The Great Zero Challenge

Nicholas Bohm ukcrypto at chiark.greenend.org.uk
Mon, 08 Sep 2008 11:42:08 +0100


Peter Fairbrother wrote:
> Roland Perry wrote:
>> In article 
>> <3452E045-84B4-4ABA-A319-4D379CA1E44A@sourcetagged.ian.co.uk>, Ian 
>> Mason <ukcrypto@sourcetagged.ian.co.uk> writes
>>> I have the appended signature in my armoury for  replies to those who 
>>> send such ludicrous legalese.
>>
>> I have one too - which seeks to illustrate that if you don't 
>> understand it, how can you comply?
> 
> As I understand it, a disclaimer that a communication is legally 
> privileged may have some legal effect in preventing third parties from 
> divulging it. Once received, the recipient can reveal the contents 
> (unless he's a solicitor under a duty to keep schtumm).
> 
> However if you send me a disclaimer attached to a communication, I can 
> just ignore it - I haven't agreed to keep it secret, and am perfectly at 
> liberty to repeat the sense of the contents, though if eg it's a poem I 
> may not have copyright to the expression, and then I can't reproduce it 
> exactly.
> 
> 
> The last situation is when the contents of an unpriviliged communication 
> comes into the hands of a third party. If it has come by interception I 
> think (haven't checked) it's illegal under RIPA to reproduce it.
> 
> However if it has come into the hands of third party by chance, or by 
> some other means not involving interception, then I'm not so sure. 
> Obviously posting to usenet implies that a communication can be freely 
> reproduced, but in other cases where eg a third party reads a 
> communication marked PRIVATE at the end, does the reader have any duty 
> to keep it private?
> 
> In common courtesy, yes, but under law I don't know. I doubt it, but - 
> Nicholas?

Last time I thought I understood "breach of confidence" (which was 
before it began its recent strenuous morphing into a basis for a remedy 
for breach of privacy), the answer was that marking things "Private" or 
"Confidential" did not by itself impose a duty on the recipient.  It 
might, if consistently used in the course of a continuing relationship, 
especially if used by both parties, provide evidence that a 
"relationship of confidence" had come into existence.  If so, disclosure 
of information exchanged within that relationship, as long as it was 
information of an intrinsically confidential kind, would be a breach of 
confidence.  Marking may also act to designate something as part of a 
class protected under a prior agreement.

In the absence of circumstances of this kind, such markings have a 
fairly limited effect.  At best, I think, an envelope marked "Private" 
or "Confidential" operates as a request to the recipient (perhaps a post 
room in a business) to apply its usual procedures for mail so marked 
(e.g. pass unopened to the addressee).

The firm where I served my articles applied such a policy.  When my 
principal went on holiday, he instructed me to open all unopened mail 
addressed to him.  "Even if marked 'Strictly Private and Confidential'?" 
I asked him.  "Even if written in violet ink and strongly perfumed," he 
replied.  "There are no skeletons in my cupboard."  Sadly, there weren't.

As to the Great Zero Challenge, with which this thread dawned, 
unfortunately absence of evidence isn't evidence of absence.  The 
challenge does at least suggest that data recovery from disks 
overwritten with zeros isn't trivial, but the challenge is hardly 
tempting if such a thing can be done but only with significant effort. 
And I doubt if any three or even four letter agencies would see any 
virtue in demonstrating in public their prowess at recovery.  So I 
prefer to keep going with Eraser (particularly handy for its capacity 
for scheduled wipes of Temp folders etc).  Another comfort is 
TrueCrypt's ability to do full disk encryption (including system disks).

Nicholas
-- 
Salkyns, Great Canfield, Takeley,
Bishop's Stortford CM22 6SX, UK

Phone  01279 870285    (+44 1279 870285)
Mobile  07715 419728    (+44 7715 419728)

PGP public key ID: 0x899DD7FF.  Fingerprint:
5248 1320 B42E 84FC 1E8B  A9E6 0912 AE66 899D D7FF