[opensiddur-tech] The BDB Outline

[David Troidl]

Hi all,

The official release of the BDB Outline in online at
http://github.com/openscriptures/HebrewLexicon/downloads
It include the XML schema and the readme.

Questions or comments welcome.

Peace,

David

--
You received this message because you are subscribed to the Google Groups "opensiddur-tech" group.
To post to this group, send email to opensid...@googlegroups.com.
To unsubscribe from this group, send email to opensiddur-te...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/opensiddur-tech?hl=en.

[Ze'ev Clementson]

Hi David,

On Wed, May 12, 2010 at 5:46 AM, David Troidl <David...@aol.com> wrote:
> Hi all,
>
> The official release of the BDB Outline in online at
> http://github.com/openscriptures/HebrewLexicon/downloads
> It include the XML schema and the readme.
>
> Questions or comments welcome.

This is excellent! Thank you for doing this (and releasing it as PD)
as this must have taken you a huge amount of time to collate. Just a
few questions (some of which I can guess at but which I just want your
confirmation on):

1. What's the difference (or significance of them having different
attributes) between the following word entries (they all have
different attributes):
<w strong="1417">גְּדוּד</w>
<w src="BDB" strong="1418">גְּדוּדָה</w>
<w src="Strong" strong="1417">גְּדֻדָה</w>
<w>גַּד</w>

2. Sometimes a ref has a word associated with it and sometimes not - why?:
<w>גִּדְגָּד</w>
<v ref="g.ao.am"/>
<v ref="x.bm.ab">חֹר</v>

3. What is the significance of the following 'type', 'cite', and the
form attributes (I saw the permissible values in the xsd schema but am
not sure of the exact usage) :
<entry id="g.cb.aa" type="root" cite="full" form="false">
   (e.g. - what's the difference between root/entry, full/partial, and
true/false values for these attributes?)

4. There is still no Page#1 - is there some reason for this or is this a bug?

5. Earlier, you indicated that you would be "numbering the WLC with
the new IDs" - is this still your intention? Presumably you will
associate both Strongs# and BDB ref with each WLC word? If so, what
attribute will you use for BDB ref?

6. In the first release (in April), you noted: "There are 521
unassigned Strong numbers.  I'll have to catalog those and look into
them. 7591 Strong numbers have a unique BDB entry.  This bodes well
for merging the IDs. The Strong numbers with multiple entries, 573, I
intend to break down by reference." - Did you resolve the issue of the
521 unassigned Strong numbers?

Thanks,
Ze'ev

[david...@aol.com]

Hi Ze'ev,

On 5/12/2010 11:33 AM, Ze'ev Clementson wrote:

Hi David,



On Wed, May 12, 2010 at 5:46 AM, David Troidl <David...@aol.com> wrote:

Hi all,



The official release of the BDB Outline in online at

http://github.com/openscriptures/HebrewLexicon/downloads

It include the XML schema and the readme.



Questions or comments welcome.

This is excellent! Thank you for doing this (and releasing it as PD)

as this must have taken you a huge amount of time to collate. Just a

few questions (some of which I can guess at but which I just want your

confirmation on):



1. What's the difference (or significance of them having different

attributes) between the following word entries (they all have

different attributes):

<w strong="1417">גְּדוּד</w>

<w src="BDB" strong="1418">גְּדוּדָה</w>

<w src="Strong" strong="1417">גְּדֻדָה</w>

<w>גַּד</w>

The first <w> is always the main entry from BDB.  After that, if BDB lists an extra form, that doesn't appear in Strong's, it gets a src="BDB".  Likewise for Strong forms that don't appear in BDB, they get src="Strong".

2. Sometimes a ref has a word associated with it and sometimes not - why?:

<w>גִּדְגָּד</w>

<v ref="g.ao.am"/>

<v ref="x.bm.ab">חֹר</v>

The entry reads "v. following & sub חֹר."  So the first <v> references the following entry, with no word, and the second references the given word.

3. What is the significance of the following 'type', 'cite', and the

form attributes (I saw the permissible values in the xsd schema but am

not sure of the exact usage) :

<entry id="g.cb.aa" type="root" cite="full" form="false">

   (e.g. - what's the difference between root/entry, full/partial, and

true/false values for these attributes?)

'type' distinguishes roots from other entries.
'cite' is used where BDB has a dagger, indicating that the entry lists all occurrences of the word in the Hebrew Bible.
'form' is for the square brackets, that indicate that the lemma form never appears in the Hebrew Bible.

4. There is still no Page#1 - is there some reason for this or is this a bug?

Not a bug.  Anything before the first page break is page 1, by default.

5. Earlier, you indicated that you would be "numbering the WLC with

the new IDs" - is this still your intention? Presumably you will

associate both Strongs# and BDB ref with each WLC word? If so, what

attribute will you use for BDB ref?

This is still about 3 stages down the line.  I have to separate out the entries I want to index, then assign the references on the ones that split a Strong's entry, and then I'll be in a position to map them to the WLC.

I want to maintain, and refine, the BDB structure, incorporate the Strong structure, and have some means of alphabetizing the words, in the midst of all that.

6. In the first release (in April), you noted: "There are 521

unassigned Strong numbers.  I'll have to catalog those and look into

them. 7591 Strong numbers have a unique BDB entry.  This bodes well

for merging the IDs. The Strong numbers with multiple entries, 573, I

intend to break down by reference." - Did you resolve the issue of the

521 unassigned Strong numbers?

All the Strong numbers have been assigned.  There are now 8129 Strong numbers that map to a unique BDB entry.  The rest will have to be divided.

Peace,

David

Thanks,

Ze'ev

[Ze'ev Clementson]

Hi David,

Ok, thanks for the clarifications.

Ze'ev

[Ze'ev Clementson]

Hi David,

Just one more followup clarification:

> 6. In the first release (in April), you noted: "There are 521
> unassigned Strong numbers.  I'll have to catalog those and look into
> them. 7591 Strong numbers have a unique BDB entry.  This bodes well
> for merging the IDs. The Strong numbers with multiple entries, 573, I
> intend to break down by reference." - Did you resolve the issue of the
> 521 unassigned Strong numbers?
>
> All the Strong numbers have been assigned.  There are now 8129 Strong
> numbers that map to a unique BDB entry.  The rest will have to be divided.

What do you mean by "The rest will have to be divided"? Do you mean
that there is no specific BDB entry for those 521 words but there is a
BDB reference from those words to a BDB word (which would have a
different Strongs#) or is there no mention of those 521 words in BDB
at all? Could you please give me an example of one of the 521 words
that don't have a specific BDB entry so that I can see how it appears
(or doesn't appear) in your BDB outline? Also, there are 8674 Strongs
numbers and you said that 8129 numbers map to a unique BDB entry -
doesn't that leave 545 words that don't map to a unique BDB entry and
not the 521 that you indicated?

[david...@aol.com]

Hi Ze'ev,

On 5/13/2010 1:11 PM, Ze'ev Clementson wrote:

Hi David,



Just one more followup clarification:

6. In the first release (in April), you noted: "There are 521

unassigned Strong numbers.  I'll have to catalog those and look into

them. 7591 Strong numbers have a unique BDB entry.  This bodes well

for merging the IDs. The Strong numbers with multiple entries, 573, I

intend to break down by reference." - Did you resolve the issue of the

521 unassigned Strong numbers?



All the Strong numbers have been assigned.  There are now 8129 Strong

numbers that map to a unique BDB entry.  The rest will have to be divided.

What do you mean by "The rest will have to be divided"? Do you mean

that there is no specific BDB entry for those 521 words but there is a

BDB reference from those words to a BDB word (which would have a

different Strongs#) or is there no mention of those 521 words in BDB

at all? Could you please give me an example of one of the 521 words

that don't have a specific BDB entry so that I can see how it appears

(or doesn't appear) in your BDB outline? Also, there are 8674 Strongs

numbers and you said that 8129 numbers map to a unique BDB entry -

doesn't that leave 545 words that don't map to a unique BDB entry and

not the 521 that you indicated?

If I remember correctly, the 521 were those that had no BDB assignment at all.  Those have been taken care of.  The remaining 545, which is less than it was before, are those that have more than one BDB entry.  So I'm taking those entries and dividing up the WLC references for the Strong number to agree with the BDB entries, wherever possible.  I've done several so far and only came across two uncertain cases.

Peace,

David

Thanks,

Ze'ev

[Ze'ev Clementson]

Hi David,

Ok, thanks!

Ze'ev

[Ze'ev Clementson]

Hi David/Efraim,

I'm writing an xslt script to extract the related Strongs# and
BDBPage# values; however, I'm having a bit of a problem with the
correct xpath syntax so I'm hoping that one of you can give me some
advice. If I use:
<xsl:value-of select="preceding-sibling::lexicon:page/@p"/>

I can get the correct page# for a lexicon:entry node so long as there
has been a lexicon:page element before that node in the same section
(since there may be multiple page#s in the preceding siblings of the
current node, I have to do some substring processing to grab the last
one; however, that's not a problem). However, if there hasn't been a
lexicon:page element before that node in the same section, I can't
depend on there being a lexicon:page entry in the preceding
lexicon:section as there can sometimes be multiple lexicon:section
nodes without a lexicon:page. In fact, the last page element might be
in the preceding part/section/page node and not in any of the
preceding section/page nodes. Here is an example of the xml data (with
some stuff removed for illustrative purposes):

<section id="f.cf">
<entry id="f.cf.aa" type="root" cite="full" form="false">
<w strong="8308">שָׂרַךְ</w>
</entry>
<entry id="f.cf.ac" cite="full">
<w strong="8310">שַׂרְסְכִים</w>
</entry>
</section>
<section id="f.cg">
<entry id="f.cg.aa" type="root" cite="full" form="false">
<w strong="8311">שָׂרַע</w>
</entry>
</section>
<section id="f.ch">
<entry id="f.ch.aa" type="root">
<w strong="8313">שָׂרַף</w>
</entry>
<page p="977"/>
<entry id="f.ch.ab" cite="full" mod="I">
<w strong="8314">שָׂרָף</w>
</entry>
</section>

As you can see, I can easily associate Page#977 with Strongs#8314 as
it is a sibling. However, for Strongs#8308-8311, there is no page
element in the section and I'm not sure how to get the page element
value from a node an arbitrary number of sections before the node that
Strongs#8308-8311 are in. Any suggestions as to the best way to deal
with this?

[David Troidl]

Hi Ze'ev,

On 5/17/2010 8:23 PM, Ze'ev Clementson wrote:
> Hi David/Efraim,
>
> I'm writing an xslt script to extract the related Strongs# and
> BDBPage# values; however, I'm having a bit of a problem with the
> correct xpath syntax so I'm hoping that one of you can give me some
> advice. If I use:
> <xsl:value-of select="preceding-sibling::lexicon:page/@p"/>
>

There's also a 'preceding' axis. It just follows document order. (Or I
could send you the file that has page numbers on every part, section and
entry, if that would be better.)

Peace,

David

[Ze'ev Clementson]

Hi David,

On Mon, May 17, 2010 at 5:35 PM, David Troidl <David...@aol.com> wrote:
> Hi Ze'ev,
>
> On 5/17/2010 8:23 PM, Ze'ev Clementson wrote:
>>
>> Hi David/Efraim,
>>
>> I'm writing an xslt script to extract the related Strongs# and
>> BDBPage# values; however, I'm having a bit of a problem with the
>> correct xpath syntax so I'm hoping that one of you can give me some
>> advice. If I use:
>> <xsl:value-of select="preceding-sibling::lexicon:page/@p"/>
>>
>
> There's also a 'preceding' axis.  It just follows document order.  (Or I
> could send you the file that has page numbers on every part, section and
> entry, if that would be better.)

Ah (slaps forehead), that did it - thanks!

- Ze'ev

[Ze'ev Clementson]

Hi all,

The latest version of my Hebrew Bible iphone/ipad app now has support
for the Brown–Driver–Briggs Hebrew Lexicon (BDB). This is as a result
of the work that David Troidl has done in creating a BDB outline
linking words defined in the Brown–Driver–Briggs Lexicon to their
Strong's Concordance numbers. By having a cross-reference of the
Brown–Driver–Briggs pages to specific Hebrew words from the Tanach, I
have been able to link directly to the relevant Brown–Driver–Briggs
page for each Hebrew word. This should be useful for anyone either
learning or wanting to improve their knowledge of Biblical Hebrew.
Anyone interested in this can get more detail (and screen shots) from
my blog post:

http://beresheit.blogspot.com/2010/05/bdb-support-in-hebrew-bible-iphoneipad.html

Thanks again to David for all of his work with both the Strongs and
BDB xml data!

- Ze'ev

On Wed, May 12, 2010 at 5:46 AM, David Troidl <David...@aol.com> wrote: