EFF Position on Trusted Computing

Seth Schoen writes "EFF has just released our analysis of Trusted Computing. We find that the technology could benefit computer security, but must be fixed to ensure that the computer owner is always in control. We also propose a specific way of fixing it. There's coverage of our position at news.com. More articles should be up in the near future at the new EFF Trusted Computing page. Thanks to all the people who helped us understand this technology!"

Comments

In short

[Anonymous Coward]

Don't trust trusted computing. It does not compute.

Re:In short

[Mod Me God]

Computing does compute... but even if it could be trusted... should we actually trust those behind it?

Re:In short

[Anonymous Coward]

I just wish to put it in simple terms as well. To all those who know MS. Why trust someone to be a trustee to your computer when they cant even stop a simple virus? I mean these are same idiots who make patches that dont work. Course dont get me wrong, updating linux aint pretty either, there is that wonderful RPM/dependency hell, but at least they give your control of your computer and most of its source lol.Whats even scarier is all three major chip developers are developing this... @_@. Scary ain't it? ahh maybe ill use one of those other wierd chipset companies that are still stuck around the 400 MHz level lol.Besides, i have an imac with 400 mhz right now and my only complaint is 3-d graphics. other than that, I can hold off buying a the new chips that come out in 2005 for palladium for a very very long time. Hell i still have my old mac lc III motorola 68040 at 25 mhz and a pentium 1 haha.

Re:In short

[cbiltcliffe]

Don't trust trusted computing.

Has anybody thought of this before?
One of the biggest uses of Palladium seems to be DRM. Wouldn't the media companies have to code the file to work specifically with your computer, and no other? Otherwise, what would be the point? You could transfer it, and it would just know it was a DRM system, rather than which one.
Wouldn't this mean then, that a media company that you're downloading a song from would need access to the unique hardware key embedded in your CPU, so that they can code the file to your machine specifically?
If this is true, then they can immediately get personal information and track every single thing you do with your computer, ever.
Suddenly the DMCA and the Patriot Act don't look too bad after all.....frightening thought.

Am I missing something here, or do I have this figured right?

Re:In short

[Anonymous Coward]

It sounds like you haven't heard of public key encryption. Someone can encrypt something for a recipient without knowing the recipient's decryption key.

Re:In short

[pyrrhonist]

Computing does compute...

It computes now! I just solved the halting problem!

Re:In short

[cbiltcliffe]

It sounds like you haven't heard of public key encryption.

I've heard of it, I just didn't realize that's how Palladium was supposed to work. Any of the literature I've read on it just mentions the hardware key that's built onto the motherboard or CPU.
Maybe I've just been reading the wrong articles.....

Bad assumption

[Jason1729]

This seems to be assuming "Trusted Computing" is intended to benefit users.

The real reason it exists is precisely to take control away from the computer owner and give it to the content owner. Given that, what is the point of the EFF proposing "fixes" to help keep the computer owner in control, when its primary design goal is the exact opposite?

Jason
ProfQuotes

Re:Bad assumption

[pla]

Given that, what is the point of the EFF proposing "fixes" to help keep the computer owner in control, when its primary design goal is the exact opposite?

Because it throws the ball back over the fence to those trying to force DRM on us.

In essence, the EFF has given these folks an ultimatum - "You want a trusted computing environment, but not the public backlash? You can fix it like this. Now put up or shut up".

Up to this point, the Palladium group et al could safely ignore most of us, since all of us opposed to DRM have basically just whined about it. Now that someone (and a respectable someone, at that) has offered them a way to get what they claim they want, choosing to ignore that will very tangibly clarify the real intent - If they ignore the EFF's recommendations completely, they all but publically admit they only care about stripping users of the right to use their own machines, rather than creating some fictional "safe" computing environment.

Re:Bad assumption

[wankledot]

The reality is that most of the time, the user is the content owner, having created the content. And for these people, having control over their documents will be a direct result of having created them. They will also have the ability to control the way they are disseminated. This point seems missed in a lot of these discussions. Someone has to create and lock down the files... and that person probably meant to. They had every right to do it, and MS is giving them what they want, control over the access rights to their content. Why is that a problem?

Re:Bad assumption

[Anonymous Coward]

Hence the suggestions for fixing the technology so the user remains in control.

Re:Bad assumption

[Zeinfeld]

This seems to be assuming "Trusted Computing" is intended to benefit users.

I don't necessarily expect there to be end users on the machines I intend to run Palladium on. I want an effective means of hardening a server against compromise.

Slashweeniedom might want to take a look at the people who have worked on Palladium before claiming that Microsoft employees know nothing about security. Butler Lampson won the Turing award for his work on computer security.

There are only two applications I can think of that make sense from an end user application. One is voting, the other is setting up a wallet type application for payments.

I don't think Palladium works for the RIAA DRM case. It has the same weakness as DeCSS, it can be broken through hardware compromise of a single machine.

Re:Bad assumption

[fermion]

The EFF is doing exactly what it should. It is taking business propaganda at face value and then compare the actual product to the propaganda. If the two match, the yea. If not, then either the company is deceiving through it's propaganda or building a deficient product.

In this case trusted computers is being billed as a way to allows owners to control their content. The opportunity for deception is provided by the interpretation of the word 'owners.'

Re:Bad assumption

[Simple-Simmian]

How can they ignore people who will not buy their hardware and OSes? I wouldn't by a trusted computing platform or OS period. I will keep all my old stuff untill it dies and then movie on to something else besided computing if it means giving control over what processes run on my computer to someone else. I had a life before computers and I will have one after them as well. I purchase about 3 complete computer systems and OSes for everyone Joe Sixpack buys and I influence a few otehrs in there purchases as well. I am not going to purchase any "trusted computing" system developed by Microsoft and IBM with endorsements form the RIAA, MPAA and Government. So how can the safely ignore us? They might get the fortune 500 to buy Palladium but getting real people to purchase it will not be easy or I suspect sucessful once they find out what up.

Re:Bad assumption

[red floyd]

This seems to be assuming "Trusted Computing" is intended to benefit users.

I don't necessarily expect there to be end users on the machines I intend to run Palladium on. I want an effective means of hardening a server against compromise.

Yeah, but DRM is being proposed in the hardware and OS for *ALL* computers, not just servers.

Re:Bad assumption

[mentin]

if it means giving control over what processes run on my computer to someone else

It does not. It means being able to prove what processes run on your computer to someone else, if you want this - if you need some services from that someone one. If you can't, that someone else simply would not deal with you, but it would not be able to control what is run on your machine.

EFF proposal is stupidiest I've ever saw (from CNET):

The EFF proposes amending the trusted computing initiative to include a feature called "owner override," which would allow computer owners, whether individuals or companies, to essentially lie to an organization that attempts to ascertain the integrity of their content.

This ability to lie breaks the whole idea - if somebody else does not trust you, he will not deal with you - no EFF will ever force him to.

Re:Bad assumption

[pla]

How can they ignore people who will not buy their hardware and OSes?

Because most people know absolutely nothing about this, and will go out and buy the new much-hyped "Pentium 5 FX Palladium! with patented Ultra TCPA technology! To make your web experience faster over even a 300 baud packet radio modem!".

Those of us who have a clue will avoid this as long as possible, and might even make it a few years without ugrading (hey, my current desktop has lived a few years, and it still runs well), but when even grandma has a 500GHz machine with a terrabyte of 1:1 CPU-synchronous RAM and a petabyte of solid-state disk space, we simply won't have the option of not upgrading our pathetic oversized calculators.

I purchase about 3 complete computer systems and OSes for everyone Joe Sixpack buys

But for every one of us, they have three thousand joe sixpacks to buy into whatever they tell him he wants.

pretty, and I don't like it any more than you, but a geek-only boycott will simply never exert enough market pressure to make a difference.

Re:Bad assumption

[Anonymous Coward]

I have a question: How can "Trusted Computing" protect a small-time content creator like me. More importantly, how can I distribute my content [like a movie, or a sound clip, or even pictures] without it being disallowed?

As far as I can see it, truly powerfully implemented trusted computing and DRM cannot allow anything to play unless it has some sort of stamp of approval on it. Am I to pay for this, and would it have to be vetted to ensure that I wasn't trying to pass off pirated information/music/code/etc as my own?

I'm not sure this benefits content producers, unless they are big enough to play the game.

Re:Bad assumption

[Anonymous Coward]

> This seems to be assuming "Trusted Computing" is intended to benefit users.

This highlights an important issue: there seems to be a tremendous amount of confusion out there about what "Trusted Computing" is.

"Trusted Computing" means that corporations (generally the content owners) can trust your computer. Period.

"Trusted Computing" has absolutely nothing to do with the end-user's trust in their computer. (In fact, it's opposite and backward of that -- "Trusted Computing" implies that the end-user is specifically not to be trusted.)

If anyone uses the term "Trusted Computing" in the future, I recommend that they take the extra time to point out what is being trusted (your computer) and who is doing the trusting (the content owners). This will help reduce the confusion.

Re:Bad assumption

[ShadowDrake]

There are ways to lie and still play the game.

Let's say everyone who doesn't want authorization says "100% accurate copy of SpargleBlaster 9.64 on Baikal DOS 611.82" The person you're working with can either say:

"I won't deal with Spargleblaster users at all" (his loss)
or
"I will take them, but will require different terms" (you may need different account-establishment approach, or they can sell 'anonymous content' at a higher price"
or
"2/3 of our customers are calling themselves Spargleblasters now... perhaps we'd better lay off on the trusted bilge."

Re:Bad assumption

[Alsee]

>if it means giving control over what processes run on my computer to someone else

It does not.

Actually it does when more and more websites and software simply refuse to run at all. It is essentially extortion. You are given a choice to "voluntarily" agree to give up all right to privacy and give up control over your own computer, or you are denied use of your computer.

That computer sitting on your desk is little more than a worthless lump of metal and plastic if you are denied access to most of the internet and you are denied use of virtually all new software.

This ability to lie breaks the whole idea - if somebody else does not trust you, he will not deal with you - no EFF will ever force him to.

Fine, if someone doesn't want to deal with the GERNERAL PUBLIC then they are perfectly free to go hide a hole in the ground. They have absolutely right to expect the GENERAL PUBLIC to be denied ordinary control over their own property.

You are essentially proposing to 'offer' everyone a chance to have a polygraph surgically implanted in their brain. Anyone who doesn't 'voluntarily' agree then gets locked out of all buildings, denied use of the phone, denied use of the roads, denied use of money. To quote you, "if somebody else does not trust you, he will not deal with you". You don't HAVE to vuluntarily have this device implanted in your brain, but if you decline you are effectively thrown in prison. Sure, you're free to walk around your own house, but your house is the prison cell.

Oh, and that "polygraph device" they are implanting in your brain? When you 'voluntarily' use it, it has TOTAL REMOTE CONTROL power. It can force you to do anything, it can prevent you from doing anything, it can erase or modify anything. Of course you are perfectly free to chose to live in a prison cell for the rest of your life instead.

The EFF is simply saying that your computer is your property. They are simply saying that it should not be designed as a weapon against it's owner.

As I have been saying for months, the only problem with TCPA and Palladium/NGSCB is that the design specifications require that the owner of the machine is FORBIDDEN to know his own keys (passwords). The sole purpose for that design requirement is "secure" the computer against it's rightful owner. The owner of the computer has absolutely every right to rip the hardware open and dig those passwords out with a microscope if he feels like it. And once he does that he does have full control over the system and is capable of doing exactly what the EFF proposes. The EFF isn't proposing anything that people don't already have every right and ability to do. They are just saying that there is no reason that people should need a microscope and other equipment to do it.

-

Re:Bad assumption

[WNight]

Palladium and similar "trusted" environments won't help your server at all. It's not like you've got a "Upload and run program - non-malicious only please" page on your server, where all you need is a decent way of checking digital signatures. Attacks happen when there's a hole in software. When bad data (that's all code it, until your machine decides to execute it) causes a program to do something it has the access to do, but is not supposed to be doing.

For example, Outlook has access (intentionally) to your MyDocuments folder, so you can send attachments. Outlook has the ability to send email, and to read from the address book. You should be able to write a script to get outlook to email random documents to random people. This is intentional. The only problem is that a badly formed email can cause Outlook to do this without the owner's intent.

Palladium won't stop this. It's not like people are going to submit their emails to Microsoft for a round of testing and a "Certified safe data" signature to apply. EMail will still come from untrusted sources and will still contain information that breaks tricks outlook into doing something it shouldn't.

For better security, run your services (all of them ideally) in a chroot jail, using remount to expose files to them on a case by case basis. Give them read access only, accept to a few key files, probably logs. This way, even if someone does trick the service into doing bad things, its ability to do anything is severely limited.

And there are ways of keeping Outlook locked in its own directory, unable to read your files, yet able to send them at your command. Have it request a system file-selector dialog. This dialog would be a system tool, not an Outlook window. When a file was selected the contents (not a pointer to it) would be passed to outlook in a manner similar to the clipboard. Even if you got an Outlook worm it would only be able to request this system dialog, not actually open a file and send it.

This, without any hardware protection of cryptographic keys, would be more secure than Palladium (or any other "trusted" architecture). If there was an OS exploit (an ability to fake console input) you'd be at risk, but you would with a trusted architecture too. The difference is that only an OS leak would cause improper application access. A mere application bug (95%+ of the bugs out there) would only cause problems for that appliciation.

But then they will argue

[placeclicker]

That users are ignorant of Computer Security, so it must be controlled by a more intelligent source, like Microsoft. (It's true most are, but does anyone believe MS will fix it?)

Re:But then they will argue

[Geek of Tech]

> But then they will argue That users are ignorant of Computer Security, so it must be controlled by a more intelligent source, like Microsoft. (It's true most are, but does anyone believe MS will fix it?)

Then I will argue that Microsoft knows nothing (or does nothing) about security and I really do think I'd be better holding the controls. Of course, that assumes that I still use Windows.......

trusted computing

[marine_recon]

i dont understand what the big deal over this is. all your doing is handing your computer over to peoples unknow, and they can be trusted to not wreak havoc with your machine. cant they?

Security in Fortune 500 companies

[Anonymous Coward]

I've been working in the security field for about 30 odd years, starting with securing mainframes back at Berkeley in the early 70s and am now providing consulting services to the major financial institutions in the US.

I think that any corporation that invests at least 10% of their budget wisely should be on the track to provide their clients and staff a secure environment in which to deliver their products. I have to deal with a lot of intrusions on a daily basis while overhauling the infrastructure. Currently we've implemented the .NET framework in an insurance company which has permitted them granular control of all security aspects of the deployed .NET applications. This is key, we don't just want to control the desktops but also the software running on them.

Which is nice.

Re:Security in Fortune 500 companies

[Anonymous Coward]

Good idea posting anonymously. Your real-world insights and experiences are not welcome here. Doubly-so because you're praising a product that a large, multination, non-communist software company produces.

Re:Security in Fortune 500 companies

[marine_recon]

keeping things in control is all well and good, but where do you draw the line? next are you going to keep tabs on what is on each persons screen? i dont know about you, but i sometimes might actually feel the urge to check my personal email during the day, and having people look at my personal things with out me ever knowing about it is rather disturbing.

Re:Security in Fortune 500 companies

[Anonymous Coward]

then you had better stop logging on to your sensitive "personal" e-mail accounts at work because you can expect no privacy there

Re:Security in Fortune 500 companies

[Tony-A]

This is key, we don't just want to control the desktops but also the software running on them.

Which is nice.

Good, if you have succeeded, you are safe from Microsoft worms and viruses running unpatched software.
If the systems aren't safe, then you're just fooling yourself and your clients.

Re:Security in Fortune 500 companies

[shaitand]

That is all well and good, but you surely know the masses of people at home are more important than the securtity of any fortune 500 or all of them for that matter.

So long as all DRM applications are knocked out and actively prevented. And there is some sort of guarantee no VENDER can use this technology, that it can only be used at the private level... then it would be a good thing. I could control what software runs on my home systems, but no vendor could generate a key and require ME to verify to THEM.

I really don't see how you'd manage this though,

Fear

[Davak]

In order for a computer to be more secure, it must monitor more aggressively for changes. This seems to be point 4 in the article (remote attestation).

However, by intuition, this would mean that your computer system would know and monitor your system and thus the user more and more.

Misconceptions about this design abound. The most common misconception denies that the trusted computing PCs would really be backwards-compatible or able to run existing software.

Well, crap... of course there is going to be compatibility problems... I am much more concerned that my system and my massaging of that system is going to be tracked and recorded at higher and higher resolution of detail.

Davak

Re:Fear

[Alsee]

In order for a computer to be more secure, it must monitor more aggressively for changes.

The EFF's point is that this is perfectly fine, so long as it is done strictly for the benefit of the owner and that the owner have actual control over it. If would be a good thing if it were a tool for the owner.

The problem is that the current design is not doing this for the owner, and that the owner does not have actual control. The only control the owner has is to kill the system entirely. This will kill much of his sytem and lock him out of his own files. The problem is that it is actualy a tool (weapon) against the owner.

Well, crap... of course there is going to be compatibility problems

Actually there should be absolutely no backwards-compatibility problems unless they screw up somehow. That's why their plan is so insidious, they plan to include this hardware in ALL new computers and you'll never know its there until they slowly start activating it. It will never interefere with old software and no one will ever have a reason not to buy a "Palladium enhanced" computer. The problem is that there is a 100% lockout against any forwards compatibility. Anyone who hasn't bought an "enhanced" computer and "voluntarily" opted into the new system gets locked out of all new software and starts getting locked out of more and more websites. Microsoft has even annouced they want to use it for new e-mail, so you will be locked out of e-mail from your friends and business contacts if you don't "voluntarily" opt in.

There has been serious discussion of a new IPsec protocol - Internet Protocol Secure. With TCPA/Palladium hardware in essentially all systems it becomes obvious to use this cryptographic co-processor to speed up this internet conection. This means you could eventually be locked out of the internet entirely if you don't "voluntarily" opt-in to TCPA/Palladium.

-

Great timing

[placeclicker]

Right on the heels of learning that Outlook Express was mostly responsible for the HL2 Source Code Leak..

EFF's position is outrageous

[Anonymous Coward]

The EFF basically wants your computer to lie to a content provider so that you can turn off the security and still receive their content. It might as well not exist in the 1st place then, which is probably their real goal.

Re:EFF's position is outrageous

[tkrotchko]

Well, its my computer, and I can damned well modify it to my hearts content. This seems more than reasonable; it seems *necessary*.

Re:EFF's position is outrageous

[Highrollr]

Having my computer do what I want it to doesn't seem particularly outrageous to me.

Re:EFF's position is outrageous

[Anonymous Coward]

Fine, but part of the point is to provide secure content. The user has shown that he can't be trusted, so the content providers have a useless security system. A computer is a tool. It's pretty stupid to say you have a god-given right to see every bit in every memory location. You're just limiting the tool's usefulness for secure content, which is the goal of the EFF. So don't be surprised and indignant, when people refuse to sell you content you want in a form you want, and sue you afterwards when you claim yuo stole it because you couldn't get it the way you wanted.

Re:EFF's position is outrageous

[bunnie]

To wit, Seth points out in the article that Trusted Computing still allows you to do with your computer what you want; there is nothing that prevents you from just turning off the trust mode outright and running it like a normal PC.

It is a misconception that the trusted PC hardware will be unable to run your custom code. I have been guilty of propagating this misconception as well.

More significantly, the trusted PC now enables a set of security primitives with implicit security policies that you can choose (or choose not) to subscribe to. By subscribing to these policies, you are choosing to risk being non-interoperable with software that operates outside of the prescribed policies. Consumers will decide on the benefits of this risk.

Unfortunately, corporations form a large "voting block" of consumers, and they are likely to receive the Trusted PC quite well. Perhaps the question then is can a company *force* employees to also subscribe to security policies that limit the employee's ability to interoperate with the employee's favorite set of personal tools and software; i.e., will you be allowed to run cygwin and use perl scripts to post-process "trusted" excel spreadsheets, csv files, databases, etc.? Now, your numbskull manager can set arbitrarily strict security policies on secure-domain documents...this will be a fun battle. :P

Re:EFF's position is outrageous

[Nucleon500]

Fine, but part of the point is to provide secure content. The user has shown that he can't be trusted, so the content providers have a useless security system. A computer is a tool. It's pretty stupid to say you have a god-given right to see every bit in every memory location.

Content providers do not have a God-given right to complete control of their content. They have a government given right to partial control. As history has shown, content providers are completely against fair use. (I'm talking about ripping, transcoding, timeshifting, and playing in the car, not P2P copying (which DRM doesn't stop anyway).) Putting content providers in a completely controlling position is not a good idea.

Why am I opposed to DRM? For one, why should I buy something from someone who explicitly states they don't trust me? More importantly, while I personally can choose to abstain from DRM, the vast majority of computer users are uninformed. When they choose DRM, they're choosing "music" over "no music," not "DRM music" over "unencumbered music." This is a result of the lack of competition in the content industry - because DRM lowers content's value, no single provider would sell it, because nobody'd buy it. Unfortunately, the content industry is dominated by a cartel, so DRM is being pushed down the throats of uninformed users who haven't been shown an alternative.

Re:EFF's position is outrageous

[Anonymous Coward]

So don't be surprised and indignant, when people refuse to sell you content you want in a form you want

Those people are such a tiny minority, that they'll never be noticed. I bet they're outnumbered by over a million to one.

Re:EFF's position is outrageous

[prichardson]

How about this, since I can't control my computer, why should I have to pay for it. I would be much less opposed to not controlling it if I didn't own the hardware. Perhaps Microsoft will start liscensing computers as well.

Re:EFF's position is outrageous

[Anonymous Coward]

I agree. The EFF has to understand that their box at the end of the article COMPLETELY ignores the part of the point of TC. Part of the point is to reduce software piracy. Like it or not, every time you copy windows/office/whatever and give it to your buddy, you are committing software piracy. Part of the point of TC is to prevent this - mainly by preventing the user from circumventing product activation. This is a worthwhile goal. The A+OE position has all the cons of the current status quo! How is this an improvement?

Re:EFF's position is outrageous

[westlake]

why should I buy something from someone who explicitly states they don't trust me?

It has become routine in this rural village for clerks to ask folks they've known all their lives to show a bank card and personal ID for any purchase over $50. Cash sales are the exception and no one accepts $100 bills. Welcome to the 21st Century.

Re:EFF's position is outrageous

[Anonymous Coward]

No, the point is that it is a general purpose, programmable computer, and I, as the owner, should be able to control how it works, program it, modify its behavior etc.

DRM style restrictions are totally unnatural and artificial for a programmable computer, and require that the computer not function as one.

By my definition, that wouldn't even be a computer. I would never buy such a thing, that isn't what I want.

The general public doesn't understand what a computer is, some of them may be happy to get some sort of restricted appliance instead. If they want that, fine, but calling it a computer and trying to migrate everyone onto it is deceptive.

I don't think EFF expects any changes in what TC will be like, they just want to expose what it is really about - preventing the user from using their computer in ways that computers are fundamentally meant to be used, while claiming that it is for the benefit of the users.

Re:EFF's position is outrageous

[Alsee]

but part of the point is to provide secure content.

Those advocating TCPA and Palladium are doing everything they can to deny that very fact. Forcing them to admit it is actually designed as a DRM system is a good step in defeating them.

It's pretty stupid to say you have a god-given right to see every bit in every memory location.

I'm not about to start on "god-given rights", but the fact is that my computer is MY PROPERTY and I have absolutely every LEGAL right to disect my computer and analise every single bit in every single memory location if I feel like it. The TCPA inititive just tries to make that really inconvient to do.

You're just limiting the tool's usefulness for secure content

Fine. "Secure content" is just another term for a crippled product.

don't be surprised and indignant, when people refuse to sell you content you want in a form you want

Fine, people are perfectly free not to sell their product. But they obviously aren't going to make a single cent by refusing to sell it. When someone offers the public the product they want then the public will buy it. The RIAA could have made a good profit had they started doing so several years ago. All they had to do was sell downloads for their full selection of music in NON-secure format with and easy and featurefull service at reasonable prices. Not only would they have made huge sales in the face of P2P, but P2P wouldn't have exploded on the scene like it did. The Music labels simply chose not to offer to sell music downloads. They refused to serve the market and created a vacuum, and P2P sprung up to fill that vacuum.

Capitalism and free markets are not based on FORCING people into a product they don't want. It is about what the customer wants, it is about companies serving the customer's desire. Customers do not want a crippled "secure" product. They want fully functional and non-secure MP3's. A company is free to offer or not offer anything they like, but the company that offers what the public wants is the company they gets the customer's money.

-

Re:EFF's position is outrageous

[shaitand]

The ONLY LEGITIMATE use of trusted computing is for network administrators to control what applications their users can install and run.

The other uses like DRM are not legitimate and your right they'd be more or less useless with the EFF's changes. So what's your point?

Not with the current government...

[dpilot]

Not just Executive, but Legislative, as well.

Our government responds to campaign finance, and the lion's share of that is done by large corporations and other aggregates that want to make sure that THEIR rights come first.

Most people don't understand enough about computers to understand how completely OUR rights in this realm have been trampled, already.

Re:Not with the current government...

[Negativeions101]

Corporations like Microsoft have every right to put DRM technology in Windows... it's a free country, after all.

It's up to users not to buy into the technology... Don't expect a government who is too busy looking for weapons of mass destruction to care about your computer.

Re:Not with the current government...

[dpilot]

In the theoretical purity of a free and open marketplace where informed consumers had a choice,
I would agree with you.

OTOH, in a marketplace where there is every attempt to shove products down our throats and make us pay for things we didn't necessarily intend to buy, deny us the ability to choose, and misinform,
I fear I must disagree.

I wish I could agree with you more.

I wish I didn't feel our legislature (and executive branch) was so 0wn3d.

Re:Not with the current government...

[Anonymous Coward]

quite so. i wouldn't want to force ms to keep drm out of their os any more than i'd want to be forced to use it.

Trusted Computing = IBM,INTEL,MICROsoft own you.

[zymano]

The will search your hardrive for any copyrighted material.

Next Tv's,Stereos

Doesn't that...

[chill]

...defeat the purpose? I mean, everyone knows that end users can't be trusted. Given the chance, they'll do nothing but pirate movies, music, television and software, etc.

*** END SARCASM ***

I think DRM is a *good* thing. Once people have to pay for music, movies, etc. the industry will realize exactly what they were losing to piracy -- almost nothing. If someone could wave a magic wand and people had to abide 100% by the rediculous license agreements, you'd find that instead of buying what they were sharing, they would go without.

Or does Microsoft, the BSA, MPAA and RIAA really think all those people in Asia are going to pay a few months worth of wages for software or entertainment?

Re:Doesn't that...

[Delron Da Thugg]

Yeah but if those dirty birdies get it from the companies for nearly nothing legally, so should we. Just because we can afford to pay more for goods and services, doesn't mean we should.

Re:Doesn't that...

[Anonymous Coward]

"..defeat the purpose? I mean, everyone knows that end users can't be trusted. Given the chance, they'll do nothing but pirate movies, music, television and software, etc. *** END SARCASM ***"

Do you understand what sarcasm means? What you describe currently occurs every day by millions of people. The end user can't be trusted. If you believe nothing is being lost because of piracy, you are a liar or a moron. Just look a concert tix sales vs. cd sale over the last few years. cd sale down, tix sale up.

Re:Doesn't that...

[Anonymous Coward]

Once people have to pay for music

Every currently proposed DRM scheme can be defeated by plugging an audio cable from the speaker jack on computer A into the line in on computer B.

The only way DRM could conceivably prevent music piracy is if it included a beefy guy who would watch over your shoulder while you used your computer and kick you in the nads when you tried to copy something... and even that would probably be defeated occasionally by someone who got the nads-kicker drunk or wore a cup.

Re:Doesn't that...

[Anonymous Coward]

cd sale down, tix sale up.
Sound like something good is happening in the industry for a change.

Re:Doesn't that...

[chill]

CD sales aren't down because of piracy -- they're down for several reasons.

1. Disks cost too damn much. Why can I buy a movie for $19.95, including the soundtrack on the disk -- and if I want the soundtrack CD separately it is $15? Movie production costs oodles more than music production -- tens of millions of $$ more, yet the music industry tries to sell CDs for only a fraction less.

2. Production costs on music CDs are down. Pros can and do produce CDs using laptops and home studios for a small fraction of the costs of the old rent-a-studio method. Have music costs to the consumer dropped proportionally? Has the profit given to the artists risen proportionally? (No, in both cases.)

3. A few years ago most baby boomers finished converting their old album, 8-track, cassette library to CDs. It was an artificial drive in sales -- buying the same thing on a different format. The last few years has reflected the loss of those sales. Unless the industry can convince us to all buy the stuff again, this time on DVD-Audio or something, that mechanism is gone.

4. ...Gone because the new format is digital music files, and the public no longer needs the industry to do the conversion for them. When cassettes came out, there was no easy or decent way to convert your albums. When CDs arrived, they weren't recordable at ALL and if you wanted your music, you had to fork over $$ again. NOW, the industry can kiss my ass -- I converted all my CDs to .flac files for archive purposes and MP3s for listening on the portable unit. I DON'T NEED THEM FOR THIS ANYMORE.

And the "sarcasm" part was meant for "...do nothing but..." part. People have jobs, school, and lives beyond the fucking entertainment industry -- most people's lives don't revolve around music, movies and downloading MP3s.

Re:Doesn't that...

[Anonymous Coward]

Every currently proposed DRM scheme can be defeated by plugging an audio cable from the speaker jack on computer A into the line in on computer B.

that is, assuming the audio output from A can't be automatically muted and is free of encoding that can be read by B.

Re:Doesn't that...

[Alsee]

Every currently proposed DRM scheme can be defeated by plugging an audio cable from the speaker jack on computer A into the line in on computer B.

You underestimate the stupidity of our opponents. They have in fact not only proposed such a system, they have had congressmen advocating it.

And how could they conceivably accomplish this impossible goal? Simple, they want to make it illegal to make or buy an ordinary recording device without a "Fritz chip" inside that would shut down the device when it detected specially tagged sound. They even proposed requiring that every single analog to digital converter have such lock-out technology embedded.

You could be dictating a letter into an ordinary tape recorder, and if someone walked by on the other side of the street with a radio the "Fritz chip" would pick up the special tag in the music and the tape recorder would record dead silence until they walked out of range. You only discover later that there is a five minute dead zone in the middle of your recorded dictation. Your camcorder tape of your child's first birthday goes dead silent whenever it detects tagged music in the bacground, and the video goes dead black whenever it detects a tagged TV image anywhere in the background.

Reporters might be able to get a special licence for a special video camera that doesn't go dead in this manner, but it would probably have to embed a special tracking code in everything it records.

I'm fairly certain that this proposal is far too extreme to ever get approved, but there ARE people demanding it.

-

Re:Doesn't that...

[Anonymous Coward]

I have some friends who don't own a CD player.
Have never purchased a CD, DVD or VHS tape. It's not a priority. Eat, Drink and be merry.

Re:Doesn't that...

[Anonymous Coward]

I have said this over and over again for many years now and have instituted patents on it based on postings many years ago.
All your base belong to me.
If you encrypt any audio, visual, media output digitaly so any input device recieving the encrypted code has to be able to descrambles it in order to play the media, I own you.
This includes your SPEAKERS, Monitors and TV's.
You can't hear or see it unless you have purchased it.
I have also heavily invested in companies that make digital speakers, Monitors and TV's. That's the last line of defense, wouldn't you agree.
All thing's being equal I've not found any digital media yet, I can't rip one way or another.
I suggest we all purchase a musical instrument and have a hoe down. Refuse to purchase second or third party entertainment. We don't need it, and never have needed it to live and be happy.
In fact a good argument could be made that all the glammer and glitz is the cause of much unhappiness.

Re:Doesn't that...

[WNight]

It's not only extreme, it's stupid. Both because of the implications of losing your content when someone whistles happy birthday that you mention, but because it's flawed.

For one, A/D converters are literally selling for pennies in quantity. Adding public-key decryption and watermark scanning in realtime would mean we'd need the equivalent of a pentium in every A/D converter. Now the price has gone from pennies to tens of dollars, at a minimum. This would make nearly every electronic device that consumers buy more expensive by a factor of ten, I'd guess. (After all, even A/D converters that aren't being used for audio recording could be used that way if you disassembled the device and took the components.) Trying to legislate this would never work once consumers realized that a cordless phone would cost five hundred dollars, that a microwave over would cost over a grand, etc. Worse, they'd never get as cheap as they are now.

But, finally and most importantly, there are billions of A/D converters in existance without Fritz chips. Simple ones used to read thermometers in a fridge and complex ones with the quality to record 24/96 audio. (Not that a lot of lesser ones could be made to produce a higher quality signal, with the right processing and quite a few passes.) These aren't going to disappear and even if they're declared legal it'll just mean that only criminals have them.

If even one unsecured A/D converter exists, or can be built by a electronic grad student, no content is secure from this. You can copy it, strip the watermark, and pass the unsecured data around the world for even people with Fritz chips to enjoy.

The only thing for them to do, if they want to stop this, is simply charge everyone a content tax. They assume we're all pirates and bill us accordingly at tax time. Except some retribution though if they try it.

"[T]he computer owner is always in control."

[HTH NE1]

We find that the technology could benefit computer security, but must be fixed to ensure that the computer owner is always in control.

That's simple enough to solve. The computer will just be both owned and "0wnz0red" by someone else, most likely by the entity that licensed the operating system to the user, and the hardware imprinted for that specific operating system and all others irrevocably locked out.

And it will all be done with the click on a seemingly innocuous little virtual button that reads simply, "I Agree".

Trust.

[Simple-Simmian]

The EFF is correct as usual. Trusted computing = Me knowing what the hell is running on my computer and having control over it. Anything else is untrustworthy computing. Anyone that wants to control what I can do with my own property (computer) can stuff it where the sun don't shine.

Re:Trust.

[Chazmati]

They may be correct, but they need to work on their presentation a little. I'm afraid the general population (the non-slashdot crowd) is going to read their position and disagree.

Companies (through this "Remote attestation" feature) checking to see if their software has been modified? What's wrong with that, my mom will ask? They own it. The subtleties of how this can run counter to users' interests will escape many.

And what's the EFF's solution? To LIE to these companies. That sounds wrong, doesn't it? To lie? No one should lie, my mom says. Or your mom. Whatever.

Re:Trust.

[Wesley Felter]

Yeah, lying about remote attestation seems like a convoluted solution. How about just not having it?

Companies can still lock down their computers by disabling boot from removable media and setting a BIOS password.

Re:Trust.

[ThaReetLad]

Repeat after me. You have NO right to modify copyrighted material.

Re:Trust.

[shaitand]

Repeat after me. THEY have NO right to check and see if I have.

That's what police are for, when they have sufficient evidence to get a judge to sign a warrant, THEN they may come and examine my computer as provided for explicitly in the constitution of the united states of america to prevent EXACTLY this.

I've said it once and I'll say it again, the only LEGITIMATE use for the individual, company, gov entity, etc to keep unauthorized software from running on THEIR OWN computers. Is this good stuff for a fortune 500 to keep their employees from installing winamp and kazaa... absolutely.

Should any 3rd party selling something to me as a member of the public in any way whatsoever be able to utilize this technology? NO. When I buy a piece of software, book, mp3, cd, etc, it should be completely free of this, and then left to me if I want to require signing within my organization to ensure users are only using what I've ok'd, or to keep my kids from installing garbage and spyware.

Re:Trust.

[WNight]

Wrong. You have no right to create duplicate copies of a copyrighted work. But copies implicit in its use, such as modifying it for your enjoyment or playing it, are explicitly allowed under US copyright and most other copyright acts around the world.

You can buy a DVD and legally modify it however you wish, just as you can write in the margins of a book and black out the bad words. You may violate the DMCA doing this, but that's not a copyright issue (you're allowed to do it, the DMCA just forbids certain ways of doing it and certain motivations) and the DMCA is likely going the way of the UCITA. The DMCA is being used as a blanket law to squash too many legitimate pursuits, eventually states are going to write laws that as good as render their citizens immune to prosecution under it and eventually it'll get overturned. This is in process for the UCITA already.

Trusted ... or Trustworthy?

[Anonymous Coward]

Personally, I still prefer "Trustworthy Computing" over "Trusted Computing."

It's a game -- flush out the rats of hidden agenda

[Morgaine]

The point of the EFF doing this is precisely to underline the fact that big business is attempting to take control of the end-user computing platform away from the user.

You see, the problem is not so much that big business is doing this, but that it is doing so by subterfuge rather than out in the open.

The EFF is just flushing out the rats here. If business were trying to take control of people's property openly then the EFF wouldn't need to put on an act of innocence and merely be "identifying dangers" as the proposed solutions as if business wasn't aware of them.

It's a good strategy. Big business can only respond by saying either "Oh yeah, we hadn't realized" (LOL), or else it can reply that this was indeed the intention. In both cases, the user wins.

My bet though is that the EFF will be met by total silence.

Token Based Trusted Computing

[Buzz_Litebeer]

I used to intern for a company that works with a product called "WIBU KEY" now despite the fact their is a single note of one group "getting around" WIBU key, done properly WIBU should not be feasible to break.

Now WIBU is making something called "Codemeter" in which a user will be able to have licence information for hundreds of different software packages, that means if someone has MS word on their computer, and knows someone else that has it, they can use their licence on the Codemeter stick on their friends computer.

They are USB devices that can be carried on the keychain, and places like www.securikey.com are going to start using the new codemeter product.

It is MUCH better than other trusted computing schemes, since the data is not just being hanleded on the computer, all codes are "private" and all licence data is "private" on the key itself.

The way the part is manufactured you cannot get the data off of the key, if it is stored using a certain key mode. Since the key has its own integrated chip, it can use its key to decrypt as a private key that you need never know. If you lose your key or break it, you can go on-line, register a new key that you have bought, and get the data transferred to your new codemeter stick.

I was excited when I got to look at the product in its pre-development stages, and only wish that I had been able to stay on with the company to do more work with the new product line.

I think token based authentication is the way to go for the future, simply make programs that will not run unless large chunks of them have been decoded, and make sure small but important algos in the programs have to be run through the key every so often.

And as you detail attacks (if you reply) please note, spoofing the key doesnt work when the code runs inside the key, and if you use random checking algos against the key, it wont work either.

Any WIBU key hack on the net (which i researched) is totally based on bad programming by the target company, IE they sent "999999" to the key, and expected "99999" back, which was just bad programming on thier part.

Buzz OUT

Re:Token Based Trusted Computing

[Cid Highwind]

This token-that-decrypts-a-key idea sounds a lot like the directv/dish network smartcard systems. Those have been cracked repeatedly (usually within hours of an update). You can't give bits to the users and expect them to remain secret.

Re:Token Based Trusted Computing

[Buzz_Litebeer]

actually, you assume the program can "run" without the key, and that it merely "checks" the key, thats not actually a very well designed program.

Using the WIBU key you can actually force a program to encrypt parts of itself, and it wont decrypt without the key. Yes you can make a JUMP statement around the encrypted and compiled code, but if the programmer has done it properly, the code wont run.

You can, for example, encrypt several KEY subroutines with the key, that requires the key to run, the key then loads these into memory via its specialized ASIC.

The only way to actually defeat a properly programmed bit of code would be to take a snapshot of it in ram and then run that snapshot.

And even then , if the code user made a key call within the encrypted secure block , i think it might even be impossible.

What you are thinking of is some examples, and a single example, on the net of the WIBU key being broken in a peice of software. Having looked at how that was done, i can tell you the person that programmed the WIBU code did NOT read the programmers guide for the WIBU key.

He basically said "hey WIBU key here is 999999, tell me what you think"

and then checked to see if the WIBU key replies "9999999"

the problem is, that is NOt what you are supposed to do. You are supposed to Encrypt all start up code using the WIBU asic, and within that encrypted data make a call to WIBU key to verify its the same key that decrypted it.

Since the only time you have "running non encrypted data" in the system is in the ram after the program starts, making a JUMP command around the code wont help, since it wont have loaded the proper data required by the WIBU key.

I mean, about the only other "exploit" i have seen is someone claiming to have gotten around "securikey" by logging in as administrator and uninstalling its drivers. When the trick is, they have to have a key and a password for them to be able to log into administrator.

Both are examples of groups not reading the manual, and not doing things properly.

1 was someone who threw a hack job of coding into the mix, the other was someone who thought that since they were the administrator, and knew the password and assumed that they would let their children log in as administrators and give them passwords, that that was an exploit.

It would be like making a safe and then keeping it unlocked and open at all times, then sueing the safe company because someone stole stuff from it.

Re:Token Based Trusted Computing

[ymgve]

How about taking a snapshot of the RAM after the WIBU dongle has decrypted the program, then overriding all other dongle calls in the now unencrypted program?

Everything is crackable.

Re:Token Based Trusted Computing

[Anonymous Coward]

I assumed good authorization meant having the fastest computer (alternating new numbers before a mathematical probability can exist) and having the ability to "fit" host & client keys through a reliable process.

Re:Token Based Trusted Computing

[Cid Highwind]

I still think the WIBU key looks a lot like a satellite TV smartcard. The reciever doesn't just check that the card it there, the card has to decrypt a symmetric program key with a secret key stored on the card, then pass the decrypted program key back to the reciever.

Besides taking a logic analyzer to the token (which some bored EE will do eventually), the obvious way to attack this system would be a series of RAM dumps, as ymgve said. If the decrypted code is ever on the user's PC, it can be captured.

I want a secure computer

[kfg]

Not a "trusted" one.

Just as I wish with my house. I want my house to protect me, my papers, possessions and privacy. I want it to be nobody's business what my house contains, even to the point of being able to protect myself against legitimate legal prossecution.

Oddly enough, that's what the Constitution was written to provide my house with.

It is up to me to secure my house with whatever technological measures are available to provide that security and understand how to use that technology. I'm perfectly willing to take the same responsibility for the security of my computer. Just provide me with the tools. Then go the hell away and leave me alone.

The second my house starts deciding for me what I may or may not keep in it or do inside it I get a new house.

The day my computer decides it doesn't "trust" me with what I'm storing in it or doing with it I pull the plug.

Fortunatly for me there are already hundreds of millions of "untrusted" computers already out there in the wild that do everything I might require my computer to do.

KFG

Re:I want a secure computer

[dukoids]

You object to have gouvernment controlled cameras in your house to protect you?!? Do you have anything to hide?

--
In corporate America, your computer controls you

Re:I want a secure computer

[kfg]

Do you have anything to hide?

None of your business. :)

KFG

Re:I want a secure computer

[SiliconEntity]

Just as I wish with my house. I want my house to protect me, my papers, possessions and privacy. I want it to be nobody's business what my house contains, even to the point of being able to protect myself against legitimate legal prossecution.

I don't think you understand Trusted Computing. I suggest that you read the linked article.

TC will not allow anyone else to look into your computer and see what software you are running, without your permission. What it does is to allow you to SHOW other people what software you are running. You can't do that today. You can't convincingly show me that you're running Mozilla or Internet Explorer or Opera or any other specific browser. Trusted Computing technology would allow you to make that kind of demonstration.

In your house analogy, it's as though today, your house was inpregnable, and you couldn't even show anyone else what you had. No one else could see in, even if you wanted them to. It's like a house with no windows.

TC would give you the power to let people see into your house, so they could see for themselves what you had. TC is like a window you can open or close. It's a window you can open into your computer that lets you show people what software is running.

In short, TC does not take away your privacy or invade your computer. It gives you the power to reveal information about your computer, and the EFF is afraid that you'll be forced to do so in exchange for being offered services on the net. But the control remains in your hands.

Re:I want a secure computer

[kfg]

You can't convincingly show me that you're running Mozilla or Internet Explorer or Opera. . .

Thank God. I'll do everything I can to keep it that way.

And if you feel the need to be "convinced" of what's in my house, get a warrant. I have no such need of demonstration.

My house has doors and windows. My computer has ports and file ownership. I can open them. I can close them. I can let people in. I can throw people out.

I don't have to "prove" a bloody thing. To anybody. Even if you have a warrant.

KFG

Re:I want a secure computer

[SiliconEntity]

You can't convincingly show me that you're running Mozilla or Internet Explorer or Opera. . .

Thank God. I'll do everything I can to keep it that way.
And if you feel the need to be "convinced" of what's in my house, get a warrant. I have no such need of demonstration.
My house has doors and windows. My computer has ports and file ownership. I can open them. I can close them. I can let people in. I can throw people out.
I don't have to "prove" a bloody thing. To anybody. Even if you have a warrant.

The point is, you don't HAVE TO prove anything with TC. It allows you to do it, but it doesn't force you to.

Without TC, you don't even have the choice. If you wanted to prove to someone that you were running a certain program, you can't do it today. TC gives you that choice.

It has nothing to do with warrants or being forced to prove anything. At most you may be requested to prove something in return for being offered something that you value.

Your house has doors and windows that you can open and close to let people see in. But with your computer, you don't have the option to have that kind of transparency. You can't "open a window" into your computer and show people what software you are running. It's like a house with no windows or doors.

Again, it's not a matter of being forced, coerced, or given a warrant. It's just a matter of being able to choose to open up your system and reveal its configuration, which you can't do today.

Re:I want a secure computer

[kfg]

What I have no capability of doing I cannot be forced to do either. Even with a warrant.

It has everything to do with warrants.

I can let you look in my house window, but that doesn't mean you can see in my file cabinet or dresser drawers. You cannot be sure of what is in those dresser drawers without coming into my house and looking into them. Even then you cannot be sure I didn't remove what you were looking for before you came in or falsely placed something there which is not mine.

If I wish to prove to you I have a certain book I can remove it from my house. You are still faced with having to believe it is actually mine.

I can show you any file on my computer. I can give you root. You may examine the complete configuration of my system, if I chose it, already. I can run a webserver and offer up anything I chose to show you.

Trusted Computing offer me nothing but potential intrusion. The ability to "prove" I own my own socks. I neither need nor desire that capability.

The capability creates the charge.

I understand it perfectly. Stay out of my dresser unless I place it on the curb.

KFG

Re:I want a secure computer

[Alsee]

At most you may be requested to prove something in return for being offered something that you value.

And when that "something that you value" expands to include virtually every new peice of software and even basic access to the internet? Yeah, yeah, I'm skipping to the end. I'm discussed this exact expansion process at length elsewhere.

You say it's like giving people the ability to put a window on their house. But it's more like imprisoning them in their house unless they "voluntarily" live in a glass house.

A computer is useless if they deny you the ability to run any new software or the ability to communicate with anyone else. It becomes extortion. Either chose to give up your right to own your own property and give up your right to privacy or get locked out of everything.

------

If you have a TCPA computer it is your physical property. You have every right to disect the TCPA chip and read your own key. And with that key you can paint any image on your window that you like, just like the EFF is suggesting. By letting someone look in this supposed window the only thing that you "prove" is that perhaps you made the effort to dig out the key that belongs to you in that window that belongs to you.

In other words the system proves nothing unless you want to claim that I do not actually own my computer and I don't have a right to analize my own property to read my own key.

-

Re:I want a secure computer

[ThaReetLad]

Do you have an electricity meter in your house? What about a gas meter or water etc? You cannot modify those without facing criminal charges, or having your services cut off. The same is true of a music client or a video on demand client. What about a cable or satelite TV decoder? These decrypt signals in your home, but if you try to modify them you could face fraud and DMCA charges.

Re:I want a secure computer

[WNight]

If you wish to use a house analogy, it's like having a house whose walls are completely transparent. You've got a tarp over it most of the time, but if someone asks you a question about your house, even if you don't wish them to know, you are unable to lie because they demand to look for themselves. Further, because it's one of these transparent houses (a computer - not granted the same constitutional protection as a primary residence) the police can lift the tarp and peer in. Even worse, it's impossible for you to stop this because while you have a "tell me who wants to look" feature, it can be overriden for you by the company that built your house.

However, I think it's probably best to ditch the analogy. It's a computer that anyone can query and (supposedly) get a unfakable answer about what you're running. Now instead of being able to fake your user agent string to view a site that merely "Works best in IE", you'll be unable to use it at all. Great advancement in my ability to use my computer. You'll also have DVDs that don't contain the decryption keys, they'll phone home and ask for a key, provided you aren't running anything that might let you take a screenshot, or evil hacker tools like a debugger (which is always running on my computer because I'm frequently doing driver development work for the company I work at), and if the player is one of a select few that won't let you skip advertisments or trailers. It'll also be a computer that'll secretly respond to the authorities, or whoever you OS vendor thinks are the authorities, with any information about your computer... But only if they have a warrant of course.

Yeah, that's an exciting new world.

Re:I want a secure computer

[yakovlev]

Undeniable proof is VERY bad in most cases. It's an important aspect of society that we have things that are inherently unprovable.

The prototypical example is voting. It's an important part of the voting system that you cannot prove to someone else that you voted for a certain candidate. This prevents people from being able to buy elections.

It's simply not enough to only have the OPTION not to prove something, you must have the INABILITY to prove it. If some unscrupulous person says "I'll torture and kill your family if you don't show me proof that you voted for my candidate," then you really don't have the option not to prove it to them, do you? The only way to make sure you have an option not to prove something is to have an inability to prove it.

While in most cases refusing TCPA attestation won't cause someone to kill your family, saying "At most you may be requested to prove something in return for being offered something that you value," fails to realize the value of data accessable by and stored on some computers. While the average home user can usually weather the loss of all data stored on their computer (as witness the lack of backup procedures for most home computers,) most banks can't. If their data is sufficiently locked inside a TCPA computer that only one software program can access it, then they have NO CHOICE but to continue using that program. We can argue about whether or not they should have chosen that program in the first place, but that's beside the point once they're locked in.

Furthermore, even for the home user there may in the future be sufficiently valuable information online and locked behind TCPA attestation requirements that it makes the user feel like they don't have a choice. It's hard to know what kinds of things will be made available online in the future, or the software restrictions that will be placed on those. Some people would rather not let the genie out of the bottle, as it were.

Since adding remote attestation creates an architecture that allows or even encourages an awful lot of remote software restrictions and vendor lock-in, the case needs to be made that the benefits outweigh the disadvantages. The EFF document does a pretty good job of enumerating the trade-offs. While it clearly has a slant towards preferring not to have remote attestation, I tend to agree with their reasoning.

Unfortunately, the general public will just use whatever Microsoft force-feeds them by default unless they're scared silly like they were around the Pentium 3 Chip ID, and even then they would have just taken it if Intel hadn't removed the feature. Most people don't know and don't care about the implications of hardware like this, so their choice is being taken away without their knowledge whatsoever.

Fixing it

[pheared]

"We also propose a specific way of fixing it"

Namely, removing it.

Microsoft may be changing course

[SnowWolf2003]

This article over at cnet looks like Microsoft may actually be listening to the critics of trusted computing and rather moving towards what it calls Shield Technology - basically incorporating better firewall technology into the operating system. I for one would welcome this over trusted computing.

Re:Microsoft may be changing course

[cmowire]

The question is weather they are listening or just using the thin edge of a wedge. Shield Technology looks to be a parallel path from a different bunch of researchers because the security problem is bad and they may not have enough time to stall user uproar before they can get everybody on the trusted computing bandwagon.

Although, to be fair, the more they attack some of the applications they claim are problems that TC is supposed to solve, the fewer remaining reasons other than DRM can be given.

Trusted Computing is dangerous, even with user permissioning because it takes Microsoft+Friends 90% to the point where they can roll out what they want. First release, EFF's way, next release, take away that feature and we're stuck. I have high hopes that if they ever were to create a DRM regeme that didn't leave adequate holes, the user uproar would manage to stifle it.

Re:Microsoft may be changing course

[Wesley Felter]

They look like separate issues to me. Trusted computing provides lock-in, DRM, secure data, etc., but it doesn't protect you from viruses. "Shield technology" may help protect against that stuff. I'm sure MS is not dropping trusted computing.

Re:Microsoft may be changing course

[Davoid]

While they haven't specified what "Shield Technology " is yet... it seems to be nothing more than what most companies and people are doing already... attempting to filter attacks at the firewall. IOW nothing really new.

As they say in the article you linked a major part of the problem is that people just do not upgrade fast enough. It is debatable whether they can ever patch their systems fast enough. Mainly because the vast majority of virus definitions are after the fact. There will always be some window of vulnerability. Right now, for Swen and MSBlast, that window is over two months wide... as I am still receiving and quarantining those worms at firewalls.

One solution that may be effective, but still not _completely_ effective, is digital signing of binaries. If a binary does not have a recognized signature the OS will simply refuse to run it. The limitation here is there are still a lot of systems connected to the internet that are not upgraded on a regular basis and probably never will be. These neglected systems will always be a problem as long as they still exist. Eventually their numbers will drop so low that problems associated with earlier unpatched versions of Windows just become too sparse to sustain any sort of widespread problem.

I suppose there could also be a mechanism by which a system will only trust data from digitally signed systems and binaries... ...but this is beginning to sound like what they are proposing anyhow.

Ah well, just a few rambling thoughts.

-DU-...etc...

Don't forget consumers

[hankaholic]

Whatever changes are made to DRM, just remember what the consumers' position will be when DRM is commonplace:

Bent over, taking it in the poop chute.

Re:Don't forget consumers

[Delron Da Thugg]

Exactly. How long before some dirtbag IP attorneys representing some megacorp coalition get a human being patent and claim final, definitive ownership on us all? back to slavery!

Re:EFF

[Anonymous Coward]

How is that flamebait?

Libertarians always say they don't believe in handouts, so why should I give EFF a handout then?

If you're gonna promote that ideology then you can get your own damn money thank you very much.

Sad to see EFF legitimizing this

[Atario]

You're exactly right. In "Trusted Computing", as the analysis points out:

...the computer's owner is sometimes treated as just another attacker or adversary who must be prevented from breaking in and altering the computer's software.

I can't put it any more directly than that without risking being modded "Funny". Your computer, in effect, belongs to them. (See?)

Even the proposed "Owner Override" seems to me a "how are you going to do that" issue. How are you going to assure that a change was made by you and not by some software pretending to be you?

There are other oversights too:

• "Identity" of software is determined by submitting a hash value, but how can you be sure someone's not sending a canned hash value?
• "Secure output can prevent information displayed on the screen from being recorded" -- until someone invents a screen-scraping monitor. If information exists, there's a way to copy it. That's just what information is.
• The most serious point of all -- that the EFF is lending credibility to this blatant grab for dictator-like powers by suggesting that it can be "fixed" and the problems "addressed", at which point we should all happily adopt it. Not me, brother.

I would have much preferred the factual analysis and then a great big "run away from this as fast as you can".

Re:Sad to see EFF legitimizing this

[Sloppy]

Even the proposed "Owner Override" seems to me a "how are you going to do that" issue. How are you going to assure that a change was made by you and not by some software pretending to be you?

That question is the machine owner's problem to deal with, and "Trusted Computing" spec and app developers need not concern themselves with it. It doesn't matter how it gets done (or even if it really gets done). The owner takes responsibility.

The most serious point of all -- that the EFF is lending credibility to this blatant grab for dictator-like powers by suggesting that it can be "fixed" and the problems "addressed", at which point we should all happily adopt it. Not me, brother.

Can't you see through the lines here, that EFF is giving everyone a naughty wink as they speak? Their proposal, by attacking the core value of Trusted Computing, is dripping with sarcastic irony. I love it. Their "fix" completely subverts the bad guys' intent. With friends "lending credibility" like this, Trusted Computing advocates wouldn't need enemies. :-)

Re:Sad to see EFF legitimizing this

[SiliconEntity]

Even the proposed "Owner Override" seems to me a "how are you going to do that" issue. How are you going to assure that a change was made by you and not by some software pretending to be you?

The idea would be to use the secure I/O capabilities to make sure the user approves the change/override at the keyboard, which can't be spoofed by software in a TC system.

"Identity" of software is determined by submitting a hash value, but how can you be sure someone's not sending a canned hash value?

The hash value is cryptographically signed by a key generated in the Trusted Platform Module. The key never leaves the chip and only the chip can issue such signatures. This is what makes sure that the hash values are correct.

The EFF's proposal actually amounts to letting you submit a spoofed or canned hash value, which makes the whole attestation feature useless.

"Secure output can prevent information displayed on the screen from being recorded" -- until someone invents a screen-scraping monitor. If information exists, there's a way to copy it. That's just what information is.

The (claimed) purpose of the secure I/O is to prevent software in the computer from being able to see certain parts of the screen. Obviously the user can see it, photograph it, etc.

The most serious point of all -- that the EFF is lending credibility to this blatant grab for dictator-like powers by suggesting that it can be "fixed" and the problems "addressed", at which point we should all happily adopt it.

This is just inflammatory rhetoric, something the EFF analysis was refreshingly free of. There are no dictator-like powers being grabbed here. At most, TC lets you prove your software configuration to third parties, allowing them to refuse to perform services for you unless you use certain software. That's hardly dictatorial.

Re:Sad to see EFF legitimizing this

[Alsee]

How are you going to assure that a change was made by you and not by some software pretending to be you?

Actually that is pretty easy, you press a special button/switch. Malicious software is incapable of faking actual physical control. I proposed exactly such a modification to TCPA months ago.

I e-mailed this one of the main TCPA proponents about this back in January. It was David Safford, author of Why_TCPA and TCPA_Rebuttal. I explained this system and pointed out that there every single claimed benefit of Why_TCPA works just as well with actual and full owner control like my (and the EFF's) proposed modification grants. He did not dispute this.

His only reply was to suggest this change would no longer keep laptops secure against a thief. This suggestion fails on two grounds. First of all it directly contradicts TCPA_Rebuttal where he claims TCPA is not designed to be secure against physical access and that this supposedly 'proves' that TCPA is not designed for DRM. If TCPA is not supposed to be secure against physical access then it is disingenuous to claim it is supposed to protect a laptop against theft. The second reason his 'theft' argument fails is that it is simple to combine a physical button-press with an owner ID code or password before full control is given. A theif cannot get this owner password, and software can neither get the password nor press the button.

Granting the owner of the machine to his own keys (passwords) that are locked in the TCPA chip gives the owner full control over the system. There is absolutely no justification for denying the owner access to his own keys. The only purpose for this design requirement is to use it as a weapon against the owner and for various varients of DRM.

Of course Microsoft and the TCPA proponents will never accept my proposal (and the EFF's proposal) because the only real motivation for this hardware change is for DRM-type purposes. If owners maintain actual control over their machines and it can't be used for DRM systems then the entire project is a waste of time. Everything else is just a smoke-screen. TCPA will not prevent your computer from being infected with a virus, and it will not prevent that virus from slagging your entire hard drive and everything on it. The only thing it will do is prevent the virus from distributing copies of your 'secure' music files.

-

Re:Sad to see EFF legitimizing this

[Sphere1952]

"Of course Microsoft and the TCPA proponents will never accept my proposal (and the EFF's proposal) because the only real motivation for this hardware change is for DRM-type purposes. If owners maintain actual control over their machines and it can't be used for DRM systems then the entire project is a waste of time."

Actually, it only takes some owners. Let them implement remote attestation, but make sure it remains legal to build hardware with the EFF's owner override feature. If it's legal then there'll be a market nitch for the hardware, and the hardware will be available. I'd suggest that the value added would be about $50; which should be enough to make some currently small-time vendors drool (and probably make them large-time vendors).

More people might learn how to replace chips in their machines too...

Re:Sad to see EFF legitimizing this

[WNight]

I agree. I think the EFF is saying essentially, "TCPA is anti-customer, remove [core feature that sounds minor] and it won't be." They're doing it in a way that people can't help but agree with "Owner override? Of course I trust myself," so that the TC advocates can't ignore it. But, if they comply, the intended 100% trust won't be there and none of the friendly companies advocating this will want to be involved.

It's the computer equivalent of fluffy kittens and kissing babies, of "won't someone please think of the children." In other words, it's a hand-wave, a delibrate misunderstanding and misstating of the issue, intended to redirect everyone but the frustrated original person, the TC advocates in this case.

Do they want to be seen as being anti-consumer? Do they think they've got popular support, given the results the RIAA is getting?

Re:Sad to see EFF legitimizing this

[Alsee]

Let them implement remote attestation, but make sure it remains legal to build hardware with the EFF's owner override feature.

Won't work. The remote attestation involves checking a sort of serial number (key) hidden in the chip without ever actually revealing the number to anyone. Part of attestation verifies that you have an approved key number.

If someone else makes hardware with an owner override feature it will fail all attestation requests because it doesn't have an approved key number. And the key numbers are a few dozen digits long. It is essentially impossible to guess or calculate a valid key number.

The only way to get a hold of an approved key number is inside an approved chip. The hardware specification specificly REQUIRES that that approved chips may NEVER reveal this key to it's owner.

If any approved manufacturer ever does produce a "overridable" chip then the central certificate authority can immediately revoke every chip ever made by that manufacturer. Every "Palladium enhanced" or "TCPA enhanced" computer containing a chip ever made by that manufacturer would immediately go dead. Yes, a computer you've been using for over a year would simply go dead if the manufacturer made a NEW chip with override ability. You would irretrievably lose ALL of your protected data. And it's impossible to make a backup of protected data.

If they manage to push this system on us it will be interesting to see what happens when that face that situation. They designed in an ability to revoke every key from a manufacturer if that manufacture ever intentionally or accidentally puts a hole in their security. But if they ever USE that ability it would effectively destroy millions of computers. They HAVE to rekoke the chips to maintain security, but they CAN'T revoke them because there'd be millions of people screaming for blood because their computers got slagged.

You can't implement the EFF's proposal unless "they" want to let you. And they will NEVER let you because the central design requirement was to make it impossible to "override" like that. The central design goal was for secure DRM.

Pretty much the only way to override the system they've designed is to chemicaly peel the secure chip you bought and read your key number out with areally high power microscope and lab equipment. The good news is that once to do that you can get "god level" control over the system, like the EFF override. The bad news is that each key you dig out is pretty much only good for one person. If they ever detect that you've done this they'll revoke that key.

-

Re:Sad to see EFF legitimizing this

[Sphere1952]

"Pretty much the only way to override the system they've designed is to chemicaly peel the secure chip you bought and read your key number out with areally high power microscope and lab equipment. The good news is that once to do that you can get "god level" control over the system, like the EFF override. The bad news is that each key you dig out is pretty much only good for one person. If they ever detect that you've done this they'll revoke that key."

If they try doing key per user (chip) I bet it blows up in their faces.

If they do key per manufacturer then there'll be a few chips sacrificed. The key will get out for anything worth having the key to rather quickly.

Re:Sad to see EFF legitimizing this

[Alsee]

If they try doing key per user (chip) I bet it blows up in their faces.

That is the TCPA and Palladium/NGSCB plan. I certainly HOPE it does blow up in their faces. But they have every intention of making sure that every new computer sold comes with this locked chip inside and a unique code for each computer. I've been reading the TCPA design documents. Detailed specs on Palladium/NGSCB aren't really available, but the capabilities are fundamentally the same as for TCPA and the hardware would need the same requirements for the hardware.
Microsoft's stated plan is for the DRM to be "invisible". When things work you wont even notice the DRM running in the background. And when it "doesn't work" you aren't suppossed to notice because the option to even do it simply won't be there. For example you won't notice that trying to save a JPG from a 'secure' website won't work because when you right-click the image the browser simply won't have a SaveAs option.

If they do key per manufacturer then there'll be a few chips sacrificed.

Maybe I wasn't clear here. Every chip has it's own key, but that key is also signed by the manufacturer. They can revoke one individual chip or they can revoke every chip from a manufacturer.

They have been burned many times in the past by DRM failures, so they designed the system with the ability to revoke any portion that gets compromised, and they designed in the ability to FORCE you to accept patches to the system immediately, or your computer goes dead.

How do that do that? It is explained in detail in Microsoft's DRM-OperatingSystem patent. Your computer is only fully functional while you have an active connection to the internet. If you don't have a connection, or your connection goes down for some reason, then some or all of your computer locks up. The computer needs an active connection to continously check with a cryptographicly autheticated time-server. That's so you can't simply reset your computer's clock to "cheat" on time-based DRM, like keeping music for more than a month or watching a movie after 24-hours expires.

Any time you want to play music for example, it would contact the music DRM server to see if it there are any new patchs to fix any hole discovered in the DRM. If the music player can't reach the music server, or if you refuse to accept the new software patch, then the assumption is that there IS a hole in the DRM and you are trying to "cheat" the system. All of your music files go dead until you connect and accept the patch.

The whole plan is that things simply go dead any time there is even a possibility that you might be able to get around the DRM.

-

OMG!

[Simple-Simmian]

OMG! All Libertarians are evil! OMG! OMG! Libertarian thought is evil thought and must be stameped out. OMG! OMG! Freedom is evil and bad OMG! OMG! I actually read the articles. The EFF is right. I don't think they are exactly Libertarians though. We are going to have a left right civil war in this country and the left is going to precipitate it. The A pox on the Democrats and the Republcans. Screw the left right labels. I am proud to be a Libertarian who isn't tring to force or brow beat others into "thinking correctly." That is what the left and right do. Libertarians are not "right wing" they are not leftists either, they are concerned with personal liberty.

Re:OMG!

[Anonymous Coward]

If you don't understand why libertarianism is a rightwing ideology than you are indeed quite simple.

Re:OMG!

[Simple-Simmian]

BZZZZZZT Wrong answer.

Re:OMG!

[Anonymous Coward]

I'm sorry that you have a feeble grip of politics and economics.

Don't feel bad though, most Americans are like that.

Re:OMG!

[Simple-Simmian]

BZZZZTT Wrong answer again

The trouble is...

[tkrotchko]

If this is unopposed, it will not be long until everything useful requires "trust". And so my PC, the one I paid money for, will not work the way I want anymore. Oh, theoretically it will, but in a practical sense it won't.

If a content provider wants to "trust" a device, then they should buy it for me.

My cell phone providers wants a trusted device. Great. They give me a phone, and I pay to use it.

Ask yourself this... is watching an HDTV version of Star Wars so compelling that you're willing to compromise yout ability to control your PC? If you answered "yes", then you and I simply have a completely different viewpoint on the subject that I suspect we'll never agree on.

Re:The trouble is...

[bunnie]

I completely agree with your point about everything eventually requiring "trust". I am fearful of this whole trusted computing movement for the very reasons you have stated.

However, to be fair, I think with Seth's tweaks to the attestation policies, you could have trusted computing without the intrusion on user's rights. Could, I say. I'm still not decided if in practice it would actually end up that way.

One of the things I would like out of the trusted computing initiative is the guarantee against keyloggers and/or a way to securely store my private keys. If you saw the recent news on Valve's software break-in, part of the break-in was a result of keyloggers...we are too vulnerable to really simple attacks such as keylogging.

The

[erroneus]

That's a CRAZY idea. As usual, let's compare computing on the information super highway with driving on our own freeways.

What would happen if we let people drive their own cars? They would repair their own cars, "upgrade" them too! But if they are in control, they may not make repairs as needed and then their cars would fall apart on a public super highway and cause other people to die and stuff.

Oh wait... we have a "license" to help ensure that the public has a bare minimum amount of knowledge and skill to operate a vehicle safely on public roads.

Now let's return to cyber-reality again. Instead of "trusted computing" how about "trusted users."?

Let's say that the price of admission to the information super highway should be controlled in the same or similar way to the way we control access to the roads. What a fabulous world we'd live in! "License to SPAM" wouldn't exist. Maybe there are a lot of bad things I haven't considered but is it much worse than requiring a driver's license write a check?

Wow... imagine getting a ticket and your license revoked for SPAMing... or for operating a computer with a virus...

"The Responsible Computing Initiative" is born!

Re:The

[Anonymous Coward]

Provide a way for licensing to not discriminate on race, age, social status, sex, origin, or any other non-relevant factor across all national boundaries while enforcing the same or more anonymity as currently provided by the internet, and I might be for it.

Providing even this, spam and other marketting is unlikely to be removed in a capitalist society where the costs to advertise are less than the rewards. Ignoring spam is the largest step to stopping spam. Without monetary reenforcement, spammers will have to spend more time working elsewhere to support themselves.

Re:The

[Anonymous Coward]

That's reprehensible, but let's have a license before people can have children. A baby costs a million dollars over a lifetime, and does significant ecological damage during that time.

Re:The

[shaitand]

true dat, not to mention there are simply some people who have no business breeding and spreading their genes to begin with! They simply are not on par.

Ahh, the return of dongles.

[SuperKendall]

Very fancy dongles to be sure, but here they are again. Will users reject them for the same reason?

Re:Ahh, the return of dongles.

[Buzz_Litebeer]

They are very easy to install, at least CodeMeter is, it "looks" like a USB harddrive, but the software can make calls to it through a COM interface.

The dongle "SCREAMS" ease of use, you merely stick it into your computers easiest USB port and BAM you have a working dongle. USB is also pretty robust, and the place I interned at only had a few returned WIBU USB dongles returned, and that was mostly due to children stepping on them, or in one case a child breaking it in an attempt to get back at his/her parent so they couldnt get on the computer.

Now I know that kind of thing MAY happen, but thats why licensing can be handled over the internet very easily, especially with CODEMETER which has a protocol just for that purpose.

With the amount of users accessing the internet, a site that say encoded everything comeing from it with the public key of the remote user, ONLY the remote user using his dongle that has the private key stored on it can access the data (that btw was my job, to explore the feasiblity of this) and the USER doesnt even have access to the private key, since that part of the data translation is in "hidden" sectors of the key.

So no one knows the private key, and the sites you want to get data know the public key. This makes it very secure and ensures you have the proper dongle.

If you register a new dongle, you get a new private key, and returned licencing data (which are two different things) so that you can use your new key even though an old one might have been broken.

Since it also has an internal limit counter (to be set by the licensure) the people who create products that rely on the key, can also force a remote user to update his licence every X uses/days so that he doesnt call a key in false serveral times and have offline use of his product with multiple keys and invalid licences.

Its pretty neat, and since I took several grad classes in Info Sec I can tell you its pretty darn secure, though not infalible. I mean if someone comes up with computing that can check all possible keys, then well there you go.

hehe

Trusted Computing, but who is the trusted party?

[Adrian Lopez]

Trusted computing is being promoted as a computer platform that users can trust, but it's really more about creating a platform where third parties can determine whether the users themselves can be trusted.

Trusted computing benefits content producers and service providers more than it benefits users. The reason is that producers and providers are usually the ones whose systems are being acessed, while the users are the ones accessing these well-known systems. It is the nature of the transaction that trusted computing will favor the well-known party over the party that is "anonymous".

Think People Think.....

[3seas]

when the hardware gets hackered and cracked, then you will have to upgrade....

Whats the difference between hardware and software when it comes to bit flipping?

to put something in hardware gains the advantage of speed but the inflexability of change.

Related News

[superyooser]

Microsoft Preps Major Security Strategy Shift

Microsoft's chief security strategist made the surprising statement that the company is about one-third of the way to its goals for Trustworthy Computing. I guess there's a lot more going on internally than we're aware of.

The article also says, "Microsoft's short-term strategy will shift from patch management to what the company calls 'securing the perimeter.'" What this means is that they're working more closely with firewall companies.

Re:Related News

[Anonymous Coward]

What this means is that they're working more closely with firewall companies

Only in the way that a wolf would "work closely with" an antelope. Microsoft doesn't "work with" smaller companies for long before eating them. (and they're all smaller than Microsoft!)

Here's the progression, for those that haven't been reading MS stories on /.

1: Recognize demand for some cool technology
2: Partner with small firm that owns cool technology
3: Copy small firm's cool technology, roll it into Windows
4: When small firm sues, stonewall until they go bankrupt from the legal fees.
5: Modify Windows N+1-SP3 to break small firm's still-superior-but-now-unmaintained implementation of cool technology
6: PROFIT!!!

If you're current working for Norton or Symantec or ZoneLabs now, you should start working on your resume (and sending it to to Microsoft!), otherise you're going to be working at McDonalds next year.

Etch SW into HW and Make User Data the ONLY I/O

[Anonymous Coward]

As Jason noted earlier: "...'Trusted Computing'...exists...to take control away from the computer owner and give it to the content owner... what is the point of the EFF proposing 'fixes' to help keep the computer owner in control, when its primary design goal is the exact opposite?"

Home PC users should tell scummy Big Brothers Micro$oft, Intel, Hollywood, etc. to shove it. I'm not going to pay you for a DRM'ed PC, and let you charge me usage fees, force-feed me content, view my private info, etc.

If businesses (or home users) need ultimate security, jump back to the days of a closed hardware box with etched-in software; the only I/O was user data, so, no I/O was ever considered program code (no more viruses!). This would mean that the box would have to leave the factory with a DEBUGGED, etched-in O/S; DEBUGGED, etched-in office-suite software; and hardware slots into which additional purchased software (made by any company, etched onto hardware cartridges, and memory-isolated by the hardware box) could be plugged.

This would mean NO MORE BELLS-AND-WHISTLES CRAPWARE...keep it mean, lean, and bug-free, because any patch will have to be a free replacement cartridge (or you piss off your customers).

This would mean that the closed box with hardware-cartridge expansion is a BUSINESS MACHINE. You could still buy the PC of today for your home use and program the PC to your liking...but it could never corrupt the business machine. Want to bring your work home with you? The BUSINESS MACHINE could easily be of laptop design.

The point is, the CRAPWARE and viruses of today's PC...could never touch your BUSINESS MACHINE or its user data.

END OF STORY. PROBLEM SOLVED. No more asinine "Norton Anti-Virus" and its drug-addict subscription fees. No more asinine "Microsoft Windows Updates" because of over-featured, crapified software released too early. No more script kiddies. No more employees putting WHATEVER CRAP THEY WANT onto the BUSINESS MACHINE.

Anyone who nags about:

(1) the locked-down, basic-software-etched-in-hardware box,
(2) the cost/inconvenience of cartridges versus the FREE-FOR-ALL of downloadable Web software (such as broken-software patches, utilities for things the O/S should have been doing in the first place, etc.), and
(3) lesser user freedom (to add additional, company-unapproved software to his work machine)

HAD BETTER THINK ABOUT ALL THE WASTED TIME AND MONEY WE ARE NOW SPENDING ON VIRUSES AND OUT-OF-THE-BOX-BROKEN, CRAP-FEATURE-LADEN O/S's AND SOFTWARE.

Do that, and software etched in hardware...with I/O consisting ONLY of user data...DOESN'T SEEM TO BE SUCH A BAD BUSINESS IDEA AFTER ALL...does it? :-)

Re:Etch SW into HW and Make User Data the ONLY I/O

[Anonymous Coward]

Hell don't get mad about it. Just reformat and reinstall once a year to reset your Norton subscription.

Re:EFF

[cduffy]

Libertarians always say they don't believe in handouts, so why should I give EFF a handout then?

Libertarians don't believe in handouts funded by individuals who didn't explicitly and personally agree to provide those handouts. So, say, if money that was taken from me via taxes is being given to the League of Gay Midget Eskimos without my consent, that's a bad thing. I may be more than happy to donate to said League if it were my choice -- but being forced to do it at the risk of men with guns coming and putting me in jail is a different matter.

The EFF is the same way. I don't believe in enforced handouts to the EFF from folks who don't support them -- if you don't like the EFF, you shouldn't be forced to donate to them. On the other hand, if you believe that donating to the EFF is something you wish to do -- perhaps even something which is aligned with your own enlightened self interest -- then you should be every bit as free to do that as to donate to the Gay Midget Eskimo fund. Which is to say, very.

Trusted Computing != DRM

[Anonymous Coward]

in the article it clearly states that DRM is only a small part of trusted computing.

Attestation

[TeachingMachines]

With Microsoft, IBM, and other major players involved in this process, the EFF doesn't have much of a choice but to work with what they've got. I don't think that the EFF agrees with the Trusted Computing initiative; as they say in the article, most of the changes described by the initiative can be implemented at the software level. I agree that that is where the changes should take place.

I agree with some of the other posters here and I don't really see anything useful about the attestation process (see the chart at the bottom of the page). I'm especially concerned about all of hardware specs that I know nothing about: Do you honestly expect me to think that the Bush administration isn't salivating over this? Can you say "backdoor"?

It sounds pathetic, but the only way I see out of this is through education and certification. People should be certified to connect to a network, and if they screw up, they are responsible. It's the way it works (usually) in academia.

What a mess.

Dongles != trusted computing

[Wesley Felter]

You're talking about dongles, not trusted computing. All dongles are crackable by simply modifying the application not to check the dongle at all.

Dongles don't provide sealed storage (which is pretty much the only useful feature of trusted computing), so they are not an alternative to trusted computing.

Re:Dongles != trusted computing

[Buzz_Litebeer]

in concurrence with an operating system they can be used for sealed storage, and key decryption. Right now, griffin technologies has a working (though not released) program that can encrypt/decrypt based on secure keys stored ina dongle onto the hard drive. Without the dongle you dont get access to the data. even if you try to make whatever program is accessing the data "bypass" the dongle, it will only read it in "raw" encrypted form, which doesnt help

Re:Dongles != trusted computing

[Buzz_Litebeer]

Oh and btw, these are the "future" of dongle based computing. They have thier own memory, their own processors, and their own Trusted memory which cant be accessed, and only written to by a vendor. Yes you can write a program to "re-write" the trusted sectors, but you cant read them.

This is Trusted computing, the only difference is the dongle is a small computer hooked to yours and you can carry it around. Instead of being forced to lug your machine around to listen to that copy of britaney spears, you can use the dongle as the authenticaton device, and if you have the dongle in, you can download the music from the site again and just listen to it again.

On audio/visual stuff it doesnt cover the analog hole, but it is a step to portable DRM, wherease being screwed with one machine that can run things but another cant is no fun at all.

Re:Trusted Computing, but who is the trusted party

[SiliconEntity]

The "trusted party" in trusted computing is the software. TC lets you trust software to behave in a certain way: software on your own computer, or software on remote computers, with the owner's permission.

Other Problems with Trusted Computing

[SiliconEntity]

There are some other problems with Trusted Computing that the EFF article fails to address.

One is the difficulty of dealing with upgrades, failures and replacement of computers, if your data is locked to the old machine. TCPA had a hugely complicated process you would have to go through to migrate any of your "secure" data to the new machine. It involved going back to the manufacturer, getting a special transfer key, moving the data over and having it get re-encrypted. Microsoft hasn't said what they're going to do, but it's an extremely difficult technical problem to solve while retaining the security.

Another problem is the PKI (public key infrastructure) issue. For remote attestation to work, it's necessary that the TC chips have some kind of crypto certificate that says that they are legitimate. Microsoft has said nothing about who will issue these certificates and who will revoke them if a machine gets broken into. Setting up a successful, global PKI is a prerequisite for DRM type applications and will be an enormous job.

The article also overlooks that the sealed storage feature, which the EFF mostly views favorably, can also be used to achieve lock-in and secure closed formats. Microsoft Word could store data encrypted using the TC hardware, such that only Microsoft-signed applications can access the data. This kind of lock-in does not depend on the remote attestation features that the EFF is so concerned about, and would not be addressed by their Owner Overrides.

Re:Other Problems with Trusted Computing

[hanssprudel]

Another problem is the PKI (public key infrastructure) issue. For remote attestation to work, it's necessary that the TC chips have some kind of crypto certificate that says that they are legitimate. Microsoft has said nothing about who will issue these certificates and who will revoke them if a machine gets broken into. Setting up a successful, global PKI is a prerequisite for DRM type applications and will be an enormous job.

More importantly, the embeddded certificates used for the remote attestation (TCG calls them "Endorsement Keys") are uniquely bound to the chip and computer in question. That means that every site that requires attestation can identify you and track you: think of it as Intel's processor ID on steriods!

Competition will make trusted computing OK

[SiliconEntity]

I know no one wants to hear this, but the dark picture painted by the EFF of the ills due to Trusted Computing is not likely to come to pass.

The main point that the EFF analysis overlooks is the role of competition in the marketplace. Yes, TC could allow web sites to require you to run particular software; yes, TC could allow vendors to encrypt their data formats making it impossible for you to switch to a new software package; yes, TC could be the foundation for DRM and restrictive licensing.

But the point is that not all companies would use TC to do these things. Users would have a choice between companies which impose very strong restrictions on how end users can manipulate their data, and companies which offer open and unrestricted data formats. If all those limitations which TC would allow companies to impose are so bad, customers will refuse to buy the software of those companies. Competitors which offer unrestricted data formats will win in the marketplace.

Look at what is happening today with online music. By the end of this year, there will have been several launches of online music services, each with its own tradeoffs of per-song pricing, subscription fees, and download restrictions. This is competition. The market will respond, and we will get to a situation that provides a balance between the desires of all parties involved. Some DRM will exist, but it will be in a form that customers can accept.

In the same way, TC can be used lightly to enforce DRM and other restrictions in a way that users will not find objectionable and onerous. Competition will evolve a balance between the desires of the vendors and those of the customers, just as it does for prices, features, licensing and all other elements of a software purchase. Neither side is in a position to dictate terms.

Re:Competition will make trusted computing OK

[Anonymous Coward]

You miss the point.

The problem isn't not what TC enables (verification), it's what TC needs to disable in order to achieve that.

I don't mind restricted content formats per se. The problem is that they are against the nature of programmable computers. In order to support any restricted content formats, the entire computer must be modified into something that prevents me from controlling it.

Re:Competition will make trusted computing OK

[Anonymous Coward]

The double negative was a braino, obviously.

Re:Competition will make trusted computing OK

[Anonymous Coward]

>>But the point is that not all companies would use TC to do these things.

We already know that companies will try to do this, although without TC attestation 'features', it doesn't really work. There are many sites that say you have to have particular applications, and the only reason we can access them with 'unnapproved' software(in this case browsers), is that we can mimic the approved applications in functionality with another application. Another example is IM networks, as we have seen lately, the companies that run this desire to prevent compatible third party clients from connecting to the network. If you give them a tool to be 100% sure what you are running, why would you think they wouldn't use it?

>>Competitors which offer unrestricted data formats will win in the marketplace.

I believe in the power of competition as well, and would like to believe that this would be what would happen. However, Windows platforms with the standard applications used on Windows are dominant, particularly if we are talking about IE, which is nearly ubiquitous. Most people probably won't even notice that their freedom has been restricted, because they use the 'approved' platform. And of course, there may be a couple of approved platforms for certain applications, but there is no doubt that when you give vendors the option to restrict access based on what you are running, you are reducing choice for the consumer, and, by extension, circumventing competition.

Don't get me wrong, there will still be some choices, but accessing some things will probably end up requiring you to use either one particular application or a small subset of the applications available. The opportunity cost of 'locking out' the small percentage of people who don't happen to be using 'approved' applications is small compared to the percieved 'risk' of offering services to those who use 'rogue' applications.

Re:Competition will make trusted computing OK

[Anonymous Coward]

My patent addresses this. If it's not DRM encrypted the Speakers, Monitors, TV's can't play it.
NOISE.

Re:Competition will make trusted computing OK

[Anonymous Coward]

Competition won't matter once Microsoft buys off the legislators to make TC the only legal form of computing. Sure, you could buy non-TC hardware from a foreign country, but if they detect you hooking it into the internet on US soil, they'll come get your ass.

There may also come a day when US laws affect the entire world, at which point even getting the non-TC hardware may be very difficult.

Even more spam, ugh...

[Anonymous Coward]

Considering public perceptions including the constant repetition of a less then favourable political agenda (agenda based interpretation of psychology including but not limited to criminal behaviour). Open support demonstrating opposition can mean alienation (even if it's the process of deduction and observation). When business is involved public perception can destroy a company regardless of quality, potential or community involvement (what's on your side walk). Maybe it's time to take a step back and observe. Regardless of political motivation business is business and removing peoples ability to accurately intrepret their environment does seem at present to offer the most reliable return.

In other words getting caught up in the hype, here or anywhere can threaten income generation and pay a heavier cost in demoralized.

I try to avoid manipulation (because I do love my own liberty) and it's sad to see that it's so damn effective.

"When the architects of our republic wrote the magnificent words of the Constitution and the Declaration of Independence, they were signing a promissory note to which every American was to fall heir." Some guy's most important and long overdue delivery.

Re:Trusted Computing, but who is the trusted party

[Adrian Lopez]

The "trusted party" is indeed the software, but whose software, exactly?

The point I'm making is that it's usually the user/client who must show that the software he is running on his computer may be trusted by third parties. I'm suggesting that establishing such trust relationships is the primary purpose of the trusted computing initiative, more so than users establishing trust relationships with their own software (it's easier to fool the user than it is to fool a third party's computer).

I agree it's not as annoying as dongles were but..

[SuperKendall]

What usually happens is that a cracked version not requiring the key/dongle is released, so people who want to copy the program still do so. And then you, a real user, are trying to use the program at 2am when the internet goes down or the devices fails are you are stuck with an app that will not run.

That's why if I buy a program that requires a CD key or something along those lines, I almost always download the cracked version or updater and use that instead.

If that's what you want...

[autechre]

...you can run Linux on your business desktop today. It's not quite the same, but you can have software installed which does exactly what the users need, tailor everything to your specific requirements, and disable executables on /home (most users wouldn't need them). It's even better if you use X terminals, because all the desktop machines are identical, and users can switch between them with no problems whatsoever.

The problem with the hardware solution that you propose is that it would make free-as-in-beer software almost impossible, as all software would have to be distributed on a physical medium, which costs money.

The real problem

[Dhalka226]

The real problem, as I see it, is not trustworthy computing, it is certainly not protecting the users, and it is not even corporations--rightly or not--seeking to protect their investments by invasive means.

Instead, the problem is a generally uneducated user base. I don't mean "uneducated" in the sense that they are in any way unintelligent, but that for some reason they are simply not interested in learning the intricacies of computers and related topics. They simply want things to <I>work</I>, they don't care <I>how</I> they work. And the truth is it would take an immense amount of invasion of privacy before the average computer user noticed, much less began to raise a fuss that might stop a company from heading in that direction.

The question, then, becomes how do we educate people who do not wish to be educated? If we write them off, is the cause lost? It seems even vocal critics such as the EFF go mostly ignored by companies even as the hordes of us behind them applaud. Bill Gates just smirks and buys himself another ivory back scratcher.

Can the tech-savvy win in a world of technological indifference?

Re:The real problem

[Sphere1952]

How many million filesharers are currently taking a crash course in computer security?

Re:The real problem

[SmurfButcher Bob]

Ah, this TCP stuff basically does a flip-flop of the current user issue.

See, right now... users have lots of authority, but no accountability for what happens on their machines. Send a virus to 50 million people? Got Zombied last month and *still* haven't fixed it? It's a nightmare.

TCP will change that, a lot. Once TCP is implemented, users will have little or no authority over their system's behavior, but all of the accountability for what it does.

Think about it.

Give them just enough to hang themselves.

[freality]

Well, I thank the EFF for this analysis, but I think they've missed an important tactic. Let Microsoft and Co. lock out non-MS software all they want. They're at a fundamental disadvantage. If they wish to exacerbate their tenuous position vis-a-vis monopoly, fine. If they want to gamble shareholder confidence on a risky offensive against the general good will of the net public, we should help them.

The EFF warns that Microsoft's IIS web-server could block web-browsers other than Microsoft's IE. Well, Apache can just as easily be made to block IE. After all, Apache has run the majority of Internet web-sites since 1996. In other words, if MS doesn't play nice, we shouldn't reward them by rolling out the red carpet. Kick MS off the net (maybe for just a year or so.. mercy and all). You can start sending the message now.

Re:Give them just enough to hang themselves.

[OeLeWaPpErKe]

Please include the tactic apache would use to force users to upgrade, AND the tactic to prevent a fork of the apache project.

Otherwise, you're full of shit

Incidentally...

[tkrotchko]

I bought the book a few months ago... I really like it. There's very few tech books I read cover to cover any more, yours is/was one of them.

Thanks.

Not telling

[taff^2]

From the article:

Refusing to provide the information required by remote attestation won't work, Schoen said, because such a refusal is still giving something away. "In criminal cases, you can take the Fifth Amendment," he said. "While the jury is not supposed to infer anything from that, the general public certainly infers that the person is guilty or has something to hide."

I think this will work, and this would be a feature that I would love to have. You can infer whatever you want about the contents of my hard disk. I remain innocent until proven otherwise, or is that not the way things work anymore?

Nah, coverup--

[michaelhood]

As much as I despise Outlook's vulnerable status, the consensus in our group is that Valve probably had it at www.valve.com/hl2source.tar.gz, and they were so embarassed they engineered this story.

A joke. Food for thought.
If I get modded flamebait for this, I'm going to cry at slashdot's sad state of humor.

Er, Does this work

[ColourlessGreenIdeas]

According to their idea, you can lie about changes you've made.But you can't claim that something that was never true about your computer is true. While this provides more compatibility with the dictionary definition of the word 'trusted', it doesn't solve all that much. Imagine a DRMed version of CIFS that only connects to MS clients. Under the EFFs scheme, a client that runs a hacked version of Windows is OK, but a client that never ran Windows isn't. Samba is still dead.

Re:Er, Does this work

[Sphere1952]

Want to buy a key?

I, for one

[Anonymous Coward]

welcome our new Disney Computer Overlords

Neuromancer here we come!

[Sphere1952]

Where can I buy a box with "owner override"?

It will only take one vendor doing it, and I'd pay a few extra bucks for my Linux owner override feature to work.

Ice, ice baby.

I can see the use

[ThaReetLad]

I'm a gamer. I play Ghost Recon online and an a moderator for a well respected community/clan (WGC). Our major problem is cheaters and Team killing smacktards who come in under random IP addresses, random names etc. With a decent trusted computing environment we could deny access to people who were running unverified mods or cheats or who were known to be arseholes.

That's not to say I support TCPA exactly, but when software is client side the server has to be able to trust that the client has lived up to its side of the deal, and if a trust relationship can be set up in a secure way that gives service providers a way to be certain that the client is who and what they say they are then I'm in favour so long as there are legal protections in force that prevent it being used as a way to limit competition through arbitrary assertions. i.e. checks against OS or other peripheral factors.
Also if a service is paid for providers should be forced to provide validation for 3rd party apps and an SDK for anyonewho wants one at a reasonable price.

DRM is small potatoes

[davide marney]

The DRM applications of this technology are small potatoes compared to the ability to lock-in consumers to an application suite (major score for the capitalists) and the ability to lock-out subversive information (major score for government censors).

That said, something absolutely must be done to protect end-user computers better; the current state of affairs is intolerable. I thought the EFF did a nice job not just crying Chicken Little, but making a specific suggestion on how to prevent the abuse of this important, needed technology.

Re:DRM is small potatoes

[Alsee]

something absolutely must be done to protect end-user computers better

This is actually an important question - what exactly do you want to protect them from?

Despite all of the propaganda, Palladium/TCPA are not designed to secure the computer against outside attack and intruders. They are designed to secure the computer against the owner. Despite the fact that they both use the word "secure" those two goals actually have almost nothing in common. Despite the claims to the contrary, Palladium/TCPA are nearly useless for protecting a computer against viruses and intruders. They can't prevent you from getting infected, and they can't prevent a virus from erasing your entire harddrive. The only thing they can do is prevent the virus from making copies of your music.

Not only that, but remote attestation is the lynchpin of the entire system. Once you pull that out to prevent abuses the entire system falls apart, there's almost nothing left that couldn't be done with ordinary hardware and simply writing new software.

I have a gunuine question, once the remot attestation and DRM are gone, what exactly do you see that makes it an "important, needed technology"?

-

Re:DRM is small potatoes

[yakovlev]

Didn't read the document, did you?

The important thing that it provides is secure key storage. You can store cryptographic keys in a way that they can only be accessed on the owner's computer, and only be transferred to another computer by (I believe) their owner.

Also, these keys never have to be available in cleartext on the system, another protection against broken cryptographic programs or memory protection.

Trusted Computing: Govt Censor's Dream-Machine

[davide marney]

I think folks may be focusing too much on the capitalist implications of an abused Trusted Computing Platform. Government censorship is a much more serious threat.

For the "good of the people", President Bob dictates that everyone in the United Federation must use a trusted computer platform or go to jail. Dissidents? Bye-bye. Free press? Bye-bye. Long live President Bob!

If you don't have root to your own machine, you are not free.

Re:I disagree with the EFF.

[frkiii]

Wake up.

Regards,

Fredrick

Extra Spin Cycle

[SgtChaireBourne]

Two things are going to come out of this.

First, there will be a further clamp down on what is convered in the media and how it is spun: "To Microsoft, the threat is bad publicity, and they are going to produce a security system that deals with the threat," he said. [Schneier] said.

Realize also that Microsoft in all likelihood is going to try to make the option DRM patch mandatory, if for no other reason than to lock out competitors. ""Windows 2003 may be secure, but the level of security it provides could break backwards compatibility."

The last thing MS wants is for people to go over to the new version of OpenOffice.org or to avoid the hidden payloads in WMP9 by using Ogg. Office2003 and WMP9 are essential vectors in getting the "optional" DRM patch into Windows machines.

Re:I disagree with the EFF.

[Anonymous Coward]

I've a prior patent on this idea.
Cease and desist or face the consequences.
You can not tell people to "wake up" without giving me at the very least some credit for the term "wake up".
I've also patented the idea that trash is trash and any trash whether it's digital trash or physical trash it still goddamed trash. If you for any reason decide that what you have recieved via any orifice, whether that be physical or digital is trash and should be discarded, then I want a piece of the profit.
I'm sorry that Apple actually patented the trash can before I did but I've been able to alter the ICON several million times and have several thousand ICONS I can use to enforce MY original idea that trash is trash.
So if you throw anything in the trash either digitally or physically I want my fair share of the proceeds. If you save 1Gb of a 10 Gb drive that cost $30.00 then you have to send me 0.0000045876 cent's, and I want that amount for every hard drive that stored the unwanted document on it's way to you.

hamstringing the law-abiding

[GirTheRobot]

"Trusted Computing" operates under the assumption that the system itself is trusted. Given the number of security flaws in OS software in general, this is an obviously flawed assumption. In essence, a remote attacker exploiting a security flaw in your "Trusted Computing" OS has more control over your computer than you do.

Quite literally, not only are you just leasing your software, you are giving complete access and control of your data and hardware to third parties, often without your knowledge.

How many reading this would want a technology such as this serving as the platform and network infrastructure underlying government operations? The threat is not only to your own computer, but to your information as maintained by social services and law enforcement.

Yeah, lets give all the power to software vendors (including convicted monopolists), and hackers, while the normal computer user is left in the dark.

I pledge to use only open standards in all the solutions and networks I create. Freedom is far more important than "promised" security.

Trusted computing isn't about your trust.

[shaitand]

Trusted computing is about creating a platform the content providers can trust. Not the computer owners.

Whether the content provider is a network admin rightfully protecting the company owned computers on his network. Or microsoft/riaa wrongfully protecting the computer YOU own from copy infringing materials and from things they just don't like even though you have every legal right to do them.

It works fine

[Anonymous Coward]

You can attest to having a Microsoft software environment when connecting with smbclient. The point of owner override is that there need not be any connection between the PCR values you attest to and the reality. So you have a client that speaks the protocol and the other side can't tell that it's not the original client.

Return to the Dark Ages

[Alex Belits]

All that "trusted" platform does is preventing things that were developed over decades of technology development, and replaces them with things that no one thought about because they are useless.

1. Make sure that the user is typing at the keyboard, and not another program is doing it. That means, no remote access, no automation, no pretty much any feature that allows users to use interactive software unless it's sitting on the box behind his desk. Yes, one can try to send locally encrypted data blindly over the network -- then where is going the server to stuff it to be decrypted? And if it will be able to, why someone else won't be do the same with his own keystrokes, even if it will take a bunch of mechanical relays "typing" on a "secure" physical keyboard?

2. Trust the software application to provide the "safe" data. That means, no scripts, pipes, interpreted languages, or anything else that combines multiple "products" into an application. Because anything combined will have to be trusted as every component, ane every component (including "data" that is the interpreted program and the interpreter that runs it) will be trusted just as much as the complete system.

3. No virtual machines and emulators. Does not even deserve an explanation.

4. No user-created OS-level software, no matter in what language. Same.

Any of those features, if can be overriden by the user, undermines the system in its very core -- user may have a big red switch, but unless he can discern which particular software is running at the moment he is flipping it, he can not distinguish between bypassing the controls for his own program, for someone else's legitimate software, or for a worm/virus/malware/... Same applies even to self-signing system, with a nice addition of a problem in a networked environment, when one can not physically sign the application on all computers that should be able to run it, and all other methods will mean the ability to transfer and modify secret keys by the user.

So basically we will get a nice computer with all the features expected from ZX Spectrum, but in a "secure" environment. Obviously there should be something that will provide a replacement for those things. And there certainly will be -- there will be a remote access program that will be "trusted" that it already checked the validity of input on the client end, and can be "trusted" on the server. Single application signing service that will "let" the user run some software. Long explanations that emulators are only used by pirates, and that OS authors smell bad, so no self-respecting user would want to do any of those things, ever.

And the company that will bring it to you.... No, not _that_ company, the other one.

Ahhh.........sarcasm.

[Anonymous Coward]

Is it not lathered on thickly enough, or do the links offer no clue either?

Trust, Computing, Privacy, Convenience(God rights)

[h3lpdsk]

This analysis is very well scoped out to the extent that it focusses specifically on the topics of privacy, internet security and computers. But it seems to raise more issues than it defines. In a networked environment that is increasingly defined by technological convergence, the lines between physical and online presence are already becoming blurred. When the day comes that computers are able to routinely link to and manage our material comforts and all of that "stuff" is networked in, who defines our rights to privacy and the level of access we are allowed to what we already have? The EFF says the individual should have the right of ultimate access. (It's hard to disagree with that, isn't it?) But some sectors, if not most sectors of the government could argue that, "for security reasons", they have a legitimate claim on the right of access to private identity. Identity theft is one issue that tends to feed legitimacy to this argument, as do the precedents of social security numbers and drivers lisences as accepted standards if identification. Privately governments are mostly concerned with being able to identify, locate and regulate people (that's what governments do). Of course businesses that produce convergence technology will argue publicly that they should have God rights because they "understand" the technical issues better than the mere mortals they service, while privately conspiring to lock in their market and stake out their claims to access and distribution. (That's what businesses do) But who defines where the concepts of ownership and privacy starts and end, when our computers become wearable and our consumables are wired in? That's what individuals should do. But with all these other interests out there, will they?