100%

2 views
Skip to first unread message

kurosh

unread,
May 16, 2008, 2:14:22 PM5/16/08
to Android Challenge
A handwring recognition system with 100% recognition accuracy.
A data entry method for the visually impaired/blind, that alternately
allows data entry for sighted users when
a) The user is moving and the keys aren't visible, due to excessive
jerkiness or bad lighting conditions, when normal data entry methods
would not be suitable.
b) When a more intuitive method other than existing keyboard methods
is suitable

Effective use: In the lower 25% of all submitted applications
Polish: In the lower 25% of all submitted applications
Indispensability: In the lower 25% of all submitted applications
Originality: In the lower 25% of all submitted applications

Overall: In the lower 25% of all submitted applications


KDF Infotech
www.kdfinfotech.com

kurosh

unread,
May 16, 2008, 2:58:23 PM5/16/08
to Android Challenge
For the importance of providing accessability to the blind/visually
impaired on PCs and mobile devices refer to the following links:
http://www.robobraille.com
http://www.slideshare.net/crossana/mcsig-chi-2008
http://www.knfbreader.com/

Suffice it to say that our entry is a viable, intuitive data entry
method for the visually impaired on mobile phones, with a sufficiently
shallow learning curve for young and old alike.

kurosh

kurosh

unread,
May 16, 2008, 3:03:17 PM5/16/08
to Android Challenge
My error. The first link should be
www.robobraille.org

kurosh

On May 16, 11:58 pm, kurosh <kuro...@gmail.com> wrote:
> For the importance of providing accessability to the blind/visually
> impaired on PCs and mobile devices refer to the following links:http://www.robobraille.comhttp://www.slideshare.net/crossana/mcsig-chi-2008http://www.knfbreader.com/
>
> Suffice it to say that our entry is a viable, intuitive data entry
> method for the visually impaired on mobile phones, with a sufficiently
> shallow learning curve for young and old alike.
>
> kurosh
>
> On May 16, 11:14 pm, kurosh <kuro...@gmail.com> wrote:
>
>
>
> > A handwring recognition system with 100% recognition accuracy.
> > A data entry method for the visually impaired/blind, that alternately
> > allows data entry for sighted users when
> > a) The user is moving and the keys aren't visible, due to excessive
> > jerkiness or bad lighting conditions, when normal data entry methods
> > would not be suitable.
> > b) When a more intuitive method other than existing keyboard methods
> > is suitable
>
> > Effective use: In the lower 25% of all submitted applications
> > Polish: In the lower 25% of all submitted applications
> > Indispensability: In the lower 25% of all submitted applications
> > Originality: In the lower 25% of all submitted applications
>
> > Overall: In the lower 25% of all submitted applications
>
> > KDF Infotechwww.kdfinfotech.com- Hide quoted text -
>
> - Show quoted text -

kurosh

unread,
May 16, 2008, 3:20:35 PM5/16/08
to Android Challenge
My question is:
Is accessability to mobile devices for the blind less important/
INDISPENSIBLE than some of the winning applications?
Is there ANY Handwriting Recognition system in existence that can
deliver 100% recognition accuracy. If the answer is yes, then my
system has no ORIGINALITY.

kurosh

On May 17, 12:03 am, kurosh <kuro...@gmail.com> wrote:
> My error. The first link should bewww.robobraille.org
>
> kurosh
>
> On May 16, 11:58 pm, kurosh <kuro...@gmail.com> wrote:
>
>
>
> > For the importance of providing accessability to the blind/visually
> > impaired on PCs and mobile devices refer to the following links:http://www.robobraille.comhttp://www.slideshare.net/crossana/mcsig-ch...
>
> > Suffice it to say that our entry is a viable, intuitive data entry
> > method for the visually impaired on mobile phones, with a sufficiently
> > shallow learning curve for young and old alike.
>
> > kurosh
>
> > On May 16, 11:14 pm, kurosh <kuro...@gmail.com> wrote:
>
> > > A handwring recognition system with 100% recognition accuracy.
> > > A data entry method for the visually impaired/blind, that alternately
> > > allows data entry for sighted users when
> > > a) The user is moving and the keys aren't visible, due to excessive
> > > jerkiness or bad lighting conditions, when normal data entry methods
> > > would not be suitable.
> > > b) When a more intuitive method other than existing keyboard methods
> > > is suitable
>
> > > Effective use: In the lower 25% of all submitted applications
> > > Polish: In the lower 25% of all submitted applications
> > > Indispensability: In the lower 25% of all submitted applications
> > > Originality: In the lower 25% of all submitted applications
>
> > > Overall: In the lower 25% of all submitted applications
>
> > > KDF Infotechwww.kdfinfotech.com-Hide quoted text -
>
> > - Show quoted text -- Hide quoted text -

kurosh

unread,
May 16, 2008, 3:33:27 PM5/16/08
to Android Challenge
How much POLISH is a text editor supposed to have? Is'nt audible
feedback while editing text, gesture recognition, and adjustable
character delay sufficient polish in an editing app for the blind?
EFFECTIVE USE: Providing an alternate method to multitap and
predictive text actually would EXTEND the platform by providing a
system wide keyboard class for visually impaired, which is non-
existent on other platforms.
> > > > KDF Infotechwww.kdfinfotech.com-Hidequoted text -

kurosh

unread,
May 16, 2008, 4:11:45 PM5/16/08
to Android Challenge
The last time I was so shocked by a result was when I got an
800/800 (99 percentile) score in my GRE Math.
> > > > > KDF Infotechwww.kdfinfotech.com-Hidequotedtext -

kurosh

unread,
May 16, 2008, 4:28:40 PM5/16/08
to Android Challenge
Having said all that, and since I am a perfectionist by nature, my
only regret is that I made a spelling error.
It should read INDISPENSABLE and not INDISPENSIBLE.

kurosh

unread,
May 16, 2008, 11:10:04 PM5/16/08
to Android Challenge
On hindsight, and in all fairness to the google judging process, one
of the possible explanations for such a low score on all 4 criteria
could be that the apk did not run at all.
Could someone from Google please clarify. If that is the case then we
would like to resubmit our app to the ADC II challenge, if permitted.

Regards

kurosh

hanni

unread,
May 16, 2008, 11:37:44 PM5/16/08
to Android Challenge
Wow that sounds like a really cool app!

I'm sure that you figured out the correct explanation yourself, and
that there simply was an error with the upload or similar. You should
definately resubmit that app to the 2:nd challenge!
> > - Show quoted text -- Dölj citerad text -
>
> - Visa citerad text -

kurosh

unread,
May 16, 2008, 11:47:45 PM5/16/08
to Android Challenge
Thanks for the feedback. It is important to me as a developer that I
am acknowledged by my peers. It keeps me motivated.
It is a statistical fact the we as humans receive 90% of our sensory
feedback of the world around us through our eyes. Can you imagine the
quality of life of those millions who have a quality of life that 90%
worse than sighted people.
Accessability is a fundamental right. I have tried to do my bit. I
hope it is appreciated by someone, someday.

kurosh
> > - Visa citerad text -- Hide quoted text -

Shane Isbell

unread,
May 16, 2008, 11:58:12 PM5/16/08
to android-...@googlegroups.com
On Fri, May 16, 2008 at 8:47 PM, kurosh <kur...@gmail.com> wrote:

Thanks for the feedback. It is important to me as a developer that I
am acknowledged by my peers. It keeps me motivated.
It is a statistical fact the we as humans receive 90% of our sensory
feedback of the world around us through our eyes. Can you imagine the
quality of life of those millions who have a quality of life that 90%
worse than sighted people.
Accessability is a fundamental right. I have tried to do my bit. I
hope it is appreciated by someone, someday.
This humanitarian aspect is how Android was pitched but it just didn't stand up to the way existing industry sees things. None of the humanitarian apps made it to the top 50, so it's very possible it has nothing to do with your application. I think we have to fight this one out to change the perspective. Do you have an apk ready for people to look at.
 
Shane

kurosh

unread,
May 17, 2008, 12:08:55 AM5/17/08
to Android Challenge
Shane,

At presenet I am quite unsure how I will proceed. Once I weigh all my
options I would definitely like to be an active, contributing memer of
the Android community, and intend to showcase the app at some point in
time.

kurosh


On May 17, 8:58 am, "Shane Isbell" <shane.isb...@gmail.com> wrote:
> > > - Show quoted text -- Hide quoted text -

kurosh

unread,
May 17, 2008, 1:11:25 AM5/17/08
to Android Challenge
I like to refer to text entry as a meta application. i.e. an
application that is used by most if not all other applications. That
is how integral it is to mobile computing.

The most fundamental shortcoming of a mobile device is the data entry
interface, as compared to the PC keyboard. We have tried to address
this problem.
To quote the little girl in the original Google video.
"My magic phone would have a keyboard that fits in my pocket"

The system design is rugged, reliable and accurate and eliminates the
biggest barrier to the acceptance of handwriting recognition as a
viable text-entry method. A recognition accuracy of 100%

Although the target demographic was the blind/visually impaired, the
system is also a viable text entry method for the following use-cases:

1) It is intuitive and provides 100% recognition accuracy of all
characters stroked

2) Since it is intuitive to stroke characters, there is zero seek time
i.e. the time taken to seek a character on a keyboard before pressing
the key. This can convert to high speeds of data entry

3) Does not need a stylus, and can be easily stroked by hand

4) Complements multitap and predictive text by providing gesture
recognition and relatively easy entry of special characters

5) For data entry in a dark area or while on the move when its
difficult to see the keys on the keyboard, or the phone is jerking
while travelling, when most handwriting recognition systems would give
a large error rate, and keyboard entry methods are difficult if not
impossible.
(And we all know that mobile data entry is generally done on the move)

6) Since special character entry is very fast compared to multitap or
predictive text and recognised with 100% accuracy as compared to other
handwriting recognition systems, it can serve as a programmers editor
where special characters form a large part of the vocabulary.

7) Can be implemented on hand wearable mobile devices where
conventional keypads are impractical to implement and use

8) In an emergency - send out an SOS message intuitively and reliably
when normal sighted access to the keyboard is not possible

kurosh

kurosh

unread,
May 17, 2008, 2:39:45 AM5/17/08
to Android Challenge
From a purely marketing viewpoint:

XYZ handset data entry features:
1) multitap
2) predictive text
3) handwriting recognition with less than 100% accuracy

Android handset data entry features:
1) shapewriter (more efficient onscreen soft keyboard)
2) multitap
3) predictive text
4) handwritng recogntion with 100% recognition accuracy
5) data entry for the visually impaired/blind PLUS all the features
listed in my previous post

Which handset would you rather buy?
To put it another way. Would shapewriter along with my app help sell
more Android hansets or not, considering that the greatest shortcoming
on mobile handsets is the data entry interface.

kurosh

kurosh

unread,
May 17, 2008, 4:04:21 AM5/17/08
to Android Challenge
hanni,

I didn't know you were one of the top 50. Congratulations.
And its nice to know that someone from the top 50 seems to think my
app worthy of credit.

kurosh

kurosh

unread,
May 17, 2008, 8:28:46 AM5/17/08
to Android Challenge
I think I'd like to revise the previous post I made on handset
marketability wrt the feature set for XYZ handsets.

From a purely marketing viewpoint:

XYZ handset data entry features:
1) onscreen soft keyboard
2) multitap
3) predictive text
4) handwriting recognition with less than 100% accuracy


Android handset data entry features:
1) shapewriter (more efficient onscreen soft keyboard)
2) multitap
3) predictive text
4) handwritng recogntion with 100% recognition accuracy
5) data entry for the visually impaired/blind PLUS all the features
listed in my previous post


Which handset would you rather buy?
To put it another way. Would shapewriter along with my app help sell
more Android hansets or not, considering that the greatest
shortcoming
on mobile handsets is the data entry interface.




> ...
>
> read more »- Hide quoted text -
Reply all
Reply to author
Forward
0 new messages