Re: Virtual View User's Guide

Fri, 06 Nov 2015 13:36:56 -0800 

And just as a curosity:
Did you know, these people that does the actual solving for you, are named "Computers". That was the original meaning of the word computer. Since we now aday let a machine do a number of tasks for us, we have named that machine COMPUTER. But that is merely because it overtook the task from the human computers. 


By the way, such "living computers" you will find doing a lot of the 
things we think happen automatically on the net. We think places like 
Google has a high-tech piece of software, which "knoes" what your 
preferences are, and automatically picks the advertisements to throw on 
you; just for the bother of it all. In reality though, likely the 
software has gathered some info about your browser history, and whenever 
you are done, sends the info to some human computer. These people may 
then one way or the other - more or less labor involved - catagorize you 
in with this and that kind of advertisements, which will show up on your 
screen, next time you visit the actual page. OK, perhaps Google is not 
the right example here, but I just used it for illustrative purposes. 
Fact is, that human computers are behind a number of facilities. And, 
there even are people making a living on being human computers.



Did you think it all happened free of charge? Well, let me hasten to say 
that some of the services are based on volunteers. But in most cases, 
the human computers are paid maybe 1 or 2 cents for each job done. A 
service like WebVisum, may have hundreds or even thousands of human 
computers engaged. They may take jobs from both the WebVisum group, as 
well as certain trivial jobs that are "happening automatically" from 
your local store, library, public services, news and streaming services, 
search engines ETC. Whenever you want a captcha solved, you send it to 
the network. The one person who is first at accepting the job, solves 
it, sends it back to your computer, gets registered and credited for the 
job; and can go on with their normal job or jump to next Computer-request.



In certain cases, if you try for instance to send a captcha to webvisum, 
it takes a couple of seconds for the job to be carried out. Other times, 
it might take 10 or 15 seconds, before the result gets back to you. 
Still other cases, you will experience that the captcha does not get 
solved, and you are informed the captcha  has no solving. In those 
cases, it is because the software has a certain preset timeout. If noone 
has accepted the job, the software just lets you know it was not carried 
out. the way the whole service is constructed, in combination with the 
messages it reveals to you, all give you the feeling this was an 
automated, completely electronic service.



What's more, this likely is one of the biggest reasons why such 
captcha-solving is not always looked upon as "good practice". Depending 
on the software, the underlaying website you are visiting, the security 
settings on the network of human computers, it may reveal more or less 
of the information on the screen at the captcha time. Imagine your bank 
using a captcha for your payment to be processed. If the Captcha 
software simply distributes the captcha itself, and NO OTHER INFO from 
the site - not even the site address itself - it may be considered fully 
safe. all the human computer sees is a small "picture", holding some 
characters. On the other hand, what if the whole screenshot is being 
distributed to the human network, - maybe consisting of hundreds of people?



Does all of this mean, that when you enter a search query on places like 
Google or Yahoo, it will be processed by human computers? No. That kind 
of search engines have long time been developed in fully electronic 
format. It is taken care of, by some algorithms, and some piece of 
software which is called WebCrawlers, or WebSpiders. These pieces of 
software will scan the net for new websites, updates on existing 
websites, and store some of the info from each website in a database. 
Your query will then be tested against these databases, and the 
algorithms will determine which hits you will get. Yet, human computers 
may be used to catch the typical queries, and set the database to give 
results that normally would not fall in under what you were querying. 
Such could for instance be, when a word or phrase typically will be 
misspelled by many users. Or, in cases where you most likely will need 
additional info. The human computers do not see your query in itself. 
When a thousand users have searched for something similar, without 
getting any hits, the query will be send to a human computer. This 
person then determines what you and the others may be looking for, and 
enters info about what kind of alternatives should be displayed upon 
such a request into the database. Next time you try the same search 
term, all the sudden, the search engine will return you 10000 results, 
all based on some catagories that were stored for that search query.



Did you think the human computers are just on their own - one person 
here, one there? Sure, many of them may be homeworkers, even all ages. 
And they are spread all over the globe. As a matter of facts, to have 
them spread widely in geographical terms, will prove beneficial. Not 
only does this ensure the service being operative to the highest extent, 
independetnt of whether one local line is out of service. But you as a 
user of the service, want your job done NOW, at ANY time of the day, ALL 
year round. When it is night locally with you, and all local human 
computers likely are away from their desk, the job is taken care of by 
someone on the other side of the globe. You just get the job done in a 
matter of seconds, and will be joyfully going on with your tasks.



OK, don't be fooled. A number of services that are provided over the 
net, REALLY are automatic services,all carried out by completely 
electronic pieces of hardware and software. they are more or less NEVER 
touched by any human, and will be done without any human even knowing 
thereof. But some of the services we use every day, will still be - and 
may even need to be - carried out by human computers. One example may be 
certain bank transactions, another could be captcha solving, thirdly we 
could mention services where security reasons or physical and technical 
reasons make it necessary with some human interaction. many of the 
services may be partially automated, only calling for human computers 
when  there is a given need, or if some unusual activity is ongoing.



OK, long message, but just wanted to let you know that there really is 
alot going on behind the scene of modern internetting.



On 11/6/2015 7:10 PM, Tom Kingston via Talk wrote:

You're right, Kevin. There's no magic in Webvisum. It's a human being 
looking at it, typing it in, and sending it out to you.


Tom


On 11/6/2015 1:04 PM, Kevin Huber via Talk wrote:

Hi Rod:
It has been a while since I had to solve captchas, but I used Webvisom
quite successfully.   I think that when you press the hotkey that
sends the captcha, there are one or more humans on the other end, one
of them solves the captcha and the software automatically sends it to
your clipboard so you can paste it into the proper area.
Kevin Huber

On 11/6/15, Rod Hutton via Talk <[email protected]> wrote:

And yet it is my understanding that webvisum, the Firefox add-on, does
precisely this in order to solve captchas.

Rod

-----Original Message-----
From: Talk
[mailto:[email protected]] On
Behalf
Of Aaron Smith via Talk
Sent: November 6, 2015 12:45 PM
To: [email protected]; Window-Eyes Discussion List
<[email protected]>
Subject: RE: Virtual View User's Guide

Captchas are design to elude OCR, so, probably not.

Thanks,

Aaron

--
Aaron Smith
Web Development * App Development * Product Support Specialist
Ai Squared * 725 Airport North Office Park, Fort Wayne, IN 46825
260-489-3671 * www.aisquared.com

To insure that you receive proper support, please include all past

correspondence (where applicable), and any relevant information 
pertinent

to

your situation when submitting a problem report to the Ai Squared 
Technical

Support Team.



-----Original Message-----
From: Joe Paton [mailto:[email protected]]
Sent: Friday, November 6, 2015 12:42 PM
To: Aaron Smith <[email protected]>; Window-Eyes Discussion List
<[email protected]>
Subject: Re: Virtual View User's Guide

Thanks Aaron.

One question if I may, will virtual view work with graphic capchas?

Great job, and thanks for taking the trouble.

Joe Paton


On Fri, 6 Nov 2015 16:14:16 +0000
Aaron Smith via Talk <[email protected]> wrote:

Greetings,

The following user's guide will ship with the next version of Virtual

View, but I

wanted to provide it for you here as many of you have been asking for
more
information about what Virtual View is, and how to use it.

Thanks,

Aaron



Virtual View

1     Overview

Virtual View is an app designed to provide a textual representation 
of a

program window, dialog box, message box or file using various text

retrieval

methods. The resulting text can be reviewed in the Virtual View window

without disturbing the contents of the screen or file. In some 
cases, the
buttons, links and other controls found in the original window that 
was
virtualized can be acted on directly from the Virtual View window 
using

simple keyboard commands.
This app will come in handy in a variety of situations including:

*         Reading image based PDF files

*         Reading text presented as an image


*         Selecting and copying static text found in dialogs and 
program

windows


*         Interacting with programs that lack proper keyboard

accessibility


*         Needing quick access to the text found in a PowerPoint

presentation


2     Getting Started

The default hotkey, Control-Alt-V, will open the source selection 
dialog.

You

may also choose to access Virtual View from the Apps menu in the 
Window-

Eyes Control Panel. Virtual View may also be activated by pressing

Alt-Print

Screen. Alt-Print Screen is an operating system hotkey that takes a

snapshot

of the active window. You have the option of letting Virtual View

virtualize

this image when the hotkey is pressed.
3     Selecting a Source

Pressing the Virtual View hotkey, or choosing Virtual View from the 
apps
menu, will cause the "Select Window or File to Virtualize" dialog 
to be
displayed. You can use this dialog to tell Virtual View what you 
want to

virtualize. You may choose from one of three sources:


1.       Active Window (A) - Selecting this radio button will cause 
the

current

application window to be used as the source.


2.       Focused Window (F) - Selecting this radio button will 
cause the

current

control in the active application to be used as the source.


3.       File (I) - Selecting this radio button will cause Virtual 
View

to

prompt for

a file using a standard Windows File Open dialog.

3.1   Supported File Types

Virtual View has the ability to virtualize a large number of file 
types.

Exactly

which files can be virtualized depends on the number of available file

"filters"

installed in your operating system. Filters are programmatic 
interfaces

for

accessing contents of files. Several filters come pre-installed 
with your

operating system. The most common file types include:

*         Text

*         XML

*         HTML

*         Zip

Installing Microsoft Office provides you with additional filters 
allowing

Virtual

View to virtualize many Office document formats, including:

*         Docx

*         Xlsx

*         Pptx

4     Choosing a Method

After you've selected a source, whether it be a window or a file, 
you can
then choose how you want Virtual View to obtain the text from the 
source.

The methods available will differ depending on which source type is

selected.



1.       Clips (C) - The clips method will use the OSM (Off-Screen 
Model)

to

retrieve text from a window.


2.       UIA (A) - This UIA method will use the UI Automation 
interface

to
retrieve text from a window.


3.       OCR (O) - This method uses OCR (optical character 
recognition)

to
retrieve text from either a window or a file.

4.       Version (V) - This method provides version information for a

window or

file.
Which method you choose depends on a number of factors, and there are
advantages of each.

Clips is the most common method for accessing the text of a window. 
They

provide a good way for accessing precise locations of text for easy
manipulation with Virtual View's quick keys (discussed below).
The advantage of using UIA is evident when obtaining text from windows
that do not show up in the OSM. WPF applications, and other Direct2D
programs, are good examples of windows that have rich UIA information
while at the same time are completely void of clips. The UIA method

obtains

text directly from the window rather than from the OSM, often 
resulting

in
higher text accuracy.
OCR is a preferred choice when working with graphical interfaces, or
files
containing images. While you don't have the benefit of knowing exactly
where the text is presented on the screen, you do have access to
otherwise
completely inaccessible text.
5     The Virtual View Window

After Virtual View has obtained the text from the selected source, 
using

the

chosen method, it will display the resulting text in a read only edit
box.

The

Virtual View dialog contains a menu bar with the following options:

*         File (F)

o   Open (O) - Control-O - Opens support file types.

o   Save (S) - Control-S - Saves current text.

o   Exit (X) - Exits Virtual View

*         Edit (E)

o   Select All (S) - Control-A - Select all of the text.

o   Copy All (C) - Copies all of the text to the clipboard.

o   Copy Selection (O) - Control-C - Copies the selected text to the

clipboard.



o   Append All (A) - Appends all of the text to the clipboard 
contents.


o   Append Selection (P) - Appends the selected text to the clipboard
contents.

*         View (V)

o   Clips (C) - Switches to the Clips method of accessing text.

o   UIA (U) - Switches to the UIA method of accessing text.

o   OCR (O) - Switches to the OCR method of accessing text.

*         Help (H)

o   Help and Options (H) - Displays options for controlling Virtual
View's

text

recognition characteristics. Refer to Additional Options discussed 
below


5.1   Quick Keys

Virtual View provides several "quick" keys that you can use to 
interact

with

the text, and have that interaction passed on to the window where the
text
originated. For example, if a button called Close exists in the window

where

you obtained the text from, you can place your cursor on the word 
Close,
and press the letter L to activate the Close button. Available 
quick keys

are:


*         L - Single Left Click

*         R - Single Right Click

*         D  - Double Left Click

*         M - Middle Click

*         P - Route Mouse

*         F - Focus Control
While these quick keys are always available, they are most useful with

Clips,

and become less useful when using the UIA and OCR, as the correlation

between text and originating window is less reliable with those 
methods.

6     Additional Options
The Help and Options dialog contains the following options for
controlling
how Virtual View behaves and how it processes images using the OCR
method:


*         Include Labeled Graphics (I) - Check box - Default 
unchecked -

Checking

this option will cause Virtual View to include labeled graphics when

obtaining

text using the Clips method.

*         Include Window Types (N) - Check box - Default unchecked -

Checking

this option will cause Virtual View to include windows types (edit 
box,

button, etc.) when obtaining text using the Clips method.


*         Always Ask for Window Type (L) - Check box - Default 
checked -

This

option controls whether or not you're initially asked for a source and

method

when pressing the Virtual View hotkey.

*         Prompt When Alt-Print Screen Is Pressed (P) - Check Box -

Default

checked - This option controls whether or not you're prompted to have

Virtual View virtualize the image that's captured by the operating 
system

when the Alt-Print Screen hotkey is pressed.


*         Enable Image Processing (E) - Check box - Default 
unchecked -

Enables

or disables additional image processing for enhancing the results 
of text

recognition when using the OCR method.


*         Contrast (C) - Up/Down Control - Default 50 - This 
control is

only

enabled when Enable Image Processing is checked, and controls the
contrast

of the image before text processing occurs. A lower number will 
produce a

darker image, while a higher number will produce a lighter image.

*         Resize Multiplier (R) - Up/Down Control - Default 1 - This

control is only

enabled when Enable Image Processing is checked, and controls the 
final

size

of the image before text processing occurs. A larger number will 
increase

the

dimensions of the image, but also increases the size of the file, and

could

result in decreased improvement.

*         Use Grayscale (U) - Check box - Default unchecked - This
control

is only

enabled when Enable Image Processing is checked, and controls 
whether or

not the image is in color or black and white before text processing

occurs.



*         Save Image (S) - Check box - Default unchecked - This 
control

is

only

enabled when Enable Image Processing is checked, and will save the 
last

OCR'd image to the %temp% directory. This is only useful for debugging
purposes.

*         Reset Defaults (T)- Button - This control will reset the
options

to the

factory defaults.


*         Help - Button - This control will display the default app 
help,

which

includes this help text, app version details, and the Hotkey 
Manager for

controlling the default Virtual View hotkey.

7     A Note About Elevation

By default, Virtual View can only access windows that are not 
elevated,

or
opened as a non-administrative user. You can control this behavior by
checking the Require UAC (R) check box in the initial source selection

dialog.

When checked, Virtual View will become elevated, and will then be 
able to

access elevated windows.
8     Troubleshooting

Virtual View's OCR capabilities require operating system components 
that

may not be installed by default. If you find that you're consistently

getting

"No text found" after using the OCR method, try re-installing the TIFF

IFilter

by doing the following:

1.       Open an admin command prompt.


a.       You can do this by opening the start menu, typing cmd, 
pressing

Control-Shift-Enter, and (assuming you have the User Account Control
feature enabled) confirming the UAC prompt.

2.       In the command prompt window, enter the following:
dism /online /Enable-Feature /FeatureName:TIFFIFilter

3.       Press enter, and wait for the command to complete. When it's

finished,

type exit, and press enter.

4.       Reboot
Virtual View currently supports Windows 7 and up. There does, however,

seem to be an issue with Windows 8.1 where the OCR capabilities are 
not
available, even though they have been enabled. Some users have 
reported

success by following the steps outlined in the following Microsoft
Technet
article:
https://technet.microsoft.com/en-us/library/dd744701%28v=ws.10%29.aspx

--
Aaron Smith
Web Development * App Development * Product Support Specialist Ai
Squared * 725 Airport North Office Park, Fort Wayne, IN 46825
260-489-3671 * www.aisquared.com

To insure that you receive proper support, please include all past

correspondence (where applicable), and any relevant information 
pertinent

to your situation when submitting a problem report to the Ai Squared
Technical Support Team.

_______________________________________________
Any views or opinions presented in this email are solely those of the

author

and do not necessarily represent those of Ai Squared.

For membership options, visit http://lists.window-

eyes.com/options.cgi/talk-window-eyes.com/joe%40vi-ability.demon.co.uk. 


For subscription options, visit

http://lists.window-eyes.com/listinfo.cgi/talk-

window-eyes.com
List archives can be found at

http://lists.window-eyes.com/private.cgi/talk-

window-eyes.com

--
Joe Paton <[email protected]>




_______________________________________________
Any views or opinions presented in this email are solely those of the
author
and do not necessarily represent those of Ai Squared.

For membership options, visit

http://lists.window-eyes.com/options.cgi/talk-window-eyes.com/rod_hutton%40h 


otmail.com.
For subscription options, visit
http://lists.window-eyes.com/listinfo.cgi/talk-window-eyes.com
List archives can be found at
http://lists.window-eyes.com/private.cgi/talk-window-eyes.com
_______________________________________________

Any views or opinions presented in this email are solely those of 
the author

and do not necessarily represent those of Ai Squared.

For membership options, visit

http://lists.window-eyes.com/options.cgi/talk-window-eyes.com/huber.kevin7%40gmail.com. 


For subscription options, visit
http://lists.window-eyes.com/listinfo.cgi/talk-window-eyes.com
List archives can be found at
http://lists.window-eyes.com/private.cgi/talk-window-eyes.com


_______________________________________________

Any views or opinions presented in this email are solely those of the 
author and do not necessarily represent those of Ai Squared.



For membership options, visit 
http://lists.window-eyes.com/options.cgi/talk-window-eyes.com/tom.kingston%40charter.net.
For subscription options, visit 
http://lists.window-eyes.com/listinfo.cgi/talk-window-eyes.com
List archives can be found at 
http://lists.window-eyes.com/private.cgi/talk-window-eyes.com



_______________________________________________

Any views or opinions presented in this email are solely those of the 
author and do not necessarily represent those of Ai Squared.



For membership options, visit 
http://lists.window-eyes.com/options.cgi/talk-window-eyes.com/trailerdavid%40hotmail.com.
For subscription options, visit 
http://lists.window-eyes.com/listinfo.cgi/talk-window-eyes.com
List archives can be found at 
http://lists.window-eyes.com/private.cgi/talk-window-eyes.com





_______________________________________________
Any views or opinions presented in this email are solely those of the author 
and do not necessarily represent those of Ai Squared.

For membership options, visit 
http://lists.window-eyes.com/options.cgi/talk-window-eyes.com/archive%40mail-archive.com.
For subscription options, visit 
http://lists.window-eyes.com/listinfo.cgi/talk-window-eyes.com
List archives can be found at 
http://lists.window-eyes.com/private.cgi/talk-window-eyes.com






















					Reply via email to