[sword-devel] NASB module

Troy A. Griffitts scribe at crosswire.org
Tue Nov 11 14:35:35 MST 2008

Thanks DM.  Congrats with your daughter!

I'm gonna keep listing items here so we have a record of them.

it looks like our <transChange type="tenseChange"> is surrounding what 
we think is the entire verb by expanding to include the whole <w> tag.

Well, we're guessing at the words included in the <w> tag because NASB 
data only has milestone strongs numbers.

Thus text like <123> this will become <234> this:
<w lemma="123>Thus text like</w> <w lemma="234">this will *become</w> 

Guessing at the <w> is all we can do, but when we try to encode the '*' 
as a tenseChange we get this:
<transChange type="tenseChange"><w lemma="234">this will 

instead of this:

<w lemma="234">this will <transChange 

I'm guessing maybe <w> doesn't allow transChange?  Dunno.

But the fact is twofold.  We have enough info to capture the lockman 
data and produce a transChange on the correct word.  The current output 
from our software produces the wrong information to our user.  See, for 
example what John 11.11 looks like in swordweb.  Here is the lockman data:

{{43:11}}11 This<MG3778> He said<MG3004>, and after<MG3326> that He 
*said<MG3004> to them, <RS>``Our <RA>friend<MG5384> Lazarus<MG2976> 
<RB>has fallen<MG2837> asleep<MG2837>; but I go<MG4198>, so<MG2443> that 
I may awaken<MG1852> him out of sleep<MG1852>."<RT>

Here is out OSIS:

<w lemma="strong:G3778">This</w> <w lemma="strong:G3004">He said</w>, <w 
lemma="strong:G3326">and after</w> <transChange type="tenseChange"><w 
lemma="strong:G3004">that He said</w></transChange> to them, <q 
who="Jesus"><q level="1" marker="“" sID="q.7214"/>Our <note n="A" 
osisID="John.11.11.xref.A" type="crossReference"><reference 
osisRef="John.11.3">John 11:3</reference></note><w 
lemma="strong:G5384">friend</w> <w lemma="strong:G2976">Lazarus</w> 
<note n="B" osisID="John.11.11.xref.B" type="crossReference"><reference 
osisRef="Matt.27.52">Matt 27:52</reference>; <reference 
osisRef="Mark.5.39">Mark 5:39</reference>; <reference 
osisRef="John.11.13">John 11:13</reference>; <reference 
osisRef="Acts.7.60">Acts 7:60</reference></note><w 
lemma="strong:G2837">has fallen</w> <w lemma="strong:G2837">asleep</w>; 
<w lemma="strong:G4198">but I go</w>, <w lemma="strong:G2443">so</w> <w 
lemma="strong:G1852">that I may awaken</w> <w lemma="strong:G1852">him 
out of sleep</w>.<q eID="q.7214" level="1" marker="”"/></q>

And here is the swordweb output from:

11 This He said, and after *that He said to them, “Our friend Lazarus 
has fallen asleep; but I go, so that I may awaken him out of sleep.”

DM Smith wrote:
> Troy,
> This mostly falls on me to make the change. I am preparing for my  
> daughter's wedding on Nov 22 and have house guests for the entire week  
> following. Once that passes, I'll have lots of time to devote to  
> making final changes to the NASB module.
> The bigger issue to getting the NASB out, are the NAS Greek and Hebrew  
> lexicons. These have been provided to a select few for testing. These  
> are TEI modules and seem to work just fine in BibleCS and  
> BibleDesktop. In BibleCS they are quite plain because the filter is  
> rather plain (i.e. it handles very few tags). My guess is that it  
> works in a similar fashion in GnomeSword, MacSword and BibleTime.
> As to the paragraph marker, OSIS explains that <milestone> is used to  
> record information in the printed document (as well as a few other  
> discreet usages). AFAIR, the engine does not recognize the suggested  
> form of milestone. But I could be mistaken. I'd really rather use  
> something that is currently supported so that we don't have to wait  
> for another release of the SWORD engine and the front-ends.
> As to the <divineName> issue, there are two possible ways to mix it  
> with <w>:
> 1) As in the KJV <w><seg><divineName>Lord</divineName></seg></w>
> (Yes, w should change in the OSIS spec to include divineName!)
> 2) As in the NASB <divineName><w>Lord</w></divineName>
> I'll have to search my memory as to why I did NASB differently. It was  
> deliberate, solving a different scenario, but I don't remember exactly  
> what.
> In Him,
> 	DM
> On Nov 10, 2008, at 8:12 PM, Troy A. Griffitts wrote:
>> There has been a recent push to finish the NASB testing.  There have
>> been only a few people provided with the keys, as Lockman would expect
>> us to not make them freely available to a wide audience.
>> This discussion is regarding general module markup practice, so I am
>> posting here for comments.
>> Testing done with module from (I hope this is the latest):
>> lockman <PM> (paragraph marker) is being changed to:
>> <lb type="x-end-paragraph"/> <lb type="x-begin-paragraph"/>
>> The add too much whitespace in the frontends.  I'd rather use:
>> <milestone type="line" subType="x-PM"/>
>> lockman L\ORD<MH3068>/ is being changed to:
>> <divineName><w lemma="strong:H3068">Lord</w></divineName>
>> which seems fine but our KJV text uses:
>> <w lemma="strong:H03068"><seg><divineName>Lord</divineName></seg></w>
>> We need to decide what our internal OSIS will look like for this
>> construct and stick with it.  I don't like the <seg> but know why you
>> did it.  I would prefer the KJV markup and drop the <seg>.  I'll get  
>> the
>> OSIS guys to fix the spec.  We already support that markup the best
>> anyway (with or without the seg).  Currently swordweb doesn't work  
>> with
>> the NASB markup, and I'm guessing other filters also don't work.
>> _______________________________________________
>> sword-devel mailing list: sword-devel at crosswire.org
>> http://www.crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page
> _______________________________________________
> sword-devel mailing list: sword-devel at crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page

More information about the sword-devel mailing list