From the README file:

Q: I would like to have a look to FFlash's source code.
A: Sorry, the source code is not and will not be released for security reasons.

I'm sure that it's because of security reasons.

Not.

Well, my HW3Patch readme is a bit more verbose on this issue:

Q: Will you release the source code of HW3Patch?
A: No, sorry. I MIGHT send it to trusted people who give me really good reasons
why this might be a good idea, but generally the answer will be "no". The
reason for this is that the methods for disabling the FlashROM and I/O port
protection are a closely-guarded secret, because that protection is the only
one stopping you from writing a program which makes the calculator
PERMANENTLY UNBOOTABLE.

See, the TI calculators have a single protection keeping you from doing both things TI doesn't want you to do and things which are really evil. While there's hardly any risk from using a well-written program which disables the protection, does what it should and reenables it, the source code could be used to write malicious and destructive programs (and very easily, you just need to send 1 or 2 commands to the FlashROM, that means 1 or 2 instructions).

I think that anyone who wants to write a malicious program could disassemble the program and figure out which 2 instructions disable the protection. People who write stuff like that have that ability.

What I would like is to be able to put an OS receiver in my own operating system so it can receive updates to itself or even put the AMS back on the calculator. I would need the Free software-friendly source code to do that (both technically and because my OS is licensed under the GPL). If you could point me to one, I would appreciate it greatly.

By the way, how does PedroM receive TIB's? Last time I looked at it (more than a year ago), it used a binary-only receiver. But more recently I noticed some discussion about PedroM's licencing on the TI-CT message board, so has its receiver changed since then?

<< What I would like is to be able to put an OS receiver in my own operating system so it can receive updates to itself or even put the AMS back on the calculator >>
FreeFlash is as OS-independent as possible. Just provide the few ROM Calls needed by FreeFlash, and you can use it with you own OS (this can be done on PedroM-Titanium and V200 for example). There is no need integrating a receiver in the OS.

<< I would need the Free software-friendly source code to do that >>
It doesn't exist.

<< By the way, how does PedroM receive TIB's? >>
It uses an embedded version of TIB Receiver (GPL-incompatible), which has been removed from Kevin Kofler's version of PedroM.

PpHd said that Julien allowed him to release a version of TIB Receiver without the hack to unprotect the FlashROM from outside the OS under the GPL as part of PedroM 0.81. He commented the source code and it should be part of the next PedroM. So you'll be able to use that one. Maybe if you ask nicely he'll even send you a copy right now.

Thanks a lot! That's just what I need. I don't really need the hack if it already runs inside the kernel (and the hack won't work in my OS either).

Who did you suggest that I should ask, Patrick or Julien? That wasn't very clear to me.

Ask Patrick Pélissier.

Please wait until I am sure I haven't introduced new bugs in this code (JM hasn't released under GPL all the code but only the receiver part, not the link part) since a bug in this code may corrupt your calculator.

What does TI's legal department think of this?

I mean, the signature protection is there for a reason. IIRC, it's so someone can't write a custom OS that lets you run unsigned commercial FlashApps. So, does anybody else think there's a DMCA risk here?

(Then again, if they haven't gone after PedroM or VTI by now, they probably still won't.)

I think it's there to prevent people from running (or even transferring) apps on a calculator for which the app is not signed. It's a way to keep an app restricted to one calculator.

This program doesn't remove that protection.

But isn't the protection implemented at the OS level, so that a third-party OS could disable it? Or find some workaround (by, say, disabling certain memory protections)?

If not, then why did TI require OSs to be signed in the first place? Simple control-freak-ness?

I guess that's a valid point, but I think TI shouldn't have a problem with the OS signer unless (or until) someone makes their OS capable of running apps, which would probably be (unfortunately) a violation of the DMCA.

That's the problem with the DMCA: it makes otherwise good and legitimate activities illegal.

I think you're right, though. The DMCA is all about control-freak-ness. It gives too much power to those who can't handle it responsibly and who don't deserve it (see Adobe). I know _I_ didn't approve to give up my rights (as a person in the public domain) in the form of the DMCA. Did you?