parallels regardless of whether there is an OT ref. also.

---

Iver+Larsen:

What we users need is not the current design, but the design I suggested.

Unfortunately, you are not our only user. The current design is based on feedback that it’s never good to look at a parallel verse without considering all of the other parallel verses. This is why a given parallel is not ever filtered in such a way to remove verses - it’s either all or nothing.

That being said, I think the way the data is set up for Romans is not correct - there shouldn’t be two sets of the same parallels (one with the OT and one without). This defeats the purpose of reviewing them together. Unless, since I’m not a Hebrew/Greek scholar, there is actually a slightly different parallel between the NT/NT compared with OT/NT/NT for that same phrase. Maybe @anon645204 can comment on this.

---

I would say you get feedback at different times. The feedback you refer
to appears to indicate the wishes of the users before the program is
created. This is quite different from feedback once the program is
and being used in real life checking. If you only do the first
and neglect the second you create some frustrated users.

The principle that it is never good to look at a parallel verse without
considering all the others is good. But if one is working on checking
the NT and have not yet started on the OT, you should not look at Rom
13:9 without looking at the many other places in the NT which are more
or less parallel. This verse happens to have a selection of the 10
commandments as well as the commandment to love your “neighbour”. Many
other verses in the NT refer to these commandments. If a team is working
on the NT and trying to finalize it before the OT verses have been
translated, they need to compare how these commandments have been
translated in the NT. You would expect to be able to do that by
selecting NT/NT. The way it is now, this can only be checked by
selecting All Passages which defeats the idea of having a NT/NT filter
at all. Of course, if you tell everybody that they should not try to use
the NT/NT filter, and instead use the All Passages filter, that is one
way to “solve” the problem. You could also say that these commandments
might have been checked in the Biblical Terms list, but why have a tool
that does not work properly because it has not been tested sufficiently
by users? I am happy to wait for this to be improved. We have completed
our parallel checking for Romans (comparing with all other NT books that
contain parallels) by using All Passages, but it took me some time to
figure out why the NT/NT did not work as it should.

Iver+Larsen

Den [Phone Removed]kl. 17:47 skrev anon291708:

---

I joined the brand-new User Experience team after the current implementation had already been released (under 7.6 Beta), so I can’t speak to how much user input or testing went into that. But we did identify various usability issues with it after that point and pursued some research and discussion with quite a few expert consultants.

As a heads up to all of you, that problematic filter will very likely be hidden (only shown if launching PP while holding Shift) in some future release of PT 8.x, because it seemed a rather indirect way of manually keeping the bottom pane’s content manageable. It’s also risky, because users can accidentally exclude/overlook some parallels if they misunderstand these rather confusing filters. (This issue is mitigated by the fact that Assignments and Progress can let you know what’s been overlooked, once the books in question reach a certain stage.)

Most likely, we will automatically limit the bottom pane’s columns to those passages which actually exist (i.e. have been drafted) in the selected project. (The top pane will probably always show all parallels for which at least one passage has been drafted, so that you can be aware of them, but undrafted references will be dim or non-bold.)

@Iver+Larsen, I would definitely like to dialogue further with you to make sure I really am understanding the issues you describe, and whether our newer designs will do what you need (or would have needed). Please ping me on skype (jvcoombs) or email ([Email Removed]), if one of those works for you.

---

I am happy to hear that there is a User Experience team. It is only a few days ago that the team I work with at the moment migrated to P8 and I am learning as we go along partly from the Helps, partly through trial and error.
This project has drafted the NT, but still needs to go through other stages, including more parallel checks.
There is certainly a space problem on the screen, if you have about 10 parallels. I think it will be helpful to limit them to the books that have been translated, since I expect the team would want to check parallels for each book they are preparing to publish.
I have spent a couple of days setting up the Project Plan and assignments for a team of six people. We have finished Romans Parallels in the sense that we have checked with all NT passages. But there are still 40 issues remaining under the Parallel Passages check in the progress chart. These are all cases where there is an OT quote unique to Romans, so at this stage I would suggest that there are no issues left. Once the OT has been translated, we can revisit those. But let me talk to you directly. Thanks.

---

@Iver+Larsen , regarding the current list of false errors you’re seeing in Assignments and Progress, that sounds like something that needs to be fixed. I have proposed that Parallel Passages shouldn’t expect you to check off any row which doesn’t have at least two drafted passages. In fact, such a checkbox shouldn’t even be enabled, since it has little or no meaning.

Workaround: In the meantime, I don’t see any harm in checking off all ROM passages which are alone in their row because all their parallel passages are still undrafted. (The rows should show up again to be checked whenever you do draft one of them, or if you edit the ROM text.)

Answer 2

I looked at some OT quotes in the Synoptics and in Romans, and I would agree that there seems to be some inconsistencies. It might be related to the fact that The Gospels have the filter “Synoptics” available that Romans does not. So how should we understand OT quotes in the Gospels? Is Luke quoting Matthew, quoting Mark who was quoting Isaiah or a Psalmist. Or do we conceive of the scenario as each Gospel writer quoting the OT directly? And of course Paul was only quoting the OT directly so there is no debate there. I would submit that for the purposes of a software tool for Bible translation we should uniformly have all NT authors quoting the OT directly. In this way the Gospels and the Epistles will behave the same in the parallel passages tool.

Regardless the model being used we need to be able to explain it to users. If we want the Gospels and the Epistles to act differently for some academic reason, we need to make this more obvious to users. How about an Epistle/OT filter as a corollary to the Synoptic filter?

Comments

I have worked with the Parallel Passages tool in P8 for two days now
with a team of 5 translators. The only way to use it properly at the
moment is to choose All Passages. Any other choice will cause you to
lose out on some relevant parallel passages. Whoever has selected the
passages and/or designed it, needs to do more work on it. I was a
programmer 45 years and I know what testing a program means. You don’t
expect to get everything right the first few times.

Iver+Larsen

Den [Phone Removed]kl. 17:59 skrev anon044949 :

Answer 3

@Iver+Larsen, thank you for telling us about your experience.

We are going to need to speak with a variety of translators and translation consultants to figure out what to do about this. I think we all agree that the present situation is confusing.

I just have one data point so far. I spoke with Stephen Pattemore who is a UBS translation consultant that is here helping teach a P8 class in PNG. His preference, as I understand it, is that there be three lists.

1. Gospel Parallel Passages. Note that the current name “Synoptic” is apparently not accurate since only MAT/MRK/LUK are considered to be “synoptic”.

This list would contain only references from the Gospels. Nothing else. This is the most important list because this material is believed to derive from a common source and so has the strongest reason to be word for word consistent among the common phrases.

2. Other NT/NT references. Less important and (according to Stephen) a different kind of beast than list 1) so needing to be treated a bit differently and worth separating out.

3. OT/NT references. A list that Stephen advises most teams to ignore unless they have advanced training. The big problem as I understand it is that most of the entries in the NT are quoted from the LXX. But most translations translate the OT from the Hebrew so it is no small matter to figure out what degree of consistency is really appropriate, especially if you do not have a translation of the LXX in a language you read to refer to.

I have heard the saying “Ask 2 different translation consultants, get 3 answers” so this probably just scratches the surface of what we need to look into.

anon451647

Comments

I do not think there is much disagreement between consultants, but there
is some work to be done on the lists. Let me comment below.

Den [Phone Removed]kl. 13:41 skrev anon451647 :

[anon451647] anon451647 [Link Removed]
June 21

@Iver+Larsen [Link Removed], thank you for telling
us about your experience.

We are going to need to speak with a variety of translators and
translation consultants to figure out what to do about this. I think
we all agree that the present situation is confusing.

I just have one data point so far. I spoke with Stephen Pattemore who
is a UBS translation consultant that is here helping teach a P8 class
in PNG. His preference, as I understand it, is that there be three lists.

1. Gospel Parallel Passages. Note that the current name “Synoptic” is
   apparently not accurate since only MAT/MRK/LUK are considered to be
   "synoptic".

This list would contain only references from the Gospels. Nothing
else. This is the most important list because this material is
believed to derive from a common source and so has the strongest
reason to be word for word consistent among the common phrases.

At the moment the Synoptic Gospel filter selects passages from Mat, Mrk
and Luke, so the name is ok. There are a few parallels in John, so I
agree that it would be better to have a Gospel Parallel Passages
covering the four gospels. This means some reworking of the lists or
filter. Some of these parallels are parallels, because they quote the
same OT text, but the OT texts should not be included in this filter
(which they are not at the moment.)

2. Other NT/NT references. Less important and (according to Stephen) a
   different kind of beast than list 1) so needing to be treated a bit
   differently and worth separating out.

Yes, a different kind of beast. Or I might say two beasts. Since Paul
has written many letters, there are places where he gives the same or
similar advice in different letters or the same letter in different
places. So, we are checking Paul against Paul. (Or Paul against James
and Peter and John and Hebrews). One beast. Paul and other NT writers
often quote OT texts, and they are sometimes quoted in several places by
different authors. The second beast. I ran into the first beast in
Romans. I had expected to use the NT/NT filter for that. It has 17 lines
only. This is not adequate. I gave the example of Rom 13:9. In All
Passages 4 lines are seen, but the first 3 lines are hidden in the NT/NT
filter because they go back to an OT reference. The lines need to be
adjusted so that the NT/NT filter shows these other NT passages
regardless of whether they go back to an OT quote. I do not think that
the NT/NT filter needs to include the OT reference which will be handled
in another filter later.

3. OT/NT references. A list that Stephen advises most teams to ignore
   unless they have advanced training. The big problem as I understand it
   is that most of the entries in the NT are quoted from the LXX. But
   most translations translate the OT from the Hebrew so it is no small
   matter to figure out what degree of consistency is really appropriate,
   especially if you do not have a translation of the LXX in a language
   you read to refer to.

A team working on only the NT does not need to be overly concerned with
the OT, but once they work on the OT they do face the problem of
differences between the Hebrew and LXX. I assume any NT will be revised
when the OT is done, and some parallels may change slightly. So for
these teams, they do need both an OT/NT and an OT/OT list. The All
Passages list seems to be a combination of everything and that may be
useful once in a while, but if the other lists are done well, it may not
be needed.

I have heard the saying “Ask 2 different translation consultants, get
3 answers” so this probably just scratches the surface of what we need
to look into.

Much work has already been done and we appreciate that. But there is
room for improvement. This is to be expected when something this
complicated is under development.

Answer 4

One other thing that StephenP told me was that when he had to check the gospel parallel passages using just a word processor and a book it took him 3 months. Anyone that manages to do it in less time than that should IM(H?)O sign up for annual be-nice-to-a-developer day.