With help from igorsky I'm about ready to release a firmware flasher that works via the SD/MS card slot. I picked up a fresh new Reader today and I'm using it as the test subject.

Question to the rest of you: is Sony right to be worried? Is someone going to use the technology to break the DRM? I enjoy being able to buy reasonably fresh main stream ebooks and I'm not interested in producing and releasing a tool that results in the destruction of that ability.

I can assure you that flashing custom firmware is a pretty cumbersome way to try breaking the DRM when the desktop version is a much much easier target.

Yeah, it would have been much easier to disassemble CONNECT, step through it with a Windows debugger and try to break the DRM that way. If they used strong encryption like they do with firmware update, this process would have been rather futile any way though, wouldn't it.

I don't see what point breaking the DRM in the Reader itself would even have -- it's not like the Reader really has an output mechanism. I suppose you could download the 'broken' files back to your computer, but why would anyone bother? Somebody might try it just to say they did it, but I think igorsk and porkupan have the nub of it: if somebody breaks the Connect DRM it will be done on a more versatile (not to mention powerful) piece of hardware than the Reader itself. :shrug:

I just don't see that being able to (re)flash the Reader would make much difference in the matter either way.

I agree with the concern... break DRM and lose the ability to buy the e-books we want at Connect, plus potentially face additional restrictions on developing for the device.

I don't have the technical knowledge to understand the risks of Scotty's flasher, but was really glad to hear igorsk's thoughts (that it would not open any doors toward DRM-breaking)!

P.S. No - this doesn't mean I/we like DRM. But let's please not turn this into yet another DRM bashing thread. That's a separate topic that needs a separate thread.

P.P.S. I wonder if Scotty's wife knows that he's a flasher?

One of the easiest ways to "break" the DRM would be to write an app that automates the connect software and does window captures to bitmaps, then runs OCR on those. That wouldn't require any crypto smarts or reverse engineering, and any half-dencent Windows programmer could do it in a few hours (assuming they had some OCR software to use). Taking the firmware route (or attacking the desktop software) may result in a class break but its much harder to do, especially the firmware route (just contrast the number of people who can read x86 disassembly with symbolic debug info available for many of the DLLs with the number that can read ARM assembly).

P.S. No - this doesn't mean I/we like DRM. But let's please not turn this into yet another DRM bashing thread. That's a separate topic that needs a separate thread.



How about a DRM-praising thread? ;)

DRM is actually a necessary and potentially good technology if it ever gets done right. I think its fair for a content creator to be paid by each consumer, but the media costs should be borne only once, and all costs other than the royalties should be transferable along with ownership. Unfortunately the people who inflict the DRM are not the content creators, but the middlemen - exactly the people who should be cut out of the deal when ownership is transferred. As a result, DRM currently is focused on preventing transfer of ownership, rather than encouraging it (and the latter, if done properly, would actually maximize the revenue accrued to the content creator).

Unfortunately the people who inflict the DRM are not the content creators, but the middlemen - exactly the people who should be cut out of the deal when ownership is transferred.I've been given to understand that the DRM itself is a not insignificant fraction of the cost of DRMed content. There's a price cutting opportunity, right there. And customers would love it! :nice:

I think Baen has proven DRM isn't needed.

Amazon and Fictionwise have both proven to me that some merchants can't be trusted with DRM, they've both ripped me off.

The first Reader DRM attack I'd think people would sell out Sony for would be to allow the Reader to be attached to multiple CONNECT accounts so people could "lend" ebooks purchased from the Sony CONNECT store.

I think Baen has proven DRM isn't needed.If you're looking for an argument on that, don't look at me. :nice:

The problem is that while they've demonstrated that a non-DRM approach works fine, there are eighty-eleven other pubs who don't find that demonstration convincing. :shrug:

Fortunately for me, a lot of what I want to read comes from or through Baen. :nice:

I think that being able to flash the reader will work somewhat akin to rockbox on ipod. Custom firmware might be enabled, but it will be limited to enthusiasts.

The first Reader DRM attack I'd think people would sell out Sony for would be to allow the Reader to be attached to multiple CONNECT accounts so people could "lend" ebooks purchased from the Sony CONNECT store.

It's already limited to two sony readers attached to the same account, not any two accounts that can share books.

It's already limited to two sony readers attached to the same account....Um ... no it's not. You can have a maximum of 5 Readers on a single account, if you're willing to only have 1 PC on the account -- the limit is 6 devices per account, with both Readers and PCs counting as "devices."

What you can't do is have a single "device" (be it Reader or PC) on more than one account, that's the sort of 'hack' that scotty1024 is theorizing might be attractive to the average person. :shrug:

Not only 6 devices but you have to connect the device to Sony's Connect store to de-authorize it. My main computer's primary hard drive recently crashed. Sony sees the same computer with a new hard drive as a new device. However I can't use that device to "de-authorise" it because it no longer works. Sony says they de-authorized it but it still shows as one of my devices. Go figure! At this rate, we will all eventually have only our reader that is usable and also authorized. :angry:

And the hack would most likely be to "clone" your friend's Reader onto your Reader so they could copy ebooks on to it without needing to add it to their account.

Your Reader would then have a "split personality" where it could either be your or your friend's Reader as needed.

Otherwise Sony might start to notice all the accounts with 5 Readers and start going "Hmmm, have we even sold that many Reader yet???". :)

My main computer's primary hard drive recently crashed. <snip> Sony says they de-authorized it but it still shows as one of my devices.I had a PC crash the first week of January, and hit the same problem, however, when I called the number they de-authed the old PC just fine and it dropped off my devices list. Maybe you ought to try calling them again, it sounds like it didn't go through correctly. I agree, though, that's frustrating. :sad2:

drm is irritating, if anything, i want to be able to quote passages from the CONNECT software into the clipboard.

This is a tricky bit, but IMO, I don't think it would be abused to exploit the DRM of the device. While any new technology can be exploited to do things its not supposed to, I cannot think of anything I'd need to exploit on my Reader that I can't do from my desktop first. It's cool that you've figured out how to do this, and judging from peoples' threads in this group, I think everyone here is responsible enough that they wouldn't abuse your work.

Scotty, do you have any news? :)

The DRM of the Reader is not the point at all. The real issue is what rights, if any, do we have if we change readers. There are many people out there with DRM Palm files and IMP files that cannot use these files on the Reader. (PDF and DRM PDF is still another issue.)

The better question is what are we going to do with the DRM BBeB files if a new and better Reader comes along that (surprise) does not support them. Sure the non-DRM content can be converted (again), as it has been in the past. Are we to keep multiple readers around to read our full collections?

Even Microsoft changed their music DRM system when they launched the Zume. They want us to pay for each item (book, music, film) on each platform. The Reader is just an output device.

I've got a toolchain cobbled together that makes executable files for the Reader.

However, the trick forwarded from igorsky to execute that code is experiencing technical issues. As soon as that gets ironed out we should have some very rapid progress.

That news makes me very happy. :grin:

drm is irritating, if anything, i want to be able to quote passages from the CONNECT software into the clipboard.
http://www.structurise.com/kleptomania/

With this SW you can copy and paste ANY text ANYWHERE on the screen.
It works by making a printscreen and running an OCR on it ;-)

scotty1024
just wonder to know if there is any news?

I'd really like to try writing software for the reader. (with... omg... pages enterable via the keypad! or... prerendering the next page... or using using 1-bit mode for faster turns... or refreshing the page every ten seconds for a crisper display... or...)

How's the toolchain coming along?

Alex, you lost me. How would refreshing a static display of e-ink every ten seconds make for a crisper display?

It kinda fades after a while, but I'd think a refresh every 10 minutes would be sufficient for my eyes. :wink:

It depends at what font size you're reading. With large fonts (ie, regular ebooks), it takes maybe 10-20 minutes for contrast to get annoyingly bad. For teeny tiny fonts (ie, pdfs), it takes less than a minute for the text to become considerably harder to read. Often, this is quicker than the time for me to do the actual reading. Even 10s after a page turn there is a noticeable decline in quality in a situation where you're trying to get all the readability you can.

anyway, a more interesting use would be to quickly page turn using 1-bit mode and then follow up with 2-bit refreshes to regain antialiasing and images. At the extreme, a 1-bit page turn without the negative-positive transition (and with precaching) would be less than 1/4 of a second! I'm drooling already... (the point here being to easily flip through pdf non-fiction. again, regular ebooks are fine as is.)

The only problem would be battery life. Refreshing that often would quickly drain the battery (realiztic 4000 page flips per charge, update every 15 seconds, battery dies in 1000 minutes=12 hours). Also, the "negative image" flashed during reset would be annoying. However, I suspect for a refresh the negative image used to clear the screeen would not be needed. So maybe there would be less effect on battery life by a factor of 2 or 3 (assume little CPU time needed, and display refresh is shorter so the high-voltage power supply is needed for a shorter period.

I just use the joystick to go back and forward 1 page with RTF files. I use fairly large print so unless I leave the reader sitting for 15 minutes contrast is fine.

scotty... your private message inbox is full, i can't reply to you

sigh.. I still can't send you a private message. Hopefully you'll check back on this thread. I'll repost our conversation here (and i'm sure some people are interested in this topic too).



hey. In one thread you stated that you've put together a toolchain for creating programs for the reader. But I think you also said that maybe you got stuck in running them.

Anyway, I'm very interested in creating some of my own programs to run on it and would appreciate it if you'd let me know how that was going or if I could pick up and finish your work somehow.
I have a working tool chain but so far it's a big "So what?"

You can't execute native executables from card and the built in filesystems you can execute from have very limited space left over. :-(

Scotty
So the limit is the size of the executable? What kind of size are we talking about? I'm not trying to compile and load some big package like ghostscript so maybe that isn't a problem. Also, maybe some things (e.g. mp3 decode library) could be deleted to make room?

The bigger question, though, is whether it's possible to manipulate the display on any meaningful or advanced level? (e.g. switch to 1bit mode, access the front buffer, cause a redraw)?