Re: 2.6 proposed module: dasher

On Mon, Jan 12, 2004 at 09:22:35PM -0800, Peter Korn wrote:

> One of the hallmarks of your project has been the impressive user studies 
> you have done on typing performance.  What would be ideal, in my opinion, 
> is if you were to put together a Dasher-based interface to dialog boxes, 
> toolbars, HTML content, etc., and get feedback from users as to how this 
> worked for them.

Code for this is now in CVS. In a similar way to GOK, Dasher simply 
provides useful widget information for the application or window which 
currently has focus. Arguably this is most useful with the -o switch to 
Dasher, which causes Dasher to default to text entry mode and gets rid 
of the edit box. At that point Dasher is a dedicated Gnome accessibility 
application. Possibly this should be the default in Gnome builds?

On the other hand, I'm not sure that Dasher is ever going to adapt well 
to something like, say, Gnumeric.

Is Mozilla's widget decently accessible yet? GtkHTML ought to work with 
the current code, though I haven't checked.

(Actually, there's one issue with the current code - it only properly 
works if Dasher refuses focus. I can't see it being terribly useful 
except in the case of Dasher being used for entering text into other 
applications at which point it refuses focus anyway, so I don't see that 
as a major problem)

> Dasher as a text entry replacement is tremendous.  But until Dasher is able 
> to provide a complete alternate navigation system of a desktop for a user, 
> how can the Dasher user be independent?  Should Dasher be a companaion 
> application to GOK?  Is there interest in making Dasher a desktop-wide 
> assistive technology?

Dasher in its current form allows users to start and manipulate the 
majority of applications to some degree of functionality. At present, 
however, it does also require the ability to focus the desired window or 
text entry area. This is (a-ha ha ha) just a small matter of coding. 
Apart from that, I don't think that as of now Dasher users would be 
significantly handicapped compared to, say, GOK users.

I think it's a worthwhile inclusion in 2.6, since it demonstrably does 
increase the range of users we can support. By the 2.8 timeframe that 
range could be increased further, but I think this is equally true 
regardless of whether it's included now or not.
Matthew Garrett | mjg59 srcf ucam org

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]