[extropy-chat] FWD (SK) RFC: copy protection report

[Eugen Leitl]

On Thu, Dec 01, 2005 at 02:15:36PM -0800, Adrian Tymes wrote:
> --- Eugen Leitl <eugen at leitl.org> wrote:
> > The system is not going to be officially FIPS 140-1/140-2 certified
> > and is probably not even going to be tamper-responding. However, do 
> > you know many who could launch an attack like several described in
> > http://www.cl.cam.ac.uk/~mgk25/sc99-tamper.pdf (notice that the
> > state of the art in protection has advanced since), given that
> > you only extract *a single key*?
> 
> Actually, I do know quite a handful of people who could and would, if

I call bullshit. Your friends can convince me by booting Linux on the
Xbox360, or extracting the secret from my GSM or banking smartcard.
(All of this can be done with finite effort. I just don't think you know
anyone who can and would care, which is my point -- it's about the threat
model, stupid! I don't care if the Mossad could do it. Can Gretchen Schneider do it?).

> (and this is the kicker) they used that software or worked for someone

Hello, do you copy? I'm talking about physical attacks. Not software.
You can convince me by fpting 60 ml of fuming nitric acid, a beaker, 
a person who can handle 200 k$ worth of equipment, and said equipment.

(If you can do that, you can probably also fax 20 lb of C4 
to the assclown currently playing POTUS, detonator included.
I would be your first paying customer if you could that).

> who did, and somehow did not have the authority to reject the software
> outright because the manufacturer is obviously so worried about their
> own profits that user functionality is given short shrift, so the
> customer would get better value for their time and money elsewhere.

DRM is rampant across the console gaming and premium content industry.
 
> I've seen software where people claimed a single install license was
> worth at least $200K.  Quite a few of them still tried to used copy

The target group of 200 K$ software is very select. They don't form
a percolation network for elite warez.

> protection.  I don't believe I'm allowed to say how many of them (or

If I was in the business of selling 200 k$/license software, I'd sell them
a tamperproof appliance.

> which ones at which clients) were hacked as a matter of course, but I
> can say it wasn't zero.
> 
> The hacker isn't doing it to get the key.  The hacker is testing a
> procedure to get the key.  The hacker then publishes the method.  If
> the hacker's method isn't cheap, other people publish refinements -
> hacks of the hack, if you will - to make it so.

The methods are published already. It starts with "dissolve the epoxy
by gently agitating the package in 60 ml of fuming nitric acid, then...".

Clandestine chemists publish plenty of such methods, too. I've found that
most people are not very good at following instructions (even if they have
200 k$ spare change for a bunch of esoteric equipment, and m4d ski1llZ
to operate such).
 
> The basic architectural standards will remain the same from chip to

Yes, most chips are packaged in epoxy. Most nukes also contain
plutonium. I guess this means most people can build a working nuke
from WalMart supplies.

> chip.  This includes the location of the circuits which encode the
> key.  Simpler methods to obtain the key from similar chips can then be

You're confabulating, again. You don't know where the secret is
on the die. You have to look at the floorplan (I would be very surprised
if you personally could do very much with a 100 MTransistor die floorplan).

> deduced - say, using remote sensing which induces current through the
> packaging, or a specific (undocumented) series of inputs to the chip.

This is not a bullshit session. Please don't invent things which you
wished were true.
 
> You mean this TPM?
> http://www.hackinthebox.org/modules.php?op=modload&name=News&file=article&sid=18613&mode=thread&order=0&thold=0
> 
> (For those who don't want to click: the link goes to a news post
> detailing the latest version a set of of TPM hacking tools, implying
> it's been rather thoroughly defeated.)

Do you read the stuff you post? Are you aware that there is no
MacTel hardware yet but the developer prototypes? That there is no TPM
to defeat?
 
> > The general public a) is not aware what it is buying b) does not
> > oppose DRM because it craves premium content so badly it waives
> > its firstborn in the EULA.
> 
> You mean this type of EULA?
> http://www.theregister.co.uk/2001/11/28/us_court_ruling_nixes_software/
> 
> (For those who don't want to click: it's about a ruling in the 2001

I realize that nobody reads these URLs I post. You sure enough don't
read them.

> case of Adobe vs. Softman that software purchases be treated as sales
> transactions, rather than explicit license agreements.  In other words,
> that shrinkwrap EULAs are completely invalid.  It's only a local court
> ruling, but it seems to be the highest precedent for EULAs so far.  If
> anyone pressed it to a higher court, and the court upheld the precedent
> as many courts often do, EULAs would lose their value even as a threat
> in whatever area the court had jurisdiction over.)

Of course EULAs are not valid in the EU. Will you sue to find that out?
You, personally? How deep are your pockets?
 
> > Can *you* do it? Do you know many people who can?
> 
> If I had sufficient motivation.  (No, proving a point in discussion
> isn't enough, especially if there's doubts that even that would
> honestly convince you.)  I also know people who would do it for enough

I don't know whether you can actually do it. I'm not sure you can.
I'm pretty damn sure 99.99% of people who buy Office can't. 

> money - say, in the tens or hundreds of thousands of dollars - and a
> tightly written contract to prevent you from getting out of it with
> "clarifications of what I meant" after they produce something that will
> install Office on a Windows computer.

Allright, talk is cheap.
 
> Although, frankly, if my motivation was just to get cracked Office
> software, I'd probably get it faster (and with a lot less effort) by
> combing the Web for others who have done it.  Office isn't $200K per
> install - closer to $200 - and my time *is* valuable.  If I simply want

If your time was truly valuable, you wouldn't waste by arguing pointless
stuff on this list just for the point of arguing.

> to use the thing for less money - which is, in the end, the most common
> motivation in these cases - I don't care much about whether I crack it
> myself, except as a means to an end.
> 
> > 2) Have you ever compared two live installations?
> > 3) Have you heard of chaff? Watermarks?
> 
> Yes and yes.  So I duplicate the watermarks too.

The point of chaff is that you don't know precisely where the 
relevant difference is. The point of watermarks is that a leak can
be traced to you.
 
> *cough*  Not to brag too much, but yes, I have been paid to hack
> systems before.  Of course, I'll only cop to completely lawful
> instances, like this one where some people I built a system for
> (writing my own code) lost the administrator password, and hired me to
> hack back in through my own security to fetch it.  They owned the

If you could hack though your own security, you botched the job.

> system; the only copyright violations were with the full knowledge and
> consent of all relevant copyright owners.  Good thing they only wanted
> it to be secure against network intrusions, but were willing to give me
> physical access.
> 
> I also know hackers who are far, far better than I am, and when to turn
> to them for a job.

How do you motivate these hackers for the job?
 
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20051202/525366b2/attachment.bin>