[nvda] Re: What would it take for a graphics dictionary

well the way that Jaws does it from what I understand is it recognizes a 
graphic through either MSAA or through the video intercept then indexes it 
using a graphics number.  Now, if jaws is able to detect a tool tip associated 
with the graphic then it uses the tool tip or just the text associated with the 
graphic to give it it's name.  Otherwise it just gives it the name of it's 
index number.  Not impossible to do but keep in mind that Jaws and Window Eyes 
also rely on a video intercept to get some of their text information.
  ----- Original Message ----- 
  From: Michael Whapples 
  To: nvda@xxxxxxxxxxxxx 
  Sent: Tuesday, March 27, 2007 3:36 PM
  Subject: [nvda] What would it take for a graphics dictionary


  Hello,
  There was a question about nvda and outlook express and the status of 
messages, and the answer someone suggested is that it is done with graphics. 
This leads me to the question, if it is graphics, are the other screen readers 
doing it through graphic dictionaries, and what would this take to be 
implemented in nvda? How do they identify different graphics? Surely not just 
with a list of graphics and an associated text description, or is it really 
that basic (and if so how do you stop it swamping your HD?)

  So how valuable would a graphics dictionary be, and how many uses would it 
have? Considering that is it something that needs further consideration at the 
moment, or can it wait? I know there are certainly other areas which will 
improve the usability of NVDA, I would just like to get an idea of what will 
yield the best results for the time spent on it, so an idea of the priorities 
and where a graphics dictionary is in the list.

  From
  Michael Whapples

Other related posts: