[Blindmath] Reading and Writing Math: LEAN versus MathSpeak

[Michael Whapples]

Having spoken with John on this while developing the code, may be I can 
give some ideas as to its benefits.

The first difference I see is that LEAN is a way of representing maths 
in files rather than a user interface language, from what I can tell 
MathSpeak is very much a user interfacing system specific to a 
particular form of communication (IE. speech). This point probably will 
become more clear as this message goes on.

LEAN is compact and character based: Many of the existing systems are 
very verbose and clunky to navigate through, LEAN is unicode character 
based so its much shorter should you need to cursor through it.

An example of this: Take the fraction x/2 in LaTeX written as a proper 
fraction it would be \frac{x}{2} a total of 11 characters, in LEAN it 
would be only 5 characters (one to start the fraction, the x, one for 
change to denominator, the 2 and one for the end fraction). The way 
mathplayer presents such equations to screen readers is even more 
verbose and clunky.

LEAN builds on unicode: Not really a huge user benefit but LEAN uses 
many of the characters which are already defined by unicode (eg. Greek 
letters, math operators, etc). This means LEAN only needs to define the 
structural stuff unicode doesn't and for that it uses the unicode 
private blocks. This means the user can use any unicode aware editor for 
LEAN.

Easy to convert to user interfacing formats: One of the ideas was to try 
and make the transformation between LEAN and a form the user can easily 
understand as simple as possible. The plan is to try and have it build 
on existing technologies for the transformation. An example is the 
screen reader speech dictionary files John has produced. It should be 
possible to make other speech dictionaries for other screen readers and 
other languages. Also it should be possible to write Braille tables to 
enable translation into Braille. I would imagine with enough work one 
could make a liblouis table to produce Nemeth from LEAN but I probably 
would go with what John said about 8-dot Braille potentially being a 
better option (it probably does need development of a Braille code).

Also fonts can be used to make it look like some fairly standard linear 
form of writing maths to a sighted person. How intuitive it will look is 
something I cannot fully comment on.

If I were to make any comparison to existing systems, it would be 
triangle or LAMBDA, but LEAN is unicode based, using mostly standard 
unicode symbols where possible. As far as I know MathSpeak is not really 
dealing with the same problem as triangle or LAMBDA.

Michael Whapples
On 14/04/2012 21:29, Susan Jolly wrote:
> Hi John G.,
>
> It would be helpful for my (and perhaps others') understanding of LEAN 
> if you could explain the differences between your new LEAN notation 
> and MathSpeak.  MathSpeak, as I'm sure you know, is a spoken form of 
> Nemeth braille that is similar to the "verbal math" option in 
> MegaDots.  (John Boyer knows more about the latter than I do so it 
> would be great if he wants to add some additional information.)
>
> MathSpeak was orginally proposed by Dr. Nemeth but over the last few 
> years ghBraille has undertaken a large research effort to test and 
> update MathSpeak as necessary to ensure that the latest version of 
> MathSpeak is comprehensive, easy to learn and understand, and avoids 
> ambiguity.
>
> Sincerely,
> Susan Jolly
>
> _______________________________________________
> Blindmath mailing list
> Blindmath at nfbnet.org
> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
> To unsubscribe, change your list options or get your account info for 
> Blindmath:
> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/mwhapples%40aim.com 
>