[sword-devel] How about the OT's standard versification at Sword ?

Chris Little chrislit at crosswire.org
Wed Apr 20 04:26:15 MST 2005

C. N. R. wrote:
> I've sent an answer to Mr Little but, not unexpectedly, he didn't reply. 
> Well, he is strongly defending his own view of the purpose of the SWORD 
> Project, even if everyone else seems to have a different opinion.

I don't know which email you're referring to that you feel I failed to 
answer, but realize that answering your emails is not my sole purpose in 
life. Besides, your emails tend to repeat the same points over and over 
and over and over and over and over again.

> I am impressed by the fact that Mr Little wants to rebuild the Romanian 
> Orthodox Bible (ROB) module, a daunting task for someone who doesn't 
> understand the Romanian language. <blah blah blah...>

Bibles all have (roughly) the same 66 books (give or take a testament). 
And every book is made of some number of chapters. And every chapter is 
made of some number of verses. In the case of the Romanian Orthodox 
Bible presented on the pages you sent me, there is not much other markup 
involved. In any case, it really has nothing to do with language. 
Remember, we're trying NOT to change or re-order the text like you were. 
So knowing or not knowing the language of the text is irrelevant.

> I hear that Mr Little has begun work on transforming the Vulgate 
> (Clementine version) into the new style. I don't think it will be an 
> improvement.

What new style? Improvement from what?

> Indeed, it is likely to be as wrong as the Estonian module, 
> already on the SWORD official site, which defectively repeats the same 
> verse from Exodus to Job.

To repeat, again, the problem with the Estonian Bible module is that it 
uses the last valid verse for all following empty verses. It was caused 
by a bug in an import utility. It does not, in any way, indicate that 
text is missing from the module that occurred in the source documents. 
I've written this a couple of times (every time it comes up, in fact), 
so I wonder if you are simply not reading my messages.

> <blah blah blah> I really don’t see how Mr Little can deny that the 
> KJV standard is the basis for the entire SWORD project; this just flies 
> in the face of the evidence.

I don't exactly deny that. The KJV versification (v11n) system is the 
basis for Sword's internal v11n (at the moment, for all intents and 
purposes). It's a known problem that some of us are working to correct.

However, I'm still not going to accept Bibles that actually have some 
other v11n, but have been altered to approximate the KJV v11n system.

<blah blah blah>

> But Mr. Little hasn't refused only me, he also refused a Polish module 
> creator, Mr Wojciech, (who made the Biblia Gdanska and other Polish 
> modules), Mr Artemis State (with D. Cornilescu's complete Romanian 
> version: the one on the CrossWire site is defective at I Peter 2-5 and 
> it's in an old orthography) and I presume others too.

This is just false. These modules were not rejected, by any means. Of 
course, that was on the assumption that their versification systems 
hadn't been altered. If they have, then we would prefer versions without 

> In an earlier communication, Mr. Little wrote: "I would prefer that we 
> simply be given links to source material so that someone from CrossWire 
> can work on the modules themselves. I believe it is less likely that 
> they will editorialize as they do markup & conversion." So we -- the 
> creators of the other modules -- should give up making SWORD modules, 
> and write HTMs instead ? It's a pity that the people at CrossWire don't 
> accept the contributions of learned and meticulous volunteers, who fully 
> understand their texts, only to give themselves the same work, starting 
> from zero and without understanding the texts. It seems discriminatory, 
> to say the least. And I am still worrying about what will happen to the 
> Romanian Orthodox Bible in the hands of someone who has no idea of the 
> Romanian language at all.

Maybe I was a little harsh in that statement, but I'll explain. We want 
our module sources to be as close as possible to the original electronic 
versions of the text. Every time someone converts a document from one 
markup to another, there is a possibility that errors will creep in and 
that information will be lost. So we want whatever is nearest to the 
original since it will likely have the fewest errors and the highest 
rate of information retention. Furthermore, in my experience, 
individuals who have never encoded a document in XML do not do a very 
good job the first time. The effort I have to expend to clean up a 
document from a first-time encoder is usually more than the effort I 
would have to expend to create the document from scratch from their 
sources. In addition, I have a lot more experience with OSIS and can be 
much more confident that I am not accidentally losing information in the 
process of conversion.

We're not suggesting that people make HTML files, specifically. In the 
case of the Romanian Orthodox Bible, your sources were online, so simply 
mailing us the address would have been optimal. People who are creating 
their own texts from scratch (typing or scanning), can give us documents 
in whatever format they like (HTML is fine, as are Word, RTF, or even 
OSIS if they have the ability).

> Taking the versification out of their skilled hands and 
> transferring it to an unproven central system of limited linguistic 
> skill seems a shabby reward for their work, and a recipe for confusion 
> in the future.

See... here's the problem. You seem to think you have a right to decide 
on the versification of a Bible that you didn't translate. We don't 
believe you do. We believe that decision was already made by its actual 
translators and that you have no right to impose your own opinions on 
their work.

 From an entirely practical perspective, it is not reasonable to believe 
that work done by hand will be without mistakes. People who work on 
repetitive tasks (like adjusting a versification) get tired and begin to 
make mistakes.

There is a limited set of about 6 versification systems that are used by 
almost all Bibles. (The exceptions are mostly modern Bibles that differ 
from these systems in only a few verses.) The mappings between the NRSV 
system (basically identical to the KJV system) and the other 5 systems 
are known, documented (by people with more authority than you or me), 
and available. When we support other v11n systems in Sword, we will use 
that data to convert between systems rather than whatever ad hoc methods 
you might be imagining.


More information about the sword-devel mailing list