On Wed, Oct 02, 2002 at 02:37:23PM -0500, Mark J Roberts wrote:
> Concern about whether or not nodes might cache plaintext is misguided, and
> the answer is irrelevant, but I'll give you mine anyway. I can suggest no
> way for a node to verify that a document it caches is not meaningful. An
> image can probably be made statistically random yet still legible, thus
> frustrating the best test I can imagine.
> 
> That answer is a waste of your time and mine for a frustratingly obvious
> reason. If a node operator is responsible for verifying that files in his
> cache are legal to possess or distribute (and this is, after all, the
> premise of our debate), he'd have a far easier time doing it by checking
> their search keys against a blacklist.
> 
> Encryption of the whole datastore and routing table is popularly held to be
> a way to hide those troublesome search keys from the operator. Simply not
> printing them would do that just as well, and, in any case, you'll have to
> climb rather farther down the proverbial rabbit hole before any of that
> nonsense should relieve him of whatever responsibilities he might have.
> 
> A few people will explain that potentially incriminating evidence may be
> more easily destroyed by encrypting it and wiping the key. Unfortunately,
> there would almost certainly be more evidence outside the datastore than in
> it. Personally, I ascribe the phenomenon to encryption fetishists who should
> themselves be encrypted perhaps six or seven feet beneath the earth. They'd
> probably even endorse the idea.
No, sadly, the common case is that a node op gets busted for having
something illegal in his datastore, which he did not download. Therefore
the only evidence is _in the datastore_. Simply because the best thing
to bust a node op for is something that is relatively uncommon (no
prizes for guessing what sort of JPEGs I'm referring to here).
-- 
Matthew Toseland
[EMAIL PROTECTED]
[EMAIL PROTECTED]
Freenet/Coldstore open source hacker.
Employed full time by Freenet Project Inc. from 11/9/02 to 11/11/02.

Attachment: msg04092/pgp00000.pgp
Description: PGP signature

Reply via email to