Comment by mike_hearn | Hacker News Reader

Comment by mike_hearn | Hacker News Reader

Parent Full thread

Maybe but not necessarily.

The flip side of your view is that there are absolutely cases that are real, which don't even make it to the press at all. I was an expert witness for a terrorism case in the UK - the guy was convicted - and nothing about it ever surfaced in the media.

The reality is that there are a stream of people in the UK who try to carry out terrorist attacks, and who are stopped by the police. Attempting to argue against a bad policy by claiming terrorists are establishment propaganda is likely to be a bad strategy as a result.

A much better approach is to ask how many of these terrorists are really using sophisticated cryptography, and how many successful attacks would have been stopped if not for encryption? And there we find the answer is "not many" and "essentially none".

There is a great article on that very topic, written by a British journalist who also has acted as an expert witness in many terrorism trials:

http://privacy-pc.com/articles/how-terrorists-encrypt-threat...

It looks at many cases of busted terrorist attacks over many years, and examines the involvement of cryptography. The conclusion is that the intersection of terrorists and sophisticated users of encryption is the empty set. The closest you get is a groupie who worked on things like propaganda and funding, but who wasn't involved in any attacks themselves.

Now that article was written quite a few years ago and I suspect the new attitude of companies like Facebook towards encryption has changed the game somewhat, WhatsApp end to end encryption (assuming it's really on for everyone) makes it much easier to protect conversations than before so, it would stand to reason that cryptography does foil terrorism investigations more often than it used to. However, we don't know that, and the IC was yelling about the danger of cryptography for decades already - certainly in the time frame that Duncan Campbell's analysis was written in.

In conclusion, I'd focus more on whether real terrorist plots are happening successfully because GCHQ couldn't hack things fast enough, than on whether terrorists exist at all.

Interesting. This paper takes a different view.

https://ctc.usma.edu/how-terrorists-use-encryption/

I would be very surprised indeed if the intersection of front-line terrorists and users of industrial encryption was an empty set.

I think it's more likely the intersection of caught and prosecuted terrorists and users of industrial encryption is an empty set - or at least a much smaller set than those who use FB Messenger to coordinate attacks.

This is not an argument for backdoors. I suspect the real inefficiencies in monitoring don't come from lack of evidence, but from lack of efficient data processing and flagging.

The two papers discuss some of the same cases.

In particular your paper discusses the "Tadpole" program developed by Rajib Karim to communicate with Al-Awlaki, albeit it doesn't refer to it by that name. It's interesting to see how there are different spins on the same event.

http://privacy-pc.com/articles/how-terrorists-encrypt-7-pecu...

Both papers point out that: Police described his use of encryption as “the most sophisticated they had seen in a British terrorist case.”

In the talk by Campbell, Tadpole is described as amateur hour. It's literally a Caeser cipher implemented using Microsoft Excel, with the results copied into password protected Word documents. Campbell observes that even a very rudimentary intelligence agency would be easily able to break this code without access to any of the underlying materials ... in fact, the technique for breaking such a cipher was first described by an Arabic mathematician over a millenium ago. This was used in preference to the "Asrar" PGP GUI that was circulating amongst jihadis, because it wasn't clear to Karim that Asrar was really trustworthy. Was it an NSA plant? This problem crops up all the time with jihadis trying to use strong encryption: they can't implement it themselves, they don't trust western apps and struggle to verify the origins of programs claiming to be written by fellow jihadis.

Overall Campbell treats Tadpole as a joke: a textbook study in why terrorists+encryption are not anything worth worrying about.

In the West Point paper you link to, the same program is described in quite different terms. It's described as an "intricate system", an "unorthodox and complex technique based on cipher codes and passwords stored on Excel spreadsheets" that produced "end to end encryption". It says "Western intelligence agencies were not able, as far as is known, to intercept any of his communications in real time". The West Point author appears to be under the impression that the only mistake Karim made was not wiping his laptop in time, which allowed police to access the underlying spreadsheets he was using.

This is a fascinating study in how the capabilities of terrorists are sometimes exaggerated to build the case for all-backdoors-all-the-time. Tadpole wouldn't have stopped a clever teenager with access to some intercepts, let alone an intelligence agency as sophisticated as GCHQ. Yet it is being used as evidence of fundamental shifts that require deep social and policy changes.

Thanks for such a detailed insight.

Westpoint paper in this case appears to be clearly hiding an agenda. Don't courts have some sort of checks and balance to minimise this type of influence from expert witnesses?

Cross examination.

Comment of the year. Absolutely wonderful.

> I was an expert witness for a terrorism case in the UK - the guy was convicted - and nothing about it ever surfaced in the media.

Can you give us a rough idea why that trial was keept a secret from the public?

I don't believe there have ever been fully secret trials in the UK. In camera sessions with secret evidence have been held in a handful of terror cases. The more likely explanation is just that nothing was being kept secret, and the trial wasn't deemed newsworthy.

That's right. I don't think there was any gag orders or anything. The trial wasn't secret. I certainly didn't have to sign anything committing me to secrecy, which is why I'm willing to mention it now. My contact at the police just asked me not to blog about the details, as a friendly courtesy. It just didn't make it to the press. My guess is that court case -> media transmission is not 100% reliable as we might imagine.

The police wanted it kept quiet for the usual reasons. There's some amount of luck involved in foiling any crime, especially before it happens, and when the details of a trial are broadcast it necessarily implies teaching future criminals how their predecessors were caught. The police don't like that, it just makes their jobs harder.

Well, there were the "Diplock courts". But yes: it does seem that the newspapers have no interest in reporting actual convicted terrorists. Which is very strange given how they sensationalise the threat of terrorism.

Not at all strange - convicted terrorists convince people that the threat is being met. The threat of unknown terrorists keeps everyone anxious and biases towards granting more police powers in order to curb those unmet threats.

PsychOps 101.

Umm, I see quite a lot of reporting of it. Its just on page 7 usually since it comes down to a bloke being arrested and being found guilty of conspiracy to commit acts of terrorism. Not very newsworthy.

One of my practices is to periodically use a day or two off work to attend court. I've sat in various sessions of a magistrates' court (low seriousness offences, shoplifting, maybe a drunken fight at most) and the conveniently adjacent crown court (rape, fraud by impersonation, lots of drug dealing). I never saw a reporter even once. The courts are set up to allow a local journalist to cover them, but there's no money in it unless the defendant is famous. Usually I was the only person who wasn't essential to the mechanics of the court, in the fraud trial and some of the magistrates' stuff there were family there to support the defendant but never anybody else. Your friends and indeed family are probably not going to take the day off work to attend your rape trial. So that's worth knowing...

I am intrigued. Are the sessions publicly listed or do you attend them more or less randomly?

https://www.thelawpages.com/court-hearings-lists/crown-court...

There are listings an adjacent post links but I don't pick and choose. The staff at a court will tell you e.g. that Court #4 is done for the day, or it's doing a bunch of calendar management which you might find tedious. I guess you'd want to plan at least somewhat if you have to travel a long way to attend, for me it's just across the city.

Reply on news.ycombinator.com