[nabs-l] [nfbcs] new open source Braille translation software for the blind

Steve Jacobson steve.jacobson at visi.com
Sun Oct 11 21:37:49 UTC 2009


Joe,

The fact that I got excited when I saw the subject and then disappointed when I didn't find an announcement of an 
open source translation package already available probably says it all.  While I understand very much the complexity of 
braille translation and why translators cost so much, it is hard to swallow spending what one spends for a braill printer 
only to find one usually has to spend another $500 to use it.  There are exceptions of course.

As a person somewhat familiar with the development and evolution of NFBTRANS, I recognize that we need something 
more than NFBTRANS offers.  Be careful, though, not to under-estimate the difficulty of developing a good translator.  
You would not believe the number of strange things we had to debug and correct in NFBTRANS over the years.  Some 
problems don't make themselve apparent until someone runs into a particular situation that uncovers the problem.  
Translating a long document can really give a translator a workout and expose shortcomings in a way that doesn't 
happen with other software.  A translator really has to be solid for people to count on it.

If a better program than NFBTRANS becomes available, I would be the first to loudly applaud.  This is an old program 
that has already served people well.  Please consider two things.  First, I think there are some good possibilities for 
improving NFBTRANS without re-inventing translation.  There is a lot of room for a more complete Windows interface 
than has been developed with WINTRANS.  WINTRANS is an extremely good effort but development has stopped.  In 
addition, I think that the possibility of adding the ability to get formatting from Word documents would add a lot of 
usefulness to NFBTRANS.  These would both be projects that could be manageable without having to start from 
scratch.  NFBTRANS has already been compiled under LINUX and UNIX, so the possibility of adding some interfaces 
there exists as well.  

As stated above, NFBTRANS is old code.  If you are really up to doing a translator from scratch, please consider looking 
at the NFBTRANS source code, not as a model, but to see what kinds of problems that have been encountered and 
solved.  Also, look at the NFBTRANS table to get a clear feeling of how many different things have to be considered to 
do good translation.  Maybe you have already done something like this, but I can only say that I am a programmer and 
was still surprised at all of the different types of rules that had to be developed to do good grade II translation.  

The fact that you are interested is exciting.  Please let us know what you decide to do and if any of us can be of help.  

Best regards,

Steve Jacobson

On Sun, 11 Oct 2009 03:45:15 -0600, Joseph C. Lininger wrote:

>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA256

>Hi all,
>Please forgive the cross-post, but I wanted to reach the students since
>it's the group of the NFB that I know the largest number of, as well as
>the computer science people in hopes of eliciting a response from them.
>I will warn you now that this message is a little long, but I'd
>appreciate if as many of you as possible read it and provide me with
>your feedback. Whether or not I attempt this project will be directly
>influenced by how much feedback I get on the idea.

>I've been kicking an idea around in my head for a while and I want to
>know if there would be a large enough group that would find it of use.
>I'm thinking of possibly writing a completely extensable, flexable, and
>open source Braille translation package. I know about the nfbtrans
>software already, and it's a good peace of software, all be it a little
>dated at this stage of the game. What I'm thinking is a translation
>package that takes a different aproach. The major features would be as
>follows.

>* Modular design, allowing for Braille formats to be easily supported.
>Grade 1, grade 2, Nemith, and even foreign language Braille should not
>require complete rewrites of the software. All that need happen for a
>new format to be supported is for someone to write a module for it.
>Depending on what my research reveals, I might even be able to provide a
>generic module that allows for simple one to one print to braille
>mappings for some of the simpler Braille codes.

>* Modular design allows for a wide variety of Braille printers to be
>supported. At first, all that will be supported is to print a document
>directly to the serial or parallel port, but in the future support for
>the advanced features found on individual printers will be supported.

>* Command line interface at first, with the eventual goal of having a
>fully featured editor with Braille translation such as that which you
>find in commertial translation packages.

>* Software is able to be used directly, such as by a student or
>professional wanting to translate a document into Braille, or as a back
>end solution, being called by some other program to perform translation.

>* Licensed under the GPL or similar license. This allows people to use
>the software, free of charge, as well as to modify it. It is insane that
>translation software is so expensive.

>* High quality. Performance, features, etc. should be as good or better
>than that available in the commertial products.

>So, here are my questions to you all.

>1. Is there still a need for such software, or have devices like the
>braille note made it something that is no longer strictly necessary?

>2. If such software existed, would you use it personally or do you know
>anyone who would?

>I'll look forward to hearing back from all of you!
>- --
>Those of you who think they know everything are very annoying to those
>of us who actually do.
>Joseph C. Lininger, <jbahm at pcdesk.net>
>-----BEGIN PGP SIGNATURE-----
>Version: GnuPG v1.4.9 (MingW32)

>iQEcBAEBCAAGBQJK0akrAAoJEMh8jNraUiwqPpgH/2/ds8unDsCFSDNVAyO6s7hl
>4ChYqGjD9t6+SheiraCRULCuvEBCJb7avBban1XK1bHHVsNxKSA6BukBunLYMATt
>hFDrT12HmHRwbgB2LjYTKIqFXLnYB8x0huzbJyy51AiavqFP5o7NbPQkhN7fvBDy
>UMMyH7VLZzuNyWidnR5rx4Ugxbc5NcnsFq1czkooc3qKEcd0MTg/xu+QAMF2DLyz
>uv8wxEQYEZx5cEpM6GxjwBEZkjK5ZNukay5YxEzMD93qxnQyHibAdqVa9pMG1pwy
>9KfQQ+AeyJOS4vRNf+PzhQ21uHNDxMtztS3Fs+G4+0Xp9nrzVsFUsR0cUCPXVQ4=
>=nmE8
>-----END PGP SIGNATURE-----

>_______________________________________________
>nfbcs mailing list
>nfbcs at nfbnet.org
>http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>To unsubscribe, change your list options or get your account info for nfbcs:
>http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/steve.jacobson%40visi.com











More information about the NABS-L mailing list