RE: Recent News--not exactly music

[Reyzl Kalifowicz-Waletzky <reyzl...>]

Joel,

I haven't heard the latest on this project, but I am sure that your guess is 
wrong.

As of two years ago, the latest version of Ligature program allowed 98% 
accuracy in scanning of Hebrew and Yiddish texts, a figure that was equivalent 
to the best English programs at the time.  The 98% figure may have just risen 
to 99% or else Aaron must have finally found some time to write the grant 
proposal, or else, a large fish finally took the bait.   This project is a 
long-time coming at a very busy organization.  I was someone who reviewed 
Ligature when the soft-ware program first came out, I think, around 1993 or so 
for Aaron in the hope of doing this very project, but Ligature was just not 
good enough then (only 50% accuracy then).   It is very good now and the 
project can finally be done.   It's terrific news.  


Reyzl Kalifowicz-Waletzky 


----------
From:  Joel Bresler[SMTP:jbresler (at) ultra(dot)net]
Sent:  Wednesday, December 09, 1998 11:12 AM
To:  World music from a Jewish slant.
Subject:  Re: Recent News--not exactly music

Judy, 

I think your correspondent is mistaken about the technical aspects of this
important project. My guess it that they are scanning the books and
capturing "images" of the individual pages but NOT using character
recognition to input the contents into a database or machine searchable
corpus (the latter is a MUCH bigger task.)

There is no need for a markup and file format standard if they are taking
the latter route.

Does anyone else know for sure?

Best, Joel 

At 10:51 AM 12/9/98 -0500, you wrote:
>Forwarding this from the Book People list. Hope it is of interest....
>Not exactly music.....but reminded me of our former topic of Yiddish and
>standards..... 
>perhaps relevant to those talking about standards for transliterations and
>so forth as these items are digitized and catalogued. j.
>
>>Date: Mon, 07 Dec 1998 18:44:26 -0500
>>From: Gary Young <younggr (at) echo-on(dot)net>
>>Subject: Recent News
>>Sender: spok+ (at) gs1(dot)sp(dot)cs(dot)cmu(dot)edu
>>To: "'spok+bookpeople (at) cs(dot)cmu(dot)edu'" <spok+bookpeople (at) 
>>cs(dot)cmu(dot)edu>
>>Source-Info: Sender is really spok (at) ux4(dot)sp(dot)cs(dot)cmu(dot)edu
>>
>>
>>1998 Dec 07
>>The following item appeared in today's Toronto _Globe and Mail_
>>
>>"Fewer than 1 per cent of Jews still read Yiddish.  In January, a project
>will be launched to put the literature into computers.  "Yiddish will be
>the first literature to be 100 per cent digitized," Aaron Lansky says.  The
>digital library will be named after filmmaker Steven Spielberg, who donated
>$500,000 (U.S.) to the project."
>>
>>Good luck to them.
>>
>>Maybe they can establish a de facto standard re markup and file format.
>>
>>Gary
>>--------------------------------------------------------------------
>>This message was sent via the Book People mailing list.
>>Posting address:                 spok+bookpeople (at) cs(dot)cmu(dot)edu
>>Admin. and Unsubscribe address:  spok+bookpeople-request (at) 
>>cs(dot)cmu(dot)edu
>>Charter:                         http://www.cs.cmu.edu/~spok/bookpeople.html
>>
>
>


Joel Bresler
250 E. Emerson Rd.
Lexington, MA 02420 USA

Home:   781-862-2432
Home Office:    781-862-4104
FAX:            781-862-0498
Cell:           781-622-0309
Email:          jbresler (at) ultra(dot)net