[sword-devel] NASB status

Thu Jan 7 06:15:42 MST 2016

I really don’t want to encourage the flames.

I was asked to modify the C++ source. I did that. I didn’t rewrite it. It is still C++. I think it can be diffed with the original to understand my changes. But I don’t remember if that is the case when I updated it for the latest Lockman input.

Greg’s work is complete. BibleDesktop has a publicly available beta that works with it. (It is not appropriately packaged for each platform.) Existing BD can be patched by a user placing an xslt file in a particular place on their computer.

Regarding JSword, it has full support for the NASB. STEP and AndBible have been released with such support. BibleDesktop beta 2.0 is available with support. I’ve got a couple of nasty bugs in zLD modules that have to be squashed. (I’ve squashed a dozen or so other bugs in the last month.) Other than that, it is merely a packaging issue. So, a version of BD is available for the NASB today. A user may have to run it from a shell script.

I also have done the naslex work. It is written in Perl. Is it complete? I doubt it, as I stopped when my nasb effort wasn’t acceptable. From what I remember:

The markup of the lex is in TEI. This was one of the first dictionaries in TEI. It may need to be examined to see if it still is appropriate.

There was a question to Chris Little, way back then, about how to do linking within the module. I don’t think there was an answer.

There are entries of the form Hnnnn!b (that’s what is in the NASB; in the lex it is Hnnnnb), and the SWORD engine was changed to handle these. The SWORD engine was also modified to handle these when using any of the current Strong’s Number dictionaries.

IIRC, Lockman wanted a single module for the lex. We talked about having a single module with both H and G entries. That is the key would have a leading H or G. This of course would have Greek before Hebrew. I don’t know if the SWORD engine was changed to allow for this.

If I recall, BibleDesktop, SwordWeb and The SWORD Project for Windows displayed the “word" differently. I don’t remember whether it was the Greek or the Hebrew. Best as I can remember, it was an “accented” capital letter.

I don’t remember whether there was a display problem with JSword frontend.

On the CrossWire server, for those that have accounts, the naslex work here: ~dmsmith/newmods/lockman/nasec/naslex

Together in His Service,
	DM

Here is an example of the TEI:
<superentry id="H8163" key="H8163"><entryFree><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair</pron> or</pron><orth xml:lang="he">רעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972c">972c</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">hairy</hi></def>:― <usg>hairy(3), shaggy(2).</usg></entryFree><entryFree><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972c">972c</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">male goat, buck</hi></def>:― <usg>goat(6), goat*(1), goats(3), male(25), male goat(9), male goats(1).</usg></entryFree><entryFree><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972d">972d</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">a satyr, demon</hi></def>:― <usg>demons(1), satyrs(1).</usg></entryFree></superentry>
<entryFree id="H8163a" key="H8163a"><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair</pron> or</pron><orth xml:lang="he">רעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972c">972c</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">hairy</hi></def>:― <usg>hairy(3), shaggy(2).</usg></entryFree>
<entryFree id="H8163b" key="H8163b"><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972c">972c</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">male goat, buck</hi></def>:― <usg>goat(6), goat*(1), goats(3), male(25), male goat(9), male goats(1).</usg></entryFree>
<entryFree id="H8163c" key="H8163c"><orth xml:lang="he">ריעִשָׂ</orth> <pron>sair (<xr type="xref"><ref target="BDB:972d">972d</ref></xr>)</pron>; <etym>from the same as <xr type="see">[<ref target="H8181">8181</ref>]</xr></etym>; <def><hi rend="italic">a satyr, demon</hi></def>:― <usg>demons(1), satyrs(1).</usg></entryFree>

> On Jan 6, 2016, at 5:04 PM, Troy A. Griffitts <scribe at crosswire.org> wrote:
> 
> I hesitate to even respond to this thread, as none of this kicks the can down the road, but... I believe some of the issue has to do with typical open source volunteer problems. Specifically, in this case, none of the pumpkin holders was satisfied to build on their predecessors' code. I originally made the initial co version in C++. My script is versioned in the sword-tools repo. I'm not sure what Chris did. I know DM rewrote the conversion in Java, and it sounds like Greg has again rewritten the conversion in Python. It is not a matter of perfection before release. It is a matter of basic satisfaction of the three items I have listed and we have not attainted basic satisfaction yet.
> 
> Troy
> 
> On January 6, 2016 2:51:55 PM MST, "Matěj Cepl" <mcepl at cepl.eu <mailto:mcepl at cepl.eu>> wrote:
> On 2016-01-06, 18:53 GMT, Troy A. Griffitts wrote:
>  This is a commercial module to be sold by Lockman.  That is 
>  a different scenario from other modules.  For this module, 
>  I have reasonably asked:
> 
> I didn’t like Karl’s tone last time the flamewar went around, 
> but I have to admit he is right. Whatever you say is suspicious 
> because of one argument: “Twelve years”.
> 
>  1) That we have a scripted, reproducible way to transform 
>  their data from their pristine source to a module.
> 
> Yes, what’s the problem? Whom you asked for help (e.g., I have 
> developed completely automatic conversion of CzeCSP from one XML 
> to OSIS, and it didn’t take me twelve years to do it)? Twelve 
> years.
> 
>  2) That we support their entire dataset which they have given 
>  us for this module (base text + footnotes / crossrefs 
>  + lexica)
> 
> Yes, I have CzeCSP with footnotes, crossrefs, and notes. Twelve 
> years.
> 
>  3) That the result works generally in all major SWORD/JSword 
>  frontends.
> 
> What are the problems? Where are the bugs? Where is the progress 
> on development of the module recorded? I would be willing to buy 
> the module, if that was required to work on it. Twelve years.
> 
>  It is not as simple as-- just release it and fix it later.  
>  This is to be sold by Lockman.
> 
> What do you mean? I work for Red Hat. We sell software in tune 
> of some, let's say, two billions of USD. Of course, we don’t 
> expect it to absolutely perfect and bug-free. If there are bugs 
> in the module, we can certainly provide a fix. Where is the 
> alpha version, how do you plan to make QA working? Twelve years.
> 
>  Historically, we have had at least 4 people own this effort 
>  over the years.  It is not simply that one person has sat on 
>  this and hasn't finished it for 12 years.
> 
> Yes, so where are products of work of all these people? What did 
> they do? I mean, I honestly believe that some work was done, but 
> without some presentation of the results, how can we know what's 
> done? Here “Twelve years” makes it even more difficult to be 
> persuaded.
> 
>  I think we are close.  In my mind, the current owner (Greg) 
>  simply needs to have a list of outstanding items which keep us 
>  from satisfying 1-3 and push each of them down the road until 
>  they are done.
> 
> And where is that list? That is my question.
> 
>  I don't know what those items are.  I am just concerned that 
>  we meet 1-3 before we give the data to Lockman to sell and 
>  I feel these 3 items are reasonable requests.
> 
> I don't think Lockman expects the work to be ever done at this 
> point, but that's another thing. If they do, they have my 
> admiration (or something else).
> 
> However, if you (or Lockman) expect that you will hand over 
> finished absolutely prefect module and Lockman will never ever 
> see you (or anybody from the Sword project) again, than I think 
> you are sorely mistaken.  Unless of course they are prepared to 
> maintain the module themselves. After all, even NASB itself is 
> (according to Wikipedia) in its tenth edition.
> 
> Blessings,
> 
> Matěj
> 
> -- 
> Sent from my Android device with K-9 Mail. Please excuse my brevity._______________________________________________
> sword-devel mailing list: sword-devel at crosswire.org <mailto:sword-devel at crosswire.org>
> http://www.crosswire.org/mailman/listinfo/sword-devel <http://www.crosswire.org/mailman/listinfo/sword-devel>
> Instructions to unsubscribe/change your settings at above page

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.crosswire.org/pipermail/sword-devel/attachments/20160107/d08e09b1/attachment-0001.html>