Anna Lysyanskaya, Host

   Michael Black, IPP Co-Director 

   John Savage, IPP Co-Director
   Nikos Triandopoulos, Coordinator

   Trusted Computing Group: Goals, Achievements, and Controversies

   Thursday, March 25 2004, at 8:30 AM

   Computer Science Department, Brown University

   Lubrano Conference Room, 4th floor, room CIT 477

   Providence, RI

This one-day event is open to our partner companies and academic institutions.  If this applies to you, contact Suzi Howe to register. The information we need is name, affiliation, contact information, and dietary restrictions.
NOTE: You must register; space is limited. The deadline for registration is Friday, March 19.

If this does not apply to you, but the event sounds exciting, we want to hear from you anyway, and we will try to find a way to invite you as well! Please write to Suzi Howe, telling us your name, contact information, and information about your affiliation and why you are interested in attending the Symposium.

The Trusted Computing Group(TCG) is "an industry standards body, comprised of computer and device manufacturers, software vendors, and others with a stake in enhancing the security of the computing environment across multiple platforms and devices" (quoted from the TCG webpage).

The Trusted Computing Group (TCG) is an industry group dedicated to enabling trust and security in computing systems. The goal of TCG is the design, development and promotion of open industry standard specifications for trusted computing. The task involves the design and deployment of both crypto-enhanced hardware building blocks and secure software interfaces, across multiple platforms, e.g., PC's, servers, PDA's and digital phones. TCG is an open alliance that is formed by several companies. Almost a year ago, it replaced the Trusted Computing Platform Alliance (TCPA), adopting all the TCPA specifications ready at that time.

TCG aims at an increased level of trust and security for any computing devices, where all involved parties (users and consumers, information technology organizations, computer, device and software suppliers) share the benefits of such a trusted framework. These benefits include more secure data storage, online business transactions and remote access, and lower risk of identity theft. Based on these open standards, information technology organizations will be able to develop new secure system. TCG specifications include protection of users' privacy and individual rights.

Brown University is not affiliated with TCG.

First of all, TCG is important because computer security is a matter of growing importance. Every-day life is increasingly affected by disruptions caused by system vulnerabilities. Computers are attacked by malicious intruders and viruses. Everyone encounters these risks both at home and at work, because networked PCs and other devices are used by everyone, everywhere. Creating a real, practical, industry-wide fix to all such vulnerabilities is clearly much needed.

Second, TCG is important because it is an industry-wide development affecting most major computer companies specializing in all aspects of computer technologies: from hardware, to operating systems, to application software development. TCG is a cooperative effort among all these players to design and standardize a framework that would provide solutions towards more secure computing. This initiative is expected to have a dramatic effect on the technologies that these companies will develop in the future.

Some of these companies are Brown University's Computer Science Department IPP partners (Hewlett-Packard,IBM, Microsoft,Sun) as well as other major companies (e.g., AMD, Sony Corporation,Intel).

Since TCG is extremely important, it has attracted a lot of attention and made a lot of people think about these issues and raise interesting concerns.

One important concern was privacy. This comes up when a third party wishes to verify that a given computer is enhanced with the TCG technology. Doing this in a way that revealed nothing about the identity of a computer in question appeared a challenging problem; luckily recent advances in cryptography made it possible to resolve it! Version 1.2 of the TCG specifications successfully addresses these issues.

Another concern which is essential to this debate is the need to balance a user's right and desire to control his machine with the need to protect the machine from potential security problems an untrusted application run by a user may create. Central to this is the fundamental question of who controls a given device: its owner, or the manufacturer?

In turn, if the manufacturer assumes some of the control over a given device, yet another series of questions follow. For example: how do we make sure that computer manufacturers do not get an unfair advantage in locking the users into certain applications? And how does free software come into the picture? The question becomes more complicated where maintenance is concerned.

Yet another, unrelated, concern is this: what happens if after all this work the goals of TCG are not achieved, and TCG-based systems still have vulnerabilities? Then at the end of the day matters only get worse, since now people will have the expectation of security without actually getting it.

Brown University does not have a position on any of these issues; we simply list them above for your convenience. For more information, here is a collection of interesting links: EFF TCG links, an article at News.com, Ross Anderson's TCG FAQ page.

The main reason that TCG has created controversy, is that its goals are very ambitious, it is an extremely important development and as such attracts a lot of attention. The goal of our IPP Symposium is to present these issues in a constructive manner, in a research-oriented environment. By a series of talks by distinguished researchers from academia and industry, the symposium will address the trusted computing effort. Specific topics that will be discussed include: goals, challenges and controversies of trusted computing, and component hardware, software, and privacy technologies. At the end of the talks, a panel discussion will follow, where the panelists will discuss these issues and engage the audience in the discussion as well. We hope to have a constructive meeting that will give everyone some food for thought.

The tentative agenda:

• Joe Pato from HP Labs will talk about the role of trusted computing fundamentals in creating diversity for appliance based systems and also examine the controversy around the Trusted Computing Group and how to preserve diversity in secure environments from a variety of perspectives.
• Brian LaMacchia from Microsoft will give a talk about the Next-Generation Secure Computing Base project.
• Ernie Brickell from Intel will give a talk about the cryptographic functionality of the Trusted Platform Module 1.2, describe the cryptographic protocols used for this functionality and their purpose and implementation.
• Jan Camenisch from IBM will give a talk about the direct anonymous attestation scheme (DAA) for preserving privacy of users of the TCG platform.
• Anna Lysyanskaya from Brown CS will talk about research challenges: how can fundamental research in Computer Science help the TCG efforts?
• At the end of the day, we will have a panel discussion on the technologies and controversies of TCG that will involve all the speakers as well as members of the Brown faculty.