Is Canon's CHDK Hack Dead?

in alt.fan.rush-limbaugh, about: Is Canon's CHDK Hack Dead?;
On Wed, 20 Apr 2011 16:25:09 +0200, Wolfgang Weisselberg wrote:

Crash! wrote:

About then, I decided to get the SX120 (as you may recall,)
because of CHDK, and while it was *STILL* not complete, it was
seemingly almost complete, it seemed sure it would be
done by Xmas.. (See my below repost.)
[...]
But it's *STILL* NOT DONE, it's STILL only in Beta!

Feel free to donate money or time to the effort.
-Wolfgang

I'll give the Beta a shot. If it's missing any of the below,
I'll put some time in. I've donated about a dozen DOS
utilities and a game to the public domain way back in the day.
I cringe because knowing me, it could suck me in for 100s
of hours.

in rec.photo.digital, about: Year Later, CHDK coming SOON for
Canon Ultra-Zoom SX120. - Motion Detection, Remote Shutter, Time
Lapse, etc;
On Tue, 14 Sep 2010, Crash! wrote:

The Canon PowerShot SX120 IS 10X zoom was released in London, UK
19th August 2009.

about: Is Canon's CHDK Hack Dead?;
On Wed, 20 Apr 2011, ken d wrote:


X-Newsreader: Forte Agent 4.1/32.1088
X-Usenet-Provider: http://www.giganews.com
NNTP-Posting-Host: 4.159.165.238
X-Complaints-To:

Oh brother. THAT troll. I should have known.
Still cant get laid, huh?
How many sockpuppets you running these days?

in alt.fan.rush-limbaugh, about: Is Canon's CHDK Hack Dead?;
On Tue, 19 Apr 2011 23:58:51 -0500, ken d wrote:


On Tue, 19 Apr 2011 18:14:34 -0700, Crash! wrote:


Oh really!? You seriously think they left that
firmware accessable all these years BY ACCIDENT!?
Perhaps you know something I don't? Like......?

Like? Like I now fully know that you're a total moron.

You give a nice FYI, thanks. But you failed to answer
my question: Why haven't they plugged that hole.
After all, CHDK's overview reads a great deal
like the overview of many Trojans; "...able to
seize control...."

Communicating with computers via blinkies goes
back beyond the 1940s, every gate geek did it in the '70s.
From your description it was originally an
unplugged hole (access) for firmware debugging.
Normally in software authorship those are REMmed out
near completion, and usually removed at final release.
In the case of hardware/firmware often a plugin bus
(or even an led module) is used in the prototypes.
You also describe common reverse engineering methods.


Note: you only describe the output functions, not input,
but I get the idea, thanks again. My point is, while it's
clever, it's not new, likely nobody re-invented the wheel.
If you got off your knees, I doubt your heros would mind.


If you had at least read the history of CHDK you'd know to not even pose a
comment like that. The firmware was NEVER accidentally accessible all these
years. The very first firmware had to be read out through a blinking LED on
the camera body, reading the LED blinks with a photo-diode, that signal
then fed into a sound-card, then software written to filter the LED light
intensity signal into digital bits. This is how ALL the firmwares of the
very first cameras in the CHDK project were obtained. Each one had to have
a different blinker utility written for it by randomly poking memory
addresses until one was found that would light any LED on the camera body,
no two cameras nor firmware numbers were ever the same for useable memory
locations.

Does that sound "accidentally accessible" to you? No camera firmware author
in their right mind would have even considered that their firmware could
have been read that way one day.

It is my guess that yes it was originally "accidentally
accessible." Now it's obviously intentionally accessible,
since they have not plugged those holes: 1) input nor 2) output.
"Obtuse," as somebody said, yes. Why they choose to play
that game, who knows, I've posted some guesses.

What you may not be aware of is, there is and MUST BE a
firmware read-instruction-from-memcard to instruction
reader(cpu) function IN THE CAMERA.
That can not still be an accident. It is likely that CHDK
works because the camera is permitted to, if not designed to.
("Read-instruction" is NOT a "read-image!")



This doesn't even include finding the
right encryption keys to make the blinked-out bits legible. That's yet a
whole other process.

YOU are a ****ING MORON TROLL WITHOUT ONE CLUE.

Go find some other topic to fill your troll's desperate need for attention.
You'll not waste any more of my valuable time.

Laughing.....

On Wed, 20 Apr 2011 11:08:47 -0700, Crash! wrote:

Why haven't they plugged that hole.

Wrong, moron. New ways had to be found each year that new cameras were
released. No doubt next years' firmware and design will plug previous
leaks.

On Wed, 20 Apr 2011 11:36:03 -0700 (PDT), David Dyer-Bennet
wrote:

On Tuesday, April 19, 2011 2:10:42 PM UTC-5, ken d wrote:
On 19 Apr 2011 17:44:19 GMT, ray wrote:


You're obviously much more educated about the GPL than I am, but perhaps
you can explain to me how it would violate the GPL. I was under the
impression that GPL code could indeed be included in commercial products
as long as the source is made available.


And therein lies the problem. No corporation will release their own tweaked
source-code back into the community from which they have taken it. They
exist, subsist, and survive on a "ME ME ME ME MINE MINE MINE MINE"
principle.

And THAT is precisely why Red Hat has never released any of their
changes to Linux back into the pool. Neither has Suse. And definitely
Sun Microsystems would never consider releasing most of Solaris and
all of ZFS as (their own license of) Open Source.

Oh, but wait a minute. All those things DID happen.

Therefore, you're talking nonsense.

Three companies out of how many? You forgot to mention all the companies
that use expensive lawyer-leverage to completely cripple the originators of
their open-source code and then claim ownership of it. Just like what
happened to the code for PanoTools and why it is no longer being developed
(openly). There are hundreds if not thousands more examples of this
"corporate open-source" system to bring a clearer perspective to the three
you mention.

On Wed, 20 Apr 2011 14:52:43 -0700 (PDT), David Dyer-Bennet
wrote:

On Wednesday, April 20, 2011 1:55:39 PM UTC-5, ken d wrote:
On Wed, 20 Apr 2011 11:36:03 -0700 (PDT), David Dyer-Bennet
wrote:

On Tuesday, April 19, 2011 2:10:42 PM UTC-5, ken d wrote:
On 19 Apr 2011 17:44:19 GMT, ray wrote:


You're obviously much more educated about the GPL than I am, but perhaps
you can explain to me how it would violate the GPL. I was under the
impression that GPL code could indeed be included in commercial products
as long as the source is made available.


And therein lies the problem. No corporation will release their own tweaked
source-code back into the community from which they have taken it. They
exist, subsist, and survive on a "ME ME ME ME MINE MINE MINE MINE"
principle.

And THAT is precisely why Red Hat has never released any of their
changes to Linux back into the pool. Neither has Suse. And definitely
Sun Microsystems would never consider releasing most of Solaris and
all of ZFS as (their own license of) Open Source.

Oh, but wait a minute. All those things DID happen.

Therefore, you're talking nonsense.

Three companies out of how many? You forgot to mention all the companies
that use expensive lawyer-leverage to completely cripple the originators of
their open-source code and then claim ownership of it. Just like what
happened to the code for PanoTools and why it is no longer being developed
(openly). There are hundreds if not thousands more examples of this
"corporate open-source" system to bring a clearer perspective to the three
you mention.

Those three companies have released a LOT of code, though; just
counting companies doesn't tell the whole story (and I'm sure there
are many, many others I'm not thinking of off the top of my head).

Your statement was "No corporation will release their own tweaked
source-code back into the community from which they have taken it."
Even one counter-example falsifies an absolute claim like that, and
you've already agreed there are at least three.

If your real point is that companies tend to be profit-seeking and not
all that ethical, and that this is a danger to watch out for, and that
there are many examples of bad things being done here, I can agree with
you (and watch what Oracle is doing with Sun's Open Source projects; that's
not looking at all good).

But if you insist on exaggerating it into an easily-disproven absolute
claim, well.


My error is forgetting there's always an exception to the rule. We agree
more than not.

about: Is Canon's CHDK Hack Dead?;
On Wed, 20 Apr 2011, ken d wrote:

X-Newsreader: Forte Agent 4.1/32.1088
X-Usenet-Provider: http://www.giganews.com
NNTP-Posting-Host: 4.159.165.238X-Complaints-To:


On Wed, 20 Apr 2011 11:08:47 -0700, Crash! wrote:

Why haven't they plugged that hole.

Wrong, moron. New ways had to be found each year that new cameras were
released. No doubt next years' firmware and design will plug previous
leaks.

Before I just thought you were a looney troll.
But yer a liar.

"ken" == ken d writes:

ken Three companies out of how many? You forgot to mention all the
ken companies that use expensive lawyer-leverage to completely cripple
ken the originators of their open-source code and then claim ownership
ken of it. Just like what happened to the code for PanoTools and why it
ken is no longer being developed (openly). There are hundreds if not
ken thousands more examples of this "corporate open-source" system to
ken bring a clearer perspective to the three you mention.

You're now trolling (if not already before).

I cover all sorts of open source projects on my FLOSS Weekly podcast,
and spend a lot of time talking to a lot more people than I could ever
have on that show.

Yes, there are a *few* bad corps out there that have modified GPL
software and kept the results locked up, but they've all been pursued
successfully, and the result has been unlocked. *All* of them.

Maybe you're thinking of BSD/MIT/Apache/Artistic-style projects, where
the corporation is indeed *permitted* to lock up modifications? Guess
what? They *can* legally do that. Oddly enough, *more* companies, even
faced with the permissiveness of the BSD-style license, *still* release
as much code as they can. I know, I've worked for a number of them.

So, you're not only wrong, you're wrong.

(And your history of Panotools is wrong as well... it started out closed
source, and parts of it were opened. And it looks like the sf.net
project is still alive and well.)

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
URL:http://www.stonehenge.com/merlyn/
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.posterous.com/ for Smalltalk discussion

ray wrote:
On Wed, 20 Apr 2011 16:20:04 +0200, Wolfgang Weisselberg wrote:

ray wrote:
You're obviously much more educated about the GPL than I am, but
perhaps you can explain to me how it would violate the GPL. I was under
the impression that GPL code could indeed be included in commercial
products as long as the source is made available.

Is Canon free to release the source code under the GPL (or a
GPL-compatible license)? Or have they perhaps licensed some of the
code, maybe the denoising?

As far as I can see, canon is free to release anything they want -
however, the source to chdk is not theirs to release. But if chdk is
operating under GPL, it is their responsibility to release the source. It
is my understanding that if canon were to incorporate any or all of the
chdk source in their firmware, they would be obligated to release that
portion they used - not the whole thing.

They cannot incorporate *any* GPL's code into a specific
"program" without releasing the source to that entire
"program". In this case it pretty much means that the
entire source code for the firmware would have to be
released. There are clearly ways to avoid doing that,
but none that work well for camera firmware.

Essentially they'd have to release the camera in a form
that did not use the GPL'd code and include some way to
load that code as an additional, and optional, module.
That would mean that only the source for that module
would have to be released under the GPL.

Certainly that is possible, and just as certainly it would
not be likely to benefit Canon or any other major company
in terms of marketing. Meaning: it ain't gonna happen!

Is Canon willing to release the source code under the GPL even if they
can? After all, they'll tell the competition what they are doing, which
can be very valuable, even if you won't be copying the code ...

That's B.S. Everyone in the industry knows what they are doing - and what
everyone else is doing. 'Trade secrets' are very limited and greatly
overrated.

I don't think that is B.S. at all, nor do I think
everyone actually knows what everyone else is doing.
Trade secrets are generally not long lived (except
perhaps for the formula for Coke), but the market cycles
faster than the secrets.

--
Floyd L. Davidson http://www.apaflo.com/floyd_davidson
Ukpeagvik (Barrow, Alaska)

ray wrote:
On Wed, 20 Apr 2011 16:20:04 +0200, Wolfgang Weisselberg wrote:
ray wrote:

Is Canon free to release the source code under the GPL (or a
GPL-compatible license)? Or have they perhaps licensed some of the
code, maybe the denoising?

As far as I can see, canon is free to release anything they want -

Nope. They can only release what is theirs or for which they
have the rights to release.

however, the source to chdk is not theirs to release. But if chdk is
operating under GPL, it is their responsibility to release the source.

Huh? They cannot release the source, and it is their
responsibility to release the source?

It
is my understanding that if canon were to incorporate any or all of the
chdk source in their firmware, they would be obligated to release that
portion they used - not the whole thing.

They's only need to release the portion of the CHDK they used
--- together with the rest of the firmware. (CHDK isn't a
stand-alone program.)

Is Canon willing to release the source code under the GPL even if they
can? After all, they'll tell the competition what they are doing, which
can be very valuable, even if you won't be copying the code ...

That's B.S. Everyone in the industry knows what they are doing - and what
everyone else is doing. 'Trade secrets' are very limited and greatly
overrated.

Ah, yes. That explains why everyone has released the source code
for their firmware.

-Wolfgang