ICR "Unique Identifiers"

[Andrew Cormack]

Drafting of that definition is awful. From the context of the draft Bill I'd presumed ICRs were full URLs! If NAT/DHCP logs, then why the additional limits on access/use in c47(4)?

BTW, has anyone compared+contrasted clause 51(1)(b) with clause 1(1) of the Comms Data Bill? From a quick look I can't see anything in today's Bill that limits what a "filtering arrangement" might be, so long as it "facilitates the obtaining of communications data". If there really are no statutory limits, that purpose could stretch a *lot* wider than the "API to ISP logging systems" that I've seen mentioned on Twitter. Might a duty to use weak/backdoor crypto, even, be covered by "facilitating the obtaining of communications data"?

Andrew

> -----Original Message-----
> From: ukcrypto-bounces at chiark.greenend.org.uk [mailto:ukcrypto-
> bounces at chiark.greenend.org.uk] On Behalf Of Roland Perry
> Sent: 04 November 2015 16:19
> To: ukcrypto at chiark.greenend.org.uk
> Subject: Re: ICR "Unique Identifiers"
> 
> In article <C5A5B053-5205-4B8B-A581-769A4794B196 at batten.eu.org>, Ian
> Batten <igb at batten.eu.org> writes
> >What?  What?  What is this on about?  I’m guessing it means you
> >need to store whatever the token was that was used to issue an IP
> >number (IMEI, modem MAC, etc), but it’s surely not going to be able
> >to do anyhing about downstream NAT
> 
> It's not a secret that they are trying to "do something" about
> Carrier-Grade NAT, which has always been a characteristic of most GSM
> data access, but is now infiltrating fixed broadband, largely because of
> the exhaustion of IPv4 addressing.
> 
> It's the 2015 equivalent of wanting to know which subscriber had which
> dynamic IP address (at a specific time/date) issued from a dial-up modem
> in the 90's.
> --
> Roland Perry