Here are some files from an MS-DOS Tanach package mentioned on s.c.j by Warren Burstein in message 1164@vaccine.UUCP. The originals can be found on wuarchive.wustl.edu in the directory mirrors/msdos/hebrew. Note that not all books are present. Also, some of the books seemed to have some junk (a partial repetition of the last verse) at the end, which I have removed. The files on wuarchive were obtained from the L'Chaim BBS in New York City. Attempts to determine the ultimate source of those files met with stonewalling from the BBS's sysop, Reuven Blau. Since he demands a membership fee before one can download from his BBS, I was unable to retrieve the missing books.

I have thrown out all the specifically messy-dos stuff and written some extremely simple programs to play around with the so-called 'codes,' so that Unix users can also investigate them. No disrespect to the Tanach is intended. I hope that no one will mistake this sort of thing for actual Torah study.

The texts themselves are in the file tanach.tar.Z. Use the command

zcat tanach.tar.Z | tar xvf -

to liberate them. You can then throw out tanach.tar.Z.

To compile all the c files, just type 'make'.

To create a raw copy of the Torah, suitable for use with 'skip', type 'make torah.raw'. 'make torah.rev_nums' will reverse the numbers in the file so that they look right on left-to-right displays.

The file masoretic.tar.Z is my attempt at bringing this Tanach (so far just the Torah) into line with the masoretic version. The Qere readings are still in an anarchic state, except for those in Bereshit, which are those found in the Koren edition of the Tanach. The changes were taken from the book _Keter Aram Tzovah ve-ha-nusach ha-mekubal shel ha-Mikra_ (The Aleppo Codex and the Accepted Text of the Bible) by Mordechai Breuer. The object was to create a text conforming to the accepted version among the Yemenite Jews, which Breuer argues is free from mistakes. The accepted text among Askenazi Jews differs from this text in 9 places, each single-letter differences. I cannot guarantee that this text is free from error, but you may prefer to use it rather than the text in *.tor.

The file ENCODING shows the Roman-to-Hebrew alphabet encoding used by the 'findstring' program. In general, all of the programs have the encoding used in the texts (Aleph = 128 decimal) hard-wired into them. The display program 'hebrew' maps these characters to start at 224, which seems to be a standard for terminals which display Hebrew. Corrections to this asumption would be appreciated.

If you need to make changes to any of the texts (say, to bring them into line with a particular manuscript of interest), use trans -p to obtain an ASCII copy, make any needed changes (be careful about the spelling of the letters -- if you have another prefered spelling just change the tables in trans.c and untrans.c), and then apply untrans. Of course if your editor can handle Hebrew none of this should be necessary.

Dan Rice



webmaster@shamash.org