[BRLTTY] de.ttb

[Mario Lang]

Dave Mielke <dave at mielke.cc> writes:

>>> On first sight, this looks wrong. Given that we have 2^8 possible
>>> chars and 2^8 possible dot-patterns, we shouldn't reuse the
>>> same dot pattern for two different characters.
>
> We actually now have way more than that since the tables have become 
> unicode-based.

I know.  I was refering to the first 256 characters only.
At least in latin charsets, I think there should be no duplication of dot
patterns in the first 256 characters.  Not only because it is problematic
for those that input via braille keyboard, but also because
the ambiguity makes it harder for users to tell what
character a certain dot pattern actually represents.

> In practice, though, while you can define every single unicode
> character, the VGA font in use at any given time can only have up to
> 512 characters in it. We're unfortunately limited by that since
> Linux doesn't export its unicode image of the screen. The vcs
> devices give us font offsets which we then back translate to unicode
> characters.

I am aware of that.  However, I think it might be
useful to define a lot more unicode characters since the
screen font can be switched pretty easily.  Besides, while
there is this 512 limitation on linux console, BrlAPI clients
like Orca can feed unicode characters to BRLTTY pretty easily
without any limitation on 512 characters.  If you read a webpage
in Firefox with Orca, you might be exposed to a lot of unicode stuff.
One example that kind of interests me is the international phonetic
alphabet, which is used on Wikipedia sometimes.  A mapping
for that would be useful...  There are other things like certain
symbols are kind of duplicated in Unicode.  But I cant think
of any right now, I'd need to check.

-- 
CYa,
  ⡍⠁⠗⠊⠕ | Debian Developer <URL:http://debian.org/>
  .''`. | Get my public key via finger mlang/key at db.debian.org
 : :' : | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44
 `. `'
   `-      <URL:http://delysid.org/>  <URL:http://www.staff.tugraz.at/mlang/>