[acknowledged]BUG searching Hebrew/Greek Lemma/Roots from the search window

I am pretty sure I used to be able to search any Hebrew Bible for roots or lemmas directly from the search window, but now it only works with the Lexham Hebrew Bible. It works when the search is initiated from the resource window via the context menu, but not when I start with root:h:searchterm or lemma:h:searchterm and select them from the drop-down list.

I do this:

Then pick one and get that:

But if I select LHB:

It's not just BHW that does not work. All Hebrew Bibles aside from LHB that I have tried don't work.

I have similar problems with Greek, though it's less systematic. In the example below you can compare the results from the left (where I put in the lemma from the search window using the drop-down list) with the results on the right (using the context-menu from the resource window):

I'm using 6.1 SR-2 (it was the same with SR-1).

Find more posts tagged with

Comments

Martin Grainger Dean

I can confirm this. Running 6.2 Beta 1 (6.2.0.0011), Win7. Martin.

Vincent Setterholm

LHB is the only Hebrew Bible we have that currently has root data. (BHt will have its own root analysis). I haven't really tested the drop down word selector in the Bible Search UI. In the Morph Search tab, it's supposed to only show options that actually occur in the Bible(s) you select for searching, but there's some extra material showing up in that drop down at present. I don't know that the regular Bible Search has any such design, making it a bit more likely to search for things that don't exist in the database you're using.

Dave Hooton

It's not just BHW that does not work. All Hebrew Bibles aside from LHB that I have tried don't work.

The suggestions are for a specific morphology e.g. <Root = lbs/he/בטח:2>, where lbs = Logos Morph & he = Hebrew. So it will only work in LHB, LHI or Reverse Interlinears like ESV, LEB, NASB. Similar for Greek e.g. <Root = lbs/el/αγαπαω>, where el = Greek. But <Root = lbs/el/λογος> is wrong as it should be <Root = lbs/el/λεγω>.

Getting these in Bible Search is buggy.

Francis

The suggestions are for a specific morphology

There lies the problem: using the h:lemma: functionality only produces these morphologically-constrained choices from the drop-down list in the search window. I wonder whether the morphology should be left unspecified in the lemma syntax and determined rather by which resource one searches. In other words, the lemma choice in the list should only be <Lemma = בטח:2> (just as it is in BWS in which textual searches can run across the various morphologies).

Moreover, the search templates are not altogether clear. From the picture below, it would appear that typing "hebrew:" should have the same function as "h:" but it does not work with transliterations the way h: does. So what is it's use then? I mean, if you type מלך without prefixing it with "hebrew:" the results are exactly the same.

Dave Hooton

There lies the problem: using the h:lemma: functionality only produces these morphologically-constrained choices from the drop-down list in the search window. I wonder whether the morphology should be left unspecified in the lemma syntax and determined rather by which resource one searches.

The problem/bug is Bible Search providing morphological suggestions. Morph Search was designed to provide the right morphology for a resource but you have to feed it correctly for roots e.g. root:g:leg to get root:λεγω (you have to change lemma to root!). When you click on Bible Search it gets expanded to <Root = lbs/el/λεγω>. So all is not well with roots.

In other words, the lemma choice in the list should only be <Lemma = בטח:2> (just as it is in BWS in which textual searches can run across the various morphologies).

Searches need the morphology to be specified implicitly or explicitly. Morph Search uses the format lemma:בטח:2 and gets the morphology from the resource or morphology you select. Bible Search needs the format <Lemma = lbs/he/בטח:2> because it doesn't supply morphology.

BWS works differently. For manual input you enter בטח and select the lemma בטח 2. This leverages the morphology of your preferred bible. If you run BWS from the context menu you select the lemma בטח 2 in LHB. From Andersen-Forbes you select the lemma בטח 1 for the same word and BWS has to convert that for the translation ring using your preferred RI bible (ESV) because the morphologies are different.

Francis

But <Root = lbs/el/λογος> is wrong as it should be <Root = lbs/el/λεγω>.

Not sure, why you say this. I agree it's wrong, but I just don't know why you bring this up.

BWS works differently. For manual input you enter בטח and select the lemma בטח 2. This leverages the morphology of your preferred bible. If you run BWS from the context menu you select the lemma בטח 2 in LHB. From Andersen-Forbes you select the lemma בטח 1 for the same word and BWS has to convert that for the translation ring using your preferred RI bible (ESV) because the morphologies are different.

Thanks, Dave, for this enlightening explanation.

Searches need the morphology to be specified implicitly or explicitly. Morph Search uses the format lemma:בטח:2 and gets the morphology from the resource or morphology you select. Bible Search needs the format <Lemma = lbs/he/בטח:2> because it doesn't supply morphology.

Should it not? In other words, would it not be possible and better for Bible search to emulate the behavior of Morph search in this regard? Example: I type lemma:בטח in morph search while have having search in set to the NA28 Greek New Testament, it intelligently recognizes the mismatch and does not offer a drop-drown option. When I change it to an appropriate resource (e.g., BHW), then it offers the corresponding lemma choice in the drop-down. Bible searches would do the same.

Fr Devin Roza

From the picture below, it would appear that typing "hebrew:" should have the same function as "h:" but it does not work with transliterations the way h: does. So what is it's use then? I mean, if you type מלך without prefixing it with "hebrew:" the results are exactly the same.

The "hebrew:" is for searching within texts which are tagged as Hebrew language. This could include text within an English language resource. This could be useful precisely to distinguish the Hebrew מלך from the Aramaic מלך in a resource that doesn't have morph tagging, in an English language book, etc.

Normally you would use this on the Basic tab, not the Bible tab. It could be useful when a word appears in multiple modern languages as well, say in German and English, and you want to only search for the German hits.

Andrew Batishko

The "hebrew:" is for searching within texts which are tagged as Hebrew language. This could include text within an English language resource. This could be useful precisely to distinguish the Hebrew מלך from the Aramaic מלך in a resource that doesn't have morph tagging, in an English language book, etc.

Normally you would use this on the Basic tab, not the Bible tab. It could be useful when a word appears in multiple modern languages as well, say in German and English, and you want to only search for the German hits.

It can also be useful for searching for transliterated words.

Francis

The "hebrew:" is for searching within texts which are tagged as Hebrew language. This could include text within an English language resource. This could be useful precisely to distinguish the Hebrew מלך from the Aramaic מלך in a resource that doesn't have morph tagging, in an English language book, etc.

Normally you would use this on the Basic tab, not the Bible tab. It could be useful when a word appears in multiple modern languages as well, say in German and English, and you want to only search for the German hits.

Thanks, Devin! I guess I was inclined toward understanding as h: by the fact that the examples right above are lemma-searches and I was looking for h: examples in the templates. But I'm glad I learned this from you! Turns out there is more French in my library than I would have thought!!!

Dylan Rondeau

The suggestions are for a specific morphology

There lies the problem: using the h:lemma: functionality only produces these morphologically-constrained choices from the drop-down list in the search window. I wonder whether the morphology should be left unspecified in the lemma syntax and determined rather by which resource one searches. In other words, the lemma choice in the list should only be <Lemma = בטח:2> (just as it is in BWS in which textual searches can run across the various morphologies).

Moreover, the search templates are not altogether clear. From the picture below, it would appear that typing "hebrew:" should have the same function as "h:" but it does not work with transliterations the way h: does. So what is it's use then? I mean, if you type מלך without prefixing it with "hebrew:" the results are exactly the same.

This was overlooked in the implementation of the Autocompleter with L6. Providing lemma searches that don't work isn't terribly helpful - we'll be fixing this in a future release.

Francis

we'll be fixing this in a future release.

[Y]