I think my argument only diverts from your view and Nicks in that I never
said a blind person can't do it. I have done it on my own as far back as
1992 in school. My point is that even with Homer I am slower than a sighted
person who can layout an entire applications GUI's in less than 30 minutes.
So sure if you're not in a job where your competing against another sighted
coder who can get the job done in exponintionally faster time than you your
fine. I still feel if you are writing large applications with a time
component to release you really need a graphics arts person to do it. And
they are a dime a dozen.
This doesn't mean you can't create your own code and release it on either
IPhone or Android markets, that or sell your own software to the public
without a sighted person involved. It just means that in certain jobs you
can forget doing GUI's because you are just too slow. In those same jobs
though I have found that doing the back end code I can focus much more on
algorithms under the hood when I have a sighted person building the GUI
Ken
-----Original Message-----
From: programmingblind-bounce@xxxxxxxxxxxxx
[mailto:programmingblind-bounce@xxxxxxxxxxxxx] On Behalf Of Jamal Mazrui
Sent: Saturday, October 16, 2010 7:20 AM
To: programmingblind@xxxxxxxxxxxxx
Subject: Re: Is GUI Programming Worth While for Visually Impaired Coders?
My view on this is probably closest to Nick's.
I think we can write functional, visually acceptable GUIs. In general,
we should seek sighted assistance to ensure that the layout is visually
acceptable, since it is easy for something to look odd that we did not
realize. If we understand the principles of visual layout, however,
including general spacing and alignment guidelines, then a form can
usually be tweaked without much effort to get the rest of the way to
visually acceptable.
I agree that fancy GUIs are probably not a good use of our time. Yet,
these, are over rated. As opposed to web pages that are often
presenting a corporate image, marketing products, etc., software
applications tend to be used more directly to achieve functional
outcomes. There are exceptions, but most people use an application to
get things done, not to browse competing offerings that try to grab
one's attention and suck one in for spending money. Applications that
overdo visual effects can be overwhelming, distracting, or at least less
functional than those that concentrate on widgets for achieving a
particula task in a straightforward and convenient manner.
Guido van Rossum, the architect of the Python language, has referred to
GUIs built with Windows Presentation Foundation (WPF) as "eye candy."
That is from an immensely talented sighted developer who has probably
seen it all. In the organization where I work, I have found sighted
users to be generally satisfied with the GUIs of custom apps that I
write internally, which include GUIs. I suppose they might be polite if
they know the programmer behind the GUI is blind, but I do use a bit of
reader time before sharing a GUI to tweak it if necessary, and they come
back for more programs. Much more important to them seems to be what
the program lets them accomplish than how fancy it looks.
I agree that embedded software or speech-based interfaces are often
alternative areas to pursue rather then GUI-based apps. Another
interface is command-line, console mode interfaces, which have had some
resurgence in recent years, as people have realized that a GUI is
sometimes less efficient for tasks than a command-line utility. Of
course, this has been known in the Unix world for a long time. There is
no reason why a blind programmer cannot develop such an interface as
well as his or her sighted counterpart.
I think graphical form designers are inherently difficult for us to use
well, even with scripts. We tend to be better off using layout wizards
and containers with intelligent logic for positioning and sizing
controls, based on algorithms developed from usability research. With
dialog or text-oriented layout tools, however, I think we can build a
GUI as efficiently as sighted peers. Moreover,such a GUI will be easier
to modify and evolve because it does not hard code pixel positions
nearly as much. They will also tend to work better in cross-platform
projects because the layout containers automatically adapt to
characteristics and conventions of the target platform.
If anyone knows of an easier way to build a visually acceptable and
highly functional form or Multiple Document Interface (MDI) as the
Layout by Code techniques I developed in the Homer application
Framework, HomerAPP,
http://EmpowermentZone.com/appsetup.exe
I would like to know about it. One can create each control with a
single method call on one line, without specifying specific coordinates.
I'm sure it's not perfect, but it is the result of years of study,
programming, and debugging I did in this area. If I can get that far
with such tools as a single developer, I am sure they or other tools
could be made even better if more people got involved in creating them.
Given the complex and amazing things that software can do these days,
implementing auto-layout algorithms for GUIs is hardly considered to be
a sophisticated challenge in computer science.
Although WPF makes eye candy easy to create in apps, it is also the case
that mobile devices are emphasizing more functional GUIs because of
their constrained displays. There is an exploding market, and blind
people should be able to write acceptable GUIs on such platforms.
Mobile apps tend to be task oriented rather than visually fancy.
Jamal
__________
View the list's information and change your settings at
//www.freelists.org/list/programmingblind
__________
View the list's information and change your settings at
//www.freelists.org/list/programmingblind
