[LINK] Cloud computing and privacy

[Richard Chirgwin]

Roger Clarke wrote:
> At 19:06 +1100 25/2/09, Jan Whitaker wrote:
>   
>> Does Cloud Computing Mean More Risks to Privacy?
>> http://bits.blogs.nytimes.com/2009/02/23/does-cloud-computing-mean-more-risks-to-privacy/
>> But one recommendation seems to stand out as the most prudent: 
>> “Don’t put anything in the cloud you wouldn’t want a competitor, 
>> your government or another government to see.”
>>     
>
> That's ambiguous.  (Credit where it's due - maybe intentionally so).
>
> Some refinement could make it clearer, and a very useful statement.
>
> We can't write off crypto altogether.
>
> (Okay, that's said by someone who spent some years demolishing 
> digital signatures.  But the issue was the practicality question not 
> the maths.  Crypto protections are easier than actually creating 
> public key infrastructure, and providing a basis for bit-based trust 
> of another party).
>
> So ... 'Don't put anything in the cloud **whose content** you 
> wouldn't want [a spouse / lover / employer], a competitor, your 
> government or another government to see'
>
> I don't mind someone seeing the bit-strings that represent my 
> innermost secrets and most embarrassing medical conditions.
>
> Provided that I have a great deal of confidence that the CIA / Mafia 
> would have to focus a great deal of their computing resource on 
> little increasingly old me in order to crack the key and be able to 
> see what the bit-strings meant.
>   
Hmm. As the terms & conditions bore of the list (question: why do so few
people care about T&Cs?), I'd say privacy is only one threat from cloud
computing.

Others:
We own what's posted
We aren't responsible if you lose everything when there's a failure
We can sell whatever we please to a third party
We can change the T&Cs at any time just by notifying you

If cloud computing doesn't offer security, ownership, privacy,
continuity and robustness, then it's snake-oil. It's just bureau
computing with the important bits removed.

RC