Bible names database

[Robert Hunt]

Dear all,

    Recently I had the laborious task of checking the spelling and consistency of all the proper nouns (mostly names of people and places) in the Scriptures that we helped translate. It made me think that there must be a better (i.e., lazier  ;-) ) way using a program. But a program can't do it without a database.

    So I am proposing creating a open-source database with the following fields:

• Name (original language lemma)
• Alternate forms (original language inflected or variant forms)
• Language (Hebrew, Aramaic, Greek) What about things like Hebrewized-Egyptian, etc., e.g., Joseph's adopted name Zaphenath-paneah ???
• Type (person, deity, place, river, mountain, tribe, title (like Pharaoh), etc. -- some people's names became place names so they can be both person and place!)
• Gender (for a person)
• Race (if known, for a person)
• Derived from (optional, e.g., Ishmaelites comes from Ishmael)
• English (and room for other) transliterations (or can this be derived programmatically?)
• Translated name (in any specified Bible version / language)
• Scripture references (using what versification system? Do we need a separate field to make it easier to separate NT/OT/Apocrypha?)
• Pronominal references (i.e., other verses where pronouns are used to refer to this name)
• Root(s) (if known)
• Sounds like (if known)
• Related names (e.g., Jesus / Joshua)
• English (and room for other languages) meaning (if known)
• Encylopedic information (i.e., what else is known about this; English)
• Problems (textual or semantic difficulties in using this name, e.g., should we transliterate or translate the meaning of the name)

What else?

Some questions:

1. The best way to handle double-word entries, e.g., Pontius Pilate
2. The best way to handle OT/NT (Hebrew/Greek) variations on the same name
3. Do alternatives get their own full entries, e.g., Abram / Abraham, Mesopotamia / Padan-aram
4. Would it be more useful with a unique identifier for each entry, e.g., P0001 (then it could conceivably eventually include unnamed people like "Manoah's wife" = Samson's mother)
   
5. Is there a defined way to add point or area coordinates (GPS style) so this DB could connect to maps?
6. Should/could this link to other Open Scriptures databases, esp. the Hebrew/Greek texts (or vice versa)?
   

Also, would include any necessary Python scripts for changing formats or sort-key, separating OT, NT, Apocrypha, etc.

One way to actually do it rather than hand-editing XML would be to use the SIL Toolbox (dictionary) program, and save the files in Standard Format (Unicode). Then they could easily be converted to XML with a simple script.

Apparently there are over 3,000 unique proper names in the OT and NT not counting the Apocrypha.

What do you think? Worth trying? If so, please add your comments and suggestions on the structure.

Blessings,
Robert.

[sderose]

Hi! A friend just forwarded this to me (I have now joined the list).

I think I can save you a lot of work on this. I've been wanting such a
database for years, and finally worked on it seriously last spring. I
got most of the names in, though eventually burned out (it's awfully
tedious work after a while). But, I was able to get help from the
excellent folks at ccel.org, and a summer student there finished it.

What we now have is an OWL database of close to 4000 thousand names,
with many variant spellings and forms, and with information to
disambiguate them (there are about 30 Zechariahs, for example).
There's a specific sequence of rules we followed for how to create an
unambiguous form (the first of them is to append the father's name if
available, following "_" (I think "_" is nice because you can
pronounce it "bar", which is Hebrew for "son of"); but there are
several more rules to cover cases where that's not sufficient (turns
out that when available, the father's name is almost always
sufficient).

The entries for most individuals also have fields for father, mother,
gender, occupation (king, prophet, warrior, savior,...) , tribal
affiliation, etc. where applicable and available. There's also at
least one, and often all, verse references where the person is
mentioned. They are also linked to the names (and disambiguation
numbers) used in multiple published "all the people in the bible"
kinds of reference books, and to the particular translation that uses
the spelling. Alternates use the OWL "SameAs" consruct to link back to
one "main" spelling, and ambiguous names have a separate entry
(without any suffix to disambiguate), that specifically says they're
an AmbiguousName, and lists the people that spelling can refer to.

No doubt it has some errors (though OWL can check for many kinds of
inconsistencies), and there are some cases where you simply can't tell
what a name is -- for example in Nehemiah there are many names that
might be either cities or people. But the data is very thorough. The
data is in Manchester OWL syntax, which is awfully easy to convert to
anything else one might want, as well as being quite human-readable as-
is, and supported by appropriate applications such as the (open-
source) Protege' ontology editor.

I have additional data on cities, mountains, rivers, and so on; people
groups; languages; month names; artifacts; Bible translations; Canon
lists; and a variety of other object classes (have planned but barely
begun a comparable database of HB/GR/AR words). They classes are
linked directly in to the standard SUMO ontology, widely used in
ontological R&D.

The data does not have a few of the items you suggested, such as an
English gloss for the "meaning" of the name, or links to "related"
names (except that it *does* have alternate spellings and alternate
names for the same individual (for example "Simon" vs. "Peter", etc.).
And no "encyclopedic" information.

This data is freely available to use; I do ask that enhancements and/
or corrections be sent back so I can incorporate them and avoid having
a thousand slightly-varying versions all over the place (just because
such variation is more pain for everyone!). I meant to put this up
already for public access, but have been a bit slow on the final
proofreading.

Let me know if you're interested.

Steve DeRose

Oh, a sample entry (see what I mean about easy to read/parse?):

Individual: Abagtha
Types: Human
Facts: Tr KJV, Gender "M", Profession "Chamberlain",
Affiliation Ahasuerus

[David Troidl]

Hi Robert,

Sounds like an ambitious project.  I've been working on matching Strong's Greek entries with Liddell-Scott.  In general, Liddell-Scott doesn't list proper names, and I came up with a little over 400 entries of that type.  I'm looking to do the same thing for the LXX.  So a Biblical name database would fill the void nicely.

Doing a Google search for 'Biblical proper names', the first two hits were for the International Standard Bible Encyclopedia (ISBE) Proper Names section, and Hitchcock's Bible Names.  Both are available from the SWORD Project.  So that may be a good starting point, to check out what information is available and what format they use.

Peace,

David

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

[Robert Hunt]

On 08/09/10 01:24, sderose wrote:

Hi! A friend just forwarded this to me (I have now joined the list).
  

Great!

I think I can save you a lot of work on this. I've been wanting such a
database for years, and finally worked on it seriously last spring. I
got most of the names in, though eventually burned out (it's awfully
tedious work after a while). But, I was able to get help from the
excellent folks at ccel.org, and a summer student there finished it.
  

Yes, I was planning to research for what's already freely available to build on.

What we now have is an OWL database of close to 4000 thousand names,
with many variant spellings and forms, and with information to
disambiguate them (there are about 30 Zechariahs, for example).
There's a specific sequence of rules we followed for how to create an
unambiguous form (the first of them is to append the father's name if
available, following "_" (I think "_" is nice because you can
pronounce it "bar", which is Hebrew for "son of"); but there are
several more rules to cover cases where that's not sufficient (turns
out that when available, the father's name is almost always
sufficient).
  

There are many OWLs, but I presume you're talking about the Web Ontology Language. I'm aware of that but can't say I'm very familiar with it, so it would be good learning for me. And I'm certainly a fan for using well-accepted standards. (I was thinking XML.)

The entries for most individuals also have fields for father, mother,
gender, occupation (king, prophet, warrior, savior,...) , tribal
affiliation, etc. where applicable and available. There's also at
least one, and often all, verse references where the person is
mentioned. They are also linked to the names (and disambiguation
numbers) used in multiple published "all the people in the bible"
kinds of reference books, and to the particular translation that uses
the spelling. Alternates use the OWL "SameAs" consruct to link back to
one "main" spelling, and ambiguous names have a separate entry
(without any suffix to disambiguate), that specifically says they're
an AmbiguousName, and lists the people that spelling can refer to.
  

Yes, I was hoping this work would eventually lead to something like a brief encyclopedia of Bible facts and it seems like your work is a good start on that.

No doubt it has some errors (though OWL can check for many kinds of
inconsistencies), and there are some cases where you simply can't tell
what a name is -- for example in Nehemiah there are many names that
might be either cities or people. But the data is very thorough. The
data is in Manchester OWL syntax, which is awfully easy to convert to
anything else one might want, as well as being quite human-readable as-
is, and supported by appropriate applications such as the (open-
source) Protege' ontology editor.
  

Yes, the structure MUST  handle ambiguity, i.e., unknowns and also multiple names for the same person/object.

Hadn't heard about Protégé before. I'll check it out. I see it does RDF and XML exports as well.

I have additional data on cities, mountains, rivers, and so on; people
groups; languages; month names; artifacts; Bible translations; Canon
lists; and a variety of other object classes (have planned but barely
begun a comparable database of HB/GR/AR words). They classes are
linked directly in to the standard SUMO ontology, widely used in
ontological R&D.
  

Nice.

The data does not have a few of the items you suggested, such as an
English gloss for the "meaning" of the name, or links to "related"
names (except that it *does* have alternate spellings and alternate
names for the same individual (for example "Simon" vs. "Peter", etc.).
And no "encyclopedic" information.
  

I don't see any references to the original language spellings in your description?

It seems your database assumes English. Because of my cross-cultural work, I was hoping to do something that could have other major languages added (French, Spanish, Portuguese, etc.) and even much smaller languages if someone really wanted to do it. (Certainly minority language Bible translations are often greatly affected by the dominant languages and commonly available Bibles in the area.) This means marking many fields with a language marker. Maybe OWL assumes English also??? (Sadly, many Western inventions don't realize that there's other worlds out there.) Having it multilingual gives greater universality at the expense of greater complexity.

This data is freely available to use; I do ask that enhancements and/
or corrections be sent back so I can incorporate them and avoid having
a thousand slightly-varying versions all over the place (just because
such variation is more pain for everyone!). I meant to put this up
already for public access, but have been a bit slow on the final
proofreading.
  

It's great if you are prepared to offer the data to be publicly available with an open license. And I would be happy in theory to send back enhancements and corrections. I'm just concerned that if we take the data and rework the format too much, anything sent back might not be recognizable or of use to you???

Let me know if you're interested.
  

I'm certainly interested in investigating more, and I'll also wait with interest to hear comments from others on this list who're probably more knowledgable than me in many of these areas.

Steve DeRose
  

Can you tell us a little more about yourself and the background of this database? (For me, I have just finished a partial Bible translation for a smallish cultural group in Asia.)

Oh, a sample entry (see what I mean about easy to read/parse?):

Individual: Abagtha
        Types: Human
        Facts: Tr KJV, Gender "M", Profession "Chamberlain",
Affiliation Ahasuerus
  

Thanks. Can you just give us a few more different kinds of entries to get a better feeling of the structure and contents.

Robert.

[Art Bolstad]

I am excited about this project.  We can use it directly in a translation for the Tandroy language (a variant of Malagasy) which is in a checking of the whole scriptures right now.

For a working document, I can see a need to put into each entry the language specific decision on what to do (transliterate [how], translate [sinc Hebrew names usually have a meaning], how to handle senior/junior issues etc),

How can I get a copy for our use?

art bolstad

[Josh Lipman]

I'd love the link if you wouldn't mind sending it!

[Robin Riley]

Wow!

Steve, that sounds like a great project, would you please send me a link to it;

I'd greatly appreciate it...

Thank you, Robin

[p.juanmig...@gmail.com]

Hello. I'm also very interested on that list Steve made. Would you be so kind to share the link with me? Thank you very much.

[anonymous]

Hi,

I am also interested in the list. Are you willing to share the link with me?

V.

Op maandag 15 augustus 2016 08:00:46 UTC+2 schreef p.juanmig...@gmail.com:

[Robin Riley]

Yes. please... I too would be greatly interested in this list ... or link.

Thank you, in advance,

Robin Riley