Javascript and Microsoft Windows

[Peter Olcott]

Does JavaScript represent its controls internally as Microsoft Windows controls,
or does it build them from scratch like Java?

[Peter Olcott]

"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:XM**************@objmedia.demon.co.uk...

In message <9WMEg.895$Tl4.547@dukeread06>, Peter Olcott <ol****@att.net>
writes

Matching pixel patterns won't work. You can't tell the difference

between a

screen shot of an input button and an input button based on pixel

patterns.

>

But, when is this ever a problem is actual GUI scripting?

When you are trying to control the GUI components created by the GUI scripting
language of choice, as you have been clearly trying to do, as evidenced by
your questioning elsewhere in this thread. You need to be able to locate these
controls. This is particularly a problem when such controls are not native
controls but created by the host (Firefox just draws them as it sees fit) or
by a host runtime (the Java runtime just draws them as it sees fit, and Swing
under Java does it a different way), etc.

If you can't locate them, you can't do what you are proposing. Hence my
comments in other replies.

As long as they are consistent for one application within a platform, even if
they vary willy nilly pell mell across applications and platforms, my technology
can easily "see" it all. There are also facilities for adapting to dynamic
changes to the look and feel of an application or platform.

>
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[Peter Olcott]

"Randy Webb" <Hi************@aol.comwrote in message
news:1_******************************@comcast.com. ..

Peter Olcott said the following on 8/16/2006 8:18 PM:

>"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:pc**************@objmedia.demon.co.uk...

>>In message <P0NEg.897$Tl4.797@dukeread06>, Peter Olcott <ol****@att.net>
writes
My technology is operational and has 100% accuracy at real-time speeds.
And is it resolution independent? I doubt you very much. If you were serious
you wouldn't be asking these questions about JavaScript as you'd know
already if you could identify the buttons/whatever on screen.

Here is what I know. I know that I can match thousands of pixel patterns from
millions of alternatives in a fraction of a second for the entire screen all
at once. Although GUI controls may vary quite a bit across differing
technologies and applications they rarely vary for the same program on the
same platform. Because of this my technology can be easily programmed to
"see" most any of these sort of things.

And my point, in my other reply, is that I don't believe that. Not for one
second, and for the reason I stated. It is not enough to "see" something by
visual recognition. Especially not by comparing pixels. Create a test page and
put a button on it:

<button>This is my button</button>

Then, take a screenshot of that button and place it on the page:

<img src="theButton.jpg">

And then please, oh please, tell me how you propose that you can tell the
difference between those two simply by comparing pixels. The answer is
simple - you can't.

In this case I can tell, can you see how I can tell?

>
Nor can your "program" know what clicking it is going to do so you can't
"control" it without clicking it first and by then it may be too late to try
to "control" it.

I am not saying that my programming language is all knowing, merely that it is
compatible with every application. It will still take a programmer to write the
programs. The type of things that it is intended for is to automate tasks that
previously required a person. If someone puts a dummy screen shot on the screen
that looks like a real screen, but, is really fake, my language can know that
something is wrong because it did not get the expected result in the expected
timeframe. With buttons one of the expected results is seeing the depressed
state of the button. If it pushes a button and does not "see" the depressed
state, then it will know that something is wrong and email or phone the
programmer.

>
If your program is combining the code of the page (which it can't always do)
and the visual representation, then it still won't be enough (and yes, I can
prove that also).

My language will be able to automate anything that does not require human
judgment.

>
--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq & newsgroup weekly
Javascript Best Practices - http://www.JavascriptToolbox.com/bestpractices/

[Randy Webb]

Peter Olcott said the following on 8/16/2006 8:34 PM:

"Randy Webb" <Hi************@aol.comwrote in message
news:1_******************************@comcast.com. ..

>Peter Olcott said the following on 8/16/2006 8:18 PM:

>>"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:pc**************@objmedia.demon.co.uk...
In message <P0NEg.897$Tl4.797@dukeread06>, Peter Olcott <ol****@att.net>
writes
My technology is operational and has 100% accuracy at real-time speeds.
And is it resolution independent? I doubt you very much. If you were serious
you wouldn't be asking these questions about JavaScript as you'd know
already if you could identify the buttons/whatever on screen.
Here is what I know. I know that I can match thousands of pixel patterns from
millions of alternatives in a fraction of a second for the entire screen all
at once. Although GUI controls may vary quite a bit across differing
technologies and applications they rarely vary for the same program on the
same platform. Because of this my technology can be easily programmed to
"see" most any of these sort of things.

And my point, in my other reply, is that I don't believe that. Not for one
second, and for the reason I stated. It is not enough to "see" something by
visual recognition. Especially not by comparing pixels. Create a test page and
put a button on it:

<button>This is my button</button>

Then, take a screenshot of that button and place it on the page:

<img src="theButton.jpg">

And then please, oh please, tell me how you propose that you can tell the
difference between those two simply by comparing pixels. The answer is
simple - you can't.

In this case I can tell, can you see how I can tell?

You or your program? And, based merely on visual appearance by
"comparing pixels", no, you can't.

>Nor can your "program" know what clicking it is going to do so you can't
"control" it without clicking it first and by then it may be too late to try
to "control" it.

I am not saying that my programming language is all knowing, merely that it is
compatible with every application. It will still take a programmer to write the
programs. The type of things that it is intended for is to automate tasks that
previously required a person.

Ahhh, now I think I understand. I have one of those programs, well I did
before I uninstalled it, it's called a keystroke macro program and I can
automate anything with it. So, how is your "patented technology" going
to help with that? And yes, the macro program I had had a built in logic
mechanism for it. It was rudimentary but it had one. And, it didn't
require the "pixel comparison".

So, how is yours better than that?

If someone puts a dummy screen shot on the screen that looks like a real screen,
but, is really fake, my language can know that something is wrong because it did
not get the expected result in the expected timeframe.

That's nothing new and can be handled even in javascript. In fact, it's
the basis of programming. Try something, if you get the results you
wanted, fine, if not, then something went wrong.

With buttons one of the expected results is seeing the depressed state of the button.

So, what is the "expected results" of clicking this image?

<a href="http://www.google.com" onclick="goSomeWhereElse()" id="myLink">
<img src="bigButton.jpg">
</a>

<script type="text/javascript">
document.getElementById('myLink').onclick=someOthe rFunction;
function someOtherFunction(){
return false;
}
</script>

If it pushes a button and does not "see" the depressed state,
then it will know that something is wrong and email or phone the
programmer.

But only if the programmer programs that. So it's nothing more than
error control.

>If your program is combining the code of the page (which it can't always do)
and the visual representation, then it still won't be enough (and yes, I can
prove that also).

My language will be able to automate anything that does not require human
judgment.

My keystroke macro program could do that. How is your new scripting
language going to be better than that?

--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq & newsgroup weekly
Javascript Best Practices - http://www.JavascriptToolbox.com/bestpractices/

[Peter Olcott]

"Randy Webb" <Hi************@aol.comwrote in message
news:TK******************************@comcast.com. ..

Peter Olcott said the following on 8/16/2006 8:34 PM:

>"Randy Webb" <Hi************@aol.comwrote in message
news:1_******************************@comcast.com ...

>>Peter Olcott said the following on 8/16/2006 8:18 PM:
"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:pc**************@objmedia.demon.co.uk...
In message <P0NEg.897$Tl4.797@dukeread06>, Peter Olcott <ol****@att.net>
writes
>My technology is operational and has 100% accuracy at real-time speeds.
And is it resolution independent? I doubt you very much. If you were
serious you wouldn't be asking these questions about JavaScript as you'd
know already if you could identify the buttons/whatever on screen.
Here is what I know. I know that I can match thousands of pixel patterns
from millions of alternatives in a fraction of a second for the entire
screen all at once. Although GUI controls may vary quite a bit across
differing technologies and applications they rarely vary for the same
program on the same platform. Because of this my technology can be easily
programmed to "see" most any of these sort of things.
And my point, in my other reply, is that I don't believe that. Not for one
second, and for the reason I stated. It is not enough to "see" something by
visual recognition. Especially not by comparing pixels. Create a test page
and put a button on it:

<button>This is my button</button>

Then, take a screenshot of that button and place it on the page:

<img src="theButton.jpg">

And then please, oh please, tell me how you propose that you can tell the
difference between those two simply by comparing pixels. The answer is
simple - you can't.

In this case I can tell, can you see how I can tell?

You or your program? And, based merely on visual appearance by "comparing
pixels", no, you can't.

>>Nor can your "program" know what clicking it is going to do so you can't
"control" it without clicking it first and by then it may be too late to try
to "control" it.

I am not saying that my programming language is all knowing, merely that it
is compatible with every application. It will still take a programmer to
write the programs. The type of things that it is intended for is to automate
tasks that previously required a person.

Ahhh, now I think I understand. I have one of those programs, well I did
before I uninstalled it, it's called a keystroke macro program and I can
automate anything with it. So, how is your "patented technology" going to help
with that? And yes, the macro program I had had a built in logic mechanism for
it. It was rudimentary but it had one. And, it didn't require the "pixel
comparison".

So, how is yours better than that?

>If someone puts a dummy screen shot on the screen that looks like a real
screen, but, is really fake, my language can know that something is wrong
because it did not get the expected result in the expected timeframe.

That's nothing new and can be handled even in javascript. In fact, it's the
basis of programming. Try something, if you get the results you wanted, fine,
if not, then something went wrong.

>With buttons one of the expected results is seeing the depressed state of the
button.

So, what is the "expected results" of clicking this image?

<a href="http://www.google.com" onclick="goSomeWhereElse()" id="myLink">
<img src="bigButton.jpg">
</a>

<script type="text/javascript">
document.getElementById('myLink').onclick=someOthe rFunction;
function someOtherFunction(){
return false;
}
</script>

>If it pushes a button and does not "see" the depressed state, then it will
know that something is wrong and email or phone the programmer.

But only if the programmer programs that. So it's nothing more than error
control.

>>If your program is combining the code of the page (which it can't always do)
and the visual representation, then it still won't be enough (and yes, I can
prove that also).

My language will be able to automate anything that does not require human
judgment.

My keystroke macro program could do that. How is your new scripting language
going to be better than that?

Keyboard macro programs can not be programmed. They have no intelligence at all.
They simply mindlessly repeat a sequence of keystrokes when one keystroke is
typed. Even if they had intelligence they would not be able to use it because
they are entirely blind. They could not take different actions based upon
different results, because they have no way to "see" any results.
http://en.wikipedia.org/wiki/Scripting_language

>
--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq & newsgroup weekly
Javascript Best Practices - http://www.JavascriptToolbox.com/bestpractices/

[Randy Webb]

Peter Olcott said the following on 8/16/2006 10:38 PM:

"Randy Webb" <Hi************@aol.comwrote in message
news:TK******************************@comcast.com. ..

>Peter Olcott said the following on 8/16/2006 8:34 PM:

>>"Randy Webb" <Hi************@aol.comwrote in message
news:1_******************************@comcast.co m...
Peter Olcott said the following on 8/16/2006 8:18 PM:

<snip>

>>>If your program is combining the code of the page (which it can't always do)
and the visual representation, then it still won't be enough (and yes, I can
prove that also).
My language will be able to automate anything that does not require human
judgment.

My keystroke macro program could do that. How is your new scripting language
going to be better than that?

Keyboard macro programs can not be programmed.

Are you that gullible? My keyboard macro program was *very* programmable.

They have no intelligence at all.

What good is "intelligence" when you want to automate things that "does
not require human judgment"? You don't. Its a simple click and go.

They simply mindlessly repeat a sequence of keystrokes when one keystroke is
typed.

When you are automating things that do not require human judgment (your
criteria), there is no need for intelligence. None at all.

Even if they had intelligence they would not be able to use it because
they are entirely blind. They could not take different actions based upon
different results, because they have no way to "see" any results.

You really are that gullible......

--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq & newsgroup weekly
Javascript Best Practices - http://www.JavascriptToolbox.com/bestpractices/

[The Magpie]

Peter Olcott wrote:

>
When anyone asks the question "Is X possible?" this logically entails the
universal set of every possible assumption. How about this, under any possible
assumption to what extent is it possible to make a universal scripting language
that can control every program that will run under modern operating system
versions?

While I disagree with your figures on PC ownership (try looking at the
figures for Singapore some time), the above gets us closer to being able
to answer your question - a little.

The answer? It isn't possible - and what is more, if you do not have a
prototype or a demonstration of it in operation, then your patent is not
valid anyway.

[Peter Olcott]

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

Peter Olcott wrote:

>>
When anyone asks the question "Is X possible?" this logically entails the
universal set of every possible assumption. How about this, under any
possible
assumption to what extent is it possible to make a universal scripting
language
that can control every program that will run under modern operating system
versions?

While I disagree with your figures on PC ownership (try looking at the
figures for Singapore some time), the above gets us closer to being able
to answer your question - a little.

http://www.census.gov/prod/2005pubs/p23-208.pdf
Page 12 table F, the number of people using computers at work 76,570 million.
The number of people using computers either at work or at home is 126,937
million. The PC ownership figures are different. There were 69,912 million
households in 2003 with at least one computer, representing about 62% of the
households. Since some of these households would have more than one computer,
this is not the number of computers.

>
The answer? It isn't possible - and what is more, if you do not have a
prototype or a demonstration of it in operation, then your patent is not
valid anyway.

That is not how patents work. Legally filing a patent is equivalent to building
a prototype, the legal term for this is constructive reduction to practice.

[Stephen Kellett]

In message <TA0Fg.1163$Tl4.175@dukeread06>, Peter Olcott
<ol****@att.netwrites

>That is not how patents work. Legally filing a patent is equivalent to
building
a prototype, the legal term for this is constructive reduction to practice.

No, thats not how it works. I've done this. You must have a proof of
concept that works as well as the idea. The original poster was right.

Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[Peter Olcott]

"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:vx**************@objmedia.demon.co.uk...

In message <TA0Fg.1163$Tl4.175@dukeread06>, Peter Olcott <ol****@att.net>
writes

>>That is not how patents work. Legally filing a patent is equivalent to
building
a prototype, the legal term for this is constructive reduction to practice.

No, thats not how it works. I've done this. You must have a proof of concept
that works as well as the idea. The original poster was right.

http://patft.uspto.gov/netacgi/nph-P...&RS=PN/7046848
Apparently I know more about this process than both of you. Its a good thing
that truth is not a democracy, otherwise mediocrity would rule!

>
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[The Magpie]

Peter Olcott wrote:

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

>The answer? It isn't possible - and what is more, if you do not have a
prototype or a demonstration of it in operation, then your patent is not
valid anyway.

That is not how patents work. Legally filing a patent is equivalent to building
a prototype, the legal term for this is constructive reduction to practice.

Actually, it isn't and any patents *without* prototypes are considered
"pending" by most offices until demonstrated. I am aware that in the USA
they have made a practice of allowing such patent claims and this has
led to a disaster in patent law there... which is why so many nations
are in patent dispute with the US.

[RobG]

The Magpie wrote:

Peter Olcott wrote:

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

The answer? It isn't possible - and what is more, if you do not have a
prototype or a demonstration of it in operation, then your patent is not
valid anyway.

That is not how patents work. Legally filing a patent is equivalent to building
a prototype, the legal term for this is constructive reduction to practice.

Actually, it isn't and any patents *without* prototypes are considered
"pending" by most offices until demonstrated. I am aware that in the USA
they have made a practice of allowing such patent claims and this has
led to a disaster in patent law there... which is why so many nations
are in patent dispute with the US.

It is interesting to note that over 350,000 patents are filed each year
with the US Patent Office (Microsoft alone filed over 3,000 in 2005,
I'll bet IBM filed a similar number). It is absurd of anyone to think
that the fact that a patent is granted means that the "invention" is
therefore guaranteed to actually be patentable - there is absolutely no
way that the US Patent Office has the time or resources to thoroughly
vet each and every application. That task is left to others who wish
to challenge the patent.

The purpose of filing patents is now a defensive business practice -
note IBM's reaction to the SCO case. They immediately returned fire by
accusing SCO of breaching a bunch of IBM patents, for no other reason
than to make SCO spend a fortune defending themselves and reduce the
funds available for their case against IBM (I am not supporting SCO's
case here, just pointing out a recent use of patents as a defensive
strategy).
--
Rob

[Stephen Kellett]

In message <1K7Fg.1204$Tl4.383@dukeread06>, Peter Olcott
<ol****@att.netwrites

>Apparently I know more about this process than both of you. Its a good thing
that truth is not a democracy, otherwise mediocrity would rule!

If you were correct I could patent faster than light travel without
having a proof of concept. There are plenty of bogus USA patents that
have been granted. Looks like yours is one of them if you haven't got a
working proof. I seem to remember a story on Slashdot a while back about
someone getting a US patent on something as similarly ridiculous as
faster than light travel.

Having filed for patents myself (both UK and USA) we had to be able to
demonstrate the working concept, which of course wasn't for faster than
light travel.

Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[Peter Olcott]

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec******************@news.demon.co.uk...

Peter Olcott wrote:

>"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

>>The answer? It isn't possible - and what is more, if you do not have a
prototype or a demonstration of it in operation, then your patent is not
valid anyway.

That is not how patents work. Legally filing a patent is equivalent to
building
a prototype, the legal term for this is constructive reduction to practice.

Actually, it isn't and any patents *without* prototypes are considered
"pending" by most offices until demonstrated. I am aware that in the USA
they have made a practice of allowing such patent claims and this has
led to a disaster in patent law there... which is why so many nations
are in patent dispute with the US.

I forgot that the rest of the world might handle patents differently than in the
United States. I see no reason why the United States model would not work
without problems. If it is later demonstrated that an idea will not work, then a
patent would be of no value.

[Peter Olcott]

"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:gf**************@objmedia.demon.co.uk...

In message <1K7Fg.1204$Tl4.383@dukeread06>, Peter Olcott <ol****@att.net>
writes

>>Apparently I know more about this process than both of you. Its a good thing
that truth is not a democracy, otherwise mediocrity would rule!

If you were correct I could patent faster than light travel without having a
proof of concept. There are plenty of bogus USA patents that have been
granted. Looks like yours is one of them if you haven't got a working proof. I
seem to remember a story on Slashdot a while back about someone getting a US
patent on something as similarly ridiculous as faster than light travel.

I forgot that I sent in one example of a working prototype when I filed my
patent. I have read that this is not required in the United States. The specific
legal term is [constructive reduction to practice]. This term means that filing
a patent is considered legally the same as building a prototype.

>
Having filed for patents myself (both UK and USA) we had to be able to
demonstrate the working concept, which of course wasn't for faster than light
travel.

Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[Stephen Kellett]

In message <l7lFg.1439$Tl4.346@dukeread06>, Peter Olcott
<ol****@att.netwrites

>I forgot that the rest of the world might handle patents differently
than in the
United States. I see no reason why the United States model would not work
without problems. If it is later demonstrated that an idea will not
work, then a
patent would be of no value.

In which case you have not thought about the problems in much depth.

Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[Peter Olcott]

"Stephen Kellett" <sn***@objmedia.demon.co.ukwrote in message
news:6t**************@objmedia.demon.co.uk...

In message <l7lFg.1439$Tl4.346@dukeread06>, Peter Olcott <ol****@att.net>
writes

>>I forgot that the rest of the world might handle patents differently than in
the
United States. I see no reason why the United States model would not work
without problems. If it is later demonstrated that an idea will not work, then
a
patent would be of no value.

In which case you have not thought about the problems in much depth.

Well its far better than the alternative first-to-file, making it far easier for
big corporations to steal the ideas of individual inventors. Although
first-to-invent might be more difficult to administer, it is a much more
equitable system. Because of this I would not trust the rest of the world's
judgement against the United States on patent matters. The requirement of
providing a working prototype is yet another bias towards big business over the
individual.

>
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk/software.html
Computer Consultancy, Software Development
Windows C++, Java, Assembler, Performance Analysis, Troubleshooting

[The Magpie]

Peter Olcott wrote:

>
I forgot that the rest of the world might handle patents differently than in the
United States. I see no reason why the United States model would not work
without problems. If it is later demonstrated that an idea will not work, then a
patent would be of no value.

Unfortunately, both you and the US Patent Office are quite wrong. Almost
the entire planet has problems with the shambles that the US patent
process has become.

[Peter Olcott]

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

Peter Olcott wrote:

>>
I forgot that the rest of the world might handle patents differently than in
the
United States. I see no reason why the United States model would not work
without problems. If it is later demonstrated that an idea will not work,
then a
patent would be of no value.

Unfortunately, both you and the US Patent Office are quite wrong. Almost
the entire planet has problems with the shambles that the US patent
process has become.

Can you be more specific?

[The Magpie]

Peter Olcott wrote:

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

>Unfortunately, both you and the US Patent Office are quite wrong. Almost
the entire planet has problems with the shambles that the US patent
process has become.

Can you be more specific?

How can I possibly need to? Your country is involved in patent disputes
over pharmaceuticals with Brazil and India, for software with the entire
EU, China and Japan, with the entire planet (through the WHO) over
ludicrous claims for "genetic patents" (particularly for BRCA breast
cancer tests) and many, many more. American patents are an international
farce.

[Peter Olcott]

"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

Peter Olcott wrote:

>"The Magpie" <us****@mpreston.demon.co.ukwrote in message
news:ec*******************@news.demon.co.uk...

>>Unfortunately, both you and the US Patent Office are quite wrong. Almost
the entire planet has problems with the shambles that the US patent
process has become.

Can you be more specific?

How can I possibly need to? Your country is involved in patent disputes
over pharmaceuticals with Brazil and India, for software with the entire
EU, China and Japan, with the entire planet (through the WHO) over
ludicrous claims for "genetic patents" (particularly for BRCA breast
cancer tests) and many, many more. American patents are an international
farce.

With {first to invent} all that it takes to resolve these disputes is to prove
who was the first to invent. {first to file} makes it far too easy for big
corporations to steal the work of individual inventors.