Biblical Hebrew Forum

["Peter Kirk" <Peter_Kirk AT sil.org>]

As I understand it, the main reason for the SIL software not working
properly is that it was designed to work with the slightly different CCAT
data format rather than the OTA text. That is not a bug.

May I make a suggestion. John R has "what is, so far as I can see, a perfect
text". Perhaps either he or Henry can put this up on a website.
(Unfortunately I am not able to do so.) This would not seem to break OTA's
conditions, as it is derived from their earlier and less restricted
distribution. I think it would be OK with SIL, but please check the SIL
website for conditions on the SIL conversion stuff. This website would be,
as far as we know, the first public electronic distribution of the fully
pointed Hebrew text. It would save lots of other people trying to reinvent
the wheel which John has already successfully reinvented. (Fortunately no
one could patent the first version of this particular wheel, and they can't
copyright the original Hebrew MS.)

If you like, John, I will try to verify your "perfect text" against the text
which I have, which is derived from CCAT and so not distributable. It would
be interesting to see how many differences I find. If you would like me to
do this, please e-mail me the text as a zip file (off-list of course),
preferably this week while I have local access to my e-mail server.

Peter Kirk

PS: as the OTA text is ASCII text, there should be no problems editing it
with Word as long as it is saved as Text Only. The only odd thing which
might happen would be the double quotes for tsere being changed to curly
quotes, but Save as Text Only always converts them back to straight quotes
as required. Well, Word 2000 just might get confused if it incorrectly
guesses that the text is French or some other language which uses guillemets
(chevrons) for quotation marks. If you see this happening, disable the
automatic language detection settings and tell Word that the text is
English. DO NOT say that the text is Hebrew, this will confuse it completely
as it will expect the Unicode standard Hebrew encoding and not find it.

-----Original Message-----
From: John Richards [mailto:jhr AT universalist.worldonline.co.uk]
Sent: 13 November 2000 02:34
To: Biblical Hebrew
Subject: Re: Conversion of OTA BHS text


Henry

> I have no experience with the SIL software (I only ever used the
> little 1988 CONVBHS.COM program for the first step, and then custom
> hand-written text-processing scripts from there on out) -- but you'll
> never know whether the SIL software is buggy or not until you apply it
> to a BHS text known not to be corrupt (and any text with non-tildecode
> material on the same line following a tilde code, such as you have
> described, is in fact corrupt).

You are quite right, of course - and I did precisely that. I used for later
trials the version you have put on your site. As you say, it is far better
formatted, and avoids the need to reformat the position of the tilde
markers. However, it makes no difference to the end product using
the SIL software. Unless I went throught it at the last stage before the
conversion to Right to Left, there were always approx. 50 lines
indicated by the L2R process as being over 150 characters long, and
resulting in a garbled line in the final text (words in the wrong order, or
even omitted completely). It may be, as you suggest, that there is a bug
in the SIL software. It seems, in fact, rather likely. But the fact does
remain, that if one goes through the text at the penultimate stage visually
checking for long lines (which stand out and are easy to catch), one
then ends up with what is, so far as I can see, a perfect text. That is
all I am concerned about, both for myself and for anyone else who
wants to try and convert the OTA BHS using the SIL CC files. I have
no experience with any other conversion system. If there is a better
one, it might be a good idea to let members of the list know about it.
I would certainly like to know about it myself.

> I feel no need to ask the OTA
> people for permission, since I'm complying with the conditions that
> were placed on the text I downloaded when I downloaded it (as is amply
> and redundantly documented within the ZIP archive itself).

I have looked carefully at the conditions as stated in the files included
in your ZIP file, and compared them with the conditions now placed
on the OTA site. As Bob Barrett points out in a message to the list today,
the conditions have been changed significantly. I would say you are
perfectly justified in adhering to the conditions stated in the files as
you originally received them.

> I might
> feel slightly more inclined to comply with any retroactively-imposed
> _ex post facto_ changing of the rules that they've dreamed up if they
> were doing a good job of distributing the e-text from their site --
> but in fact they're doing an execrably wretched job of distributing
> the text from their site, making the functionality of their site
> critically-dependent on a mass of poorly-written and
> frequently-malfunctioning browser scripting code, and corrupting the
> non-SGML BHS e-text with a bunch of SGML garbage.

Quite so!

Best wishes

John Richards