[BRLTTY] Support for 'dead' or broken cells?

[Jason White]

On Wed, Mar 28, 2007 at 10:25:07AM +0200, yan at mistigri.org wrote:
> Yes this could be an interresting feature:
> In the core we could make a difference bitween the 'real length ' of the 
> display and the 'actually used length'. 
> Moreover we would have to implement something like a table refering to 
> cells that are 'down' or 'not working'..

Can this be done without affecting performance in the 99.9% of cases in which
this is not used? After all, when braille cells fail, most people have them
replaced, though of course they might have a need for this kind of feature in
the interim, especially if it is going to take weeks to have the new parts
brought in from the other side of the world.