Minister promises that Part III is coming

[Owen Lewis]

> -----Original Message-----
> From: ukcrypto-admin@chiark.greenend.org.uk
> [mailto:ukcrypto-admin@chiark.greenend.org.uk]On Behalf Of Ross Anderson
> Sent: 13 May 2006 08:33
> To: ukcrypto@chiark.greenend.org.uk
> Subject: Re: Minister promises that Part III is coming
>
>
> James Davis asked:
>
> > Possibly a daft question but I'm new to the list. Has there been a case
> > of terrorists using encryption? I can't think of any examples off the
> > top of my head, only of terrorists speaking in code.
>
> Yes, one Andrew Rowe recently got sent away for a long time for
> possessing a small piece of paper on which were scrawled a few dozen
> words, and corresponding codewords. You can inspect this desperately
> dangerous cryptographic system for yourself at
>
>   http://news.bbc.co.uk/2/hi/uk_news/4271368.stm
>
> I was asked to give an expert opinion on its fitness for alleged
> purpose, and as list members can imagine I wasn't too impressed
>
> Rowe seemed to his defence team to be a lost soul, a jihad wannabe
> who'd bumbled around the edges of various conflicts. But it's not a
> crime for a UK citizen to work as a volunteer ambulanceman in Bosnia.
> There was no evidence against him that struck me as being of any
> substance. Yet he got sent down for a long time, and the safety
> elephant crowed about it. Maybe they knew something that we didn't -
> say a wiretap showing an actual link with a real bad guy. If so, roll
> on the day when such wiretaps can be adduced in evidence. The
> conviction did not strike me as at all safe, and we are all less safe
> when prosecutors can nail people with a fantastic tale and a few
> shreds of dodgy circumstantial evidence.

This does not seem to be a sufficient answer to James's question.

Truth is that the only people who know the answer to his question are those
tasked with the interception of known and suspected terrorists - and the
terrorists themselves - and none of those is going to step forward here to
satisfy James's not unreasonable curiosity. This leaves him in a position in
which he must assuage his curiosity by considering the matter from first
principles and drawing conclusions. Is it reasonable to presume that, of all
humanity, 'terrorists' are somehow either ignorant of cryptography or else
in some way incapable of using cryptography to hide their dealings? I would
say that either of those conditions is about as probable an occurrence as a
sighting of a squadron of flying pigs.

Relatively few ciphers are broken in the sense that if even a fragmentary
intercept is obtained the meaning imparted in it can invariably be made
plain within a useful timeframe. The burden of evidence from the several
memoirs and other accounts of WWII cipher breaking is that much
cryptanalysis relies on the exploitation of some weakness not essentially of
the cipher itself so much as in the cipher's implementation. Where such
weaknesses are discovered, that information is, in terms of national
security, extremely sensitive. This is so because in almost all
circumstances, if even the existence of a weakness - let alone the precise
form of the weakness - becomes known to the cipher users, they will either
promptly change their cipher or take other remedial action that will result
in a critical loss of information.

Ross makes the point that the admission of intercepts in evidence would be a
good thing. Certainly it might make easier and more certain the matter of
conviction in some cases - as well, of course, as providing both academics
and others with a guide as to the state's capabilities in respect of both
ability to intercept and usefully to process the take. Now, if all intercept
was a straight forward as some textbooks on the subject may suggest - and
the use of codes and ciphers by black-hats was non-existent - why the Crown
should be falling over itself to get those intercepts laid out in court. The
strange thing is though that, despite the apparent keenness of some
politicians, lawyers and civil rights activists that this should happen,
somewhere there is a sheet-anchor holding back progress in that direction.
Why should it matter, if the case in point is only some awful child molester
who, when put away for 30 years, is never going to bother young Caspar
again - let alone touch a cipher? Because the exploitations that caused the
said child molester to be banged up may be the same or similar to those
applied against some terrorist group or even against some nuclear-armed and
failed or rogue state.

Which brings us back to dear RIPA Pt III. This seems to me to make possible
the disclosure of incriminating evidence in court procedures by use of a
method which is essentially straight forward and open and which has no
bearing on sensitive capabilities.


Owen