Seems like every time I go to a security conference these days there’s at least one short talk where people are proposing to start over and rebuild the computer universe from scratch and make it simple and impossible to use wrong this time and it will be so awesome. Readers, it’s not going to work. And it’s not just a case of nobody’s going to put in enough time and effort to make it work. The idea is doomed from eight o’clock, Day One.

We all know from practical experience that a software module that’s too complicated is likely to harbor internal bugs and is also likely to induce bugs in the code that uses it. But we should also know from practice that a software module that’s too simple may work perfectly itself but will also induce bugs in the code that uses it! One size fits all APIs are almost always too inflexible, and so accumulate a scar tissue of workarounds, which are liable to be buggy. Is this an accident of our human fallibility? No, it is an inevitable consequence of oversimplification.

To explain why this is so, I need to talk a little about cybernetics. In casual usage, this word is a sloppy synonym for robotics and robotic enhancements to biological life (cyborgs), but as a scientific discipline it is the study of dynamic control systems that interact with their environment, ranging in scale from a simple closed-loop feedback controller to entire societies.¹ The Wikipedia article is decent, and if you want more detail, the essay Cybernetics of Society is a good starting point. Much of the literature on cybernetics talks about interacting systems of people—firms, governments, social clubs, families, etc—but is equally applicable to systems of, around, or within computers. One of the fundamental conclusions of cybernetics, evident for instance in Stafford Beer’s viable system model, is that a working system must be as least as complex as the systems it interacts with. If it isn’t, it will be unable to cope with all possible inputs. This is a theoretical explanation for the practical observation above, and it lets us put a lower bound on the complexity of a real-world computer system.

Let’s just look at one external phenomenon nearly every computer has to handle: time. Time seems like it ought to be an easy problem. Everyone on Earth could, in principle, agree on what time it is right now. Making a good clock requires precision engineering, but the hardware people have that covered; a modern $5 wristwatch could have earned you twenty thousand pounds in 1714. And yet the task of converting a count of seconds to a human-readable date and vice versa is so hairy that people write 500-page books about that alone, and the IANA has to maintain a database of time zones that has seen at least nine updates a year every year since 2006. And that’s just one of the things computers have to do with time. And handling time correctly can, in fact, be security-critical.

I could assemble a demonstration like this for many other phenomena whose characteristics are set by the non-computerized world: space, electromagnetic waves, human perceptual and motor abilities, written language, mathematics, etc. etc. (I leave the biggest hairball of all—the global information network—out, because it’s at least nominally in-scope for these radical simplification projects.) Computers have to cope with all of these things in at least some circumstances, and they all interact with each other in at least some circumstances, so the aggregate complexity is even higher than if you consider each one in isolation. And we’re only considering here things that a general-purpose computer has to be able to handle before we can start thinking about what we want to use it for; that’ll bring in all the complexity of the problem domain.

To be clear, I do think that starting over from scratch and taking into account everything we’ve learned about programming language, OS, and network protocol design since 1970 would produce something better than what we have now. But what we got at the end of that effort would not be notably simpler than what we have now, and although it might be harder to write insecure (or just buggy) application code on top of it, it would not be impossible. Furthermore, a design and development process that does not understand and accept this will not produce an improvement over the status quo.

¹ The casual-use meaning of cybernetics comes from the observation (by early AI researchers) that robots and robotic prostheses were necessarily cybernetic systems, i.e. dynamic control systems that interacted with their environment.

Your post advocates a

□ software □ hardware □ cognitive □ two-factor □ other ___________

universal replacement for passwords. Your idea will not work. Here is why it won’t work:

□ It’s too easy to trick users into revealing their credentials
□ It’s too hard to change a credential if it’s stolen
□ It initiates an arms race which will inevitably be won by the attackers
□ Users will not put up with it
□ Server administrators will not put up with it
□ Web browser developers will not put up with it
□ National governments will not put up with it
□ Apple would have to sacrifice their extremely profitable hardware monopoly
□ It cannot coexist with passwords even during a transition period
□ It requires immediate total cooperation from everybody at once

Specifically, your plan fails to account for these human factors:

□ More than one person might use the same computer
□ One person might use more than one computer
□ One person might use more than one type of Web browser
□ People use software that isn’t a Web browser at all
□ Users rapidly learn to ignore security alerts of this type
□ This secret is even easier to guess by brute force than the typical password
□ This secret is even less memorable than the typical password
□ It’s too hard to type something that complicated on a phone keyboard
□ Not everyone can see the difference between red and green
□ Not everyone can make fine motor movements with that level of precision
□ Not everyone has thumbs

and technical obstacles:

□ Clock skew
□ Unreliable servers
□ Network latency
□ Wireless eavesdropping and jamming
□ Zooko’s Triangle
□ Computers do not necessarily have any USB ports
□ SMTP messages are often recoded or discarded in transit
□ SMS messages are trivially forgeable by anyone with a PBX
□ This protocol was shown to be insecure by ________________, ____ years ago
□ This protocol must be implemented perfectly or it is insecure

and the following philosophical objections may also apply:

□ It relies on a psychologically unnatural notion of trustworthiness
□ People want to present different facets of their identity in different contexts
□ Not everyone trusts your government
□ Not everyone trusts their own government
□ Who’s going to run this brand new global, always-online directory authority?
□ I should be able to authenticate a local communication without Internet access
□ I should be able to communicate without having met someone in person first
□ Anonymity is vital to robust public debate

To sum up,

□ It’s a decent idea, but I don’t think it will work. Keep trying!
□ This is a terrible idea and you should feel terrible.
□ You are the Russian Mafia and I claim my five pounds.

hat tip to the original

It’s professional-organization management election time again. This is my response to everyone who’s about to send me an invitation to vote for them:

When it comes to ACM and IEEE elections, I am a single-issue voter, and the issue is open access to research. I will vote for you if and only if you make a public statement committing to aggressive pursuit of the following goals within your organization, in decreasing order of priority:

1. As immediately as practical, begin providing to the general public zero-cost, no-registration, no-strings-attached online access to new publications in your organization’s venues.

2. Commit to a timetable (which should also be as quickly as practical, but could be somewhat slower than for the above) for opening up your organization’s older publications to zero-cost, no-registration, no-strings-attached online access.

3. Abandon the practice of requiring authors to assign copyright to your organization; instead, require only a license substantively similar to that requested by USENIX (exclusive publication rights for no longer than 12 months with exception for posting an electronic copy on your own website, nonexclusive right to continue disseminating afterward).

4. On a definite timetable, revert copyright to all authors who published under the old copyright policy, retaining only the rights requested under the new policy.

Thank you for your consideration.

The ACM held its annual Conference on Computer and Communications Security two weeks ago today in Raleigh, North Carolina. CCS is larger than Oakland and has two presentation tracks; I attended less than half of the talks, and my brain was still completely full afterward. Instead of doing one exhaustive post per day like I did with Oakland, I’m just going to highlight a handful of interesting papers over the course of the entire conference, plus the pre-conference Workshop on Privacy in the Electronic Society.

Continued…

I just presented the major focus of my time and effort for the past year-and-a-bit, StegoTorus, at this year’s ACM Conference on Computer and Communications Security. You can see my slides and the code (also at Github). I was going to explain in more detail but all of the brain went into actually giving the talk. My apologies.

This is an ongoing project and we are looking for help; please do get in touch if you’re interested.

Readers of this blog may find https://github.com/TheTorProject/stegotorus and https://github.com/zackw/moeller-ref of interest.

This day had a lot of interesting papers, but some of the presentations were disappointing: they spent their time on uninteresting aspects of their work, or handwaved over critical details.

That said, most of the work on passwords was compelling, and if you read to the end there’s a cranky rant about the panel discussion.

Continued…

I skipped the 8:30AM session today, it was mostly not interesting to me and I badly needed the extra hour of sleep. I’m sorry to miss On the Feasibility of Internet-Scale Author Identification, but I will read the paper. I also skipped the business meeting, so, summaries start with the 10:30 session, and end with the short talks.

Continued…

I’m attending the IEEE Symposium on Security and Privacy, 2012 and I’m going to try taking notes and posting them here, again. The last time I tried this (at CCS 2010), most of the notes didn’t ever get posted, but I paid a whole lot more attention to the talks than I do when I’m not taking notes. This time, I’m going to try to clean up the notes and post them the next morning at the latest.

S&P was at the Claremont Hotel in Oakland, California for thirty-odd years, and they didn’t really want to leave, but there wasn’t room for all the people who wanted to attend. Last year they turned nearly 200 people away. This year, it’s in San Francisco at a hotel on Union Square—amusingly, the exact same hotel that USENIX Security was at, last August—with much higher capacity, and while I still have to get up at dawn to get there on time, at least I don’t have to drive.

I have not had time to read any of the papers, so this is all based on the talks, only. However, where possible I have linked each section heading to the paper or to a related website.

Mozilla folks: I would like to draw your attention particularly to the talks entitled Dissecting Android Malware, The Psychology of Security for the Home Computer User, and User-Driven Access Control.

Continued…

I haven’t posted anything research-related in a while because I’ve been on a project that I’m not supposed to talk about till it’s done, and it’s not done yet. I can say, though, that it’s about ways to get around country-scale filtration of the Internet. I’m writing it up now, starting with the threat model, as you do:

Alice Arishat wishes to publish things for Brutus to read. Cato does not approve of what Arishat has to say, and seeks to prevent her from publishing anything.

Most online discussion of censorship starts from the premise that Cato is automatically in the wrong here. That’s one of the cypherpunk premises that underpin most discussion of theoretical Internet security. I want to play devil’s advocate today, though, and explore circumstances where we might choose to support Cato. In the offline world, we trade off free speech against all sorts of other values every day:

Continued…