Hi Neil,
Yes, I had trouble with complex fractions and multilevel scripts and decided to
put the Nemeth math out and continue working on them and other things. one
problem
was that I couldn't find a good math textbook online, so i couldn't see how
mathML was actually used for these forms. I was looking for a calculus course.
That
was many years ago.
John
On Wed, Aug 14, 2019 at 12:12:54AM -0700, Neil Soiffer wrote:
I might not have been clear in my explanation: fractions, radicals, and
scripts are all handled. However when a script is nested, Nemeth code
indicates the nesting level. For example, 'x' with a superscript that is 'y
superscript z' has the Nemeth code is x^y^^z (note the ^^ before the z),
but the current incarnation of the libraries generate x^y^z -- a single ^
before the z. Similarly, fractions that are nested need to announce their
complexity/level of nesting at the start, end, and for the fraction line.
That's not done -- all fractions are treated the same in the current code.
You can see some details of that for fractions in the issue that I filed on
it: https://github.com/liblouis/liblouisutdml/issues/62.
Did the original implementation handle these more complicated nested cases
and that got broken at some point? Or is that something that was a "version
2" feature that was never implemented?
Neil
On Tue, Aug 13, 2019 at 11:44 PM John J. Boyer <john.boyer@xxxxxxxxxxxxxxxxx>
wrote:
Hi Neil,
As i originally wrote nemethn.sem and nemeth.ctb they handled fractions,
radicals, sub and superscripts, and all the other usual forms, except for
large matrices.
John
On Tue, Aug 13, 2019 at 05:12:13PM -0700, Neil Soiffer wrote:
I've submitted a PR for a small batch of Nemeth tests and created issuesto
for the failures. In general, the converter doesn't seem to handle any of
the nested notations that Nemeth code uses (fractions, radicals, or
sub/superscripts). There are also other problems, but I think those are
more easily solved. Does anyone know if this is a problem that is going
require a significant change to the code or if it can be handled viatable
changes? Does BrailleBlaster do these correctly?john.boyer@xxxxxxxxxxxxxxxxx>
Neil
On Tue, Aug 13, 2019 at 1:22 PM John J. Boyer <
wrote:have
Hi Bert,
I hope the others can contribute something on math in liblouis. You
towards adug some clues out of the archives. Nemeth.cti might be a start
missed orstring that
liblouis can translate into any Braille math code.
Thanks,
John
On Tue, Aug 13, 2019 at 01:03:57PM +0200, Bert Frees wrote:
I looked at the archives and the common linear math format was indeed
apparently something that was talked about before. I must have
(developedforgotten it. I don't know how much of this idea has been realized.
I saw that there was a lot of talk about the "LEAN" notation
whetherby
John Gardner and Courtney Christensen), but I'm not sure whether this
notation is suitable as input for a Liblouis table, and if yes,
whichit
also ended up being used as the base for the UEB table (and if yes,
backgroundsubset of it was implemented). I would love to get some more
testfrom the BrailleBlaster guys because we would like to document and
Michaelthe
behavior of the UEB table in Liblouis for math input. Keith? Michael
Whapples? Michael Gray?
After writing the above paragraph I saw in the Git history that
with aWhapples and Michael Gray did some additions to the Nemeth table,
as asingle mentioning of LEAN. So that does indicate that LEAN was used
couldn't becommon base for both the UEB and the Nemeth conversions inBrailleBlaster.
I also saw this confirmed in an email from Keith in 2017. Why didn'tanyone
tell me this when we were discussing the nemeth.ctb table? It
themore relevant!
Because there were no changes made to Liblouisutdml, this means that
thinknotations used by Liblouisutdml and BrailleBlaster are
different, and that the Nemeth table in Liblouis supports both. I
file forit
would be a good idea to also change Liblouisutdml to use
the LEAN notation. Then we could in principle have a single .sem
testboth Nemeth and UEB.
So I would like to repeat my appeal that if BrailleBlaster has some
wouldmaterial that could be useful for either Liblouis or Liblouisutdml,
useful!you please consider sharing it with us? That would be extremely
to
Looking at the archives, I discovered that some test data for MathML
code.Nemeth was posted to the mailing list by John Gardner,
in 2013. I wonder why these test files never ended up in the source
time.They definitely should. Maybe I had a very busy period around that
them.)Dunno.(Note
To Neil: you're gonna find these files useful I think, or perhaps you
already got them via Susan Jolly? The files can be downloaded
from https://www.freelists.org/post/liblouis-liblouisxml/Nemeth,14.
that you need to rename them to .docx in order to be able to open
goals
Bert
Op di 13 aug. 2019 om 01:07 schreef John J. Boyer <
john.boyer@xxxxxxxxxxxxxxxxx>:
Hi Christian,
I am concerned about math suport in liblouis. That was one of my
beingwhen I started development. I was motivated by my experience of
liblouisunable to obtain good
STEM material in Braille. This started when I was a child.
What is the current state of math suport? Some years ago there was
discussion of translating MathML into a character string which
butsince Icould translate into
any Braille math code. Has anything been done on this?
I am pleased to see that Quality Control has been greatly improved
was working on liblouis. I can't give much help in developing now,
I
want to
encourage anyone who can.
Thanks,
John
--
John J. Boyer
Email: john.boyer@xxxxxxxxxxxxxxxxx
website: http://www.abilitiessoft.org
Status: Company dissolved but website and email addresses live.
Location: Madison, Wisconsin, USA
Mission: developing assistive technology software and providing STEM
services
that are available at no cost
For a description of the software, to download it and links to
project pages go to http://liblouis.org
--
John J. Boyer
Email: john.boyer@xxxxxxxxxxxxxxxxx
website: http://www.abilitiessoft.org
Status: Company dissolved but website and email addresses live.
Location: Madison, Wisconsin, USA
Mission: developing assistive technology software and providing STEM
services
that are available at no cost
For a description of the software, to download it and links to
project pages go to http://liblouis.org