
Re the Home Office letter:
Paras 6 and 8 seem to confirm the view that Phorm are doing interception as defined under RIPA.
Para 9, I am guessing, applies to the non-processed data from opt-outs. But I don't think it is sound; the filter belongs to the 'person' (Phorm), and even though the person elects to do nothing with it, they could have processed it, so it has been made available to them. You'll notice that Phorm talk about 'our servers' at the ISP, and not about 'our software' on the ISP's servers.
Para 13 makes it clear that *both* ends must consent to the interception, for it to be authorised. So the subsequent OIX use for ad serving is entirely legal. But that is then what the letter goes on to talk about.
Instead, it should be considering the data collection at the ISP; *I* might consent to my end of a session with 'WebHost', but unless WebHost also consents, we have unauthorised interception.
The argument in para 15, for possible implied consent by WebHost, can be rapidly dismissed. Until I contact WebHost, they have no knowledge that a message is coming, and so cannot possibly have consented to its being intercepted unless they have issued some sort of blanket permission for this, in advance; and such permission could hardly be an implied permission.
We then hardly have to consider the second leg, where WebHost reply to me and the communication is again intercepted, without their knowledge. However, if we must, I need only point out that if what WebHost provide is a paid-for, password-protected, service, then the presumption of any implied consent to interception must also fail.
Re paras 16-18, I'd suggest that the lawful interception under 3(3) doesn't apply, as the Phorm data collection is clearly additional to the services needed to provide the ISP service. (Indeed, if it wasn't, then I couldn't be posting here now). And it's stretching the definition to breaking point to interpret it otherwise.
However, if what Phorm are planning is allowable under 3(3), then no sender or receiver permissions would be required. and the recommendation in para 20 would be just that - a recommendation. But it seems clearly wrong that this should be so, and para 20 should be enforceable in law, in my view.
Para 21 remains wrong about being able to assume the implied consent of web hosts. Especially, I would imagine, rival advertising services.
Para 22 I find wrong as well. However, I then have a difficulty in that the spam-blocking service provided by my other ISP, and which I have cheerfully opted into, would also seem to me not to be lawful interception under 3(3). And if not, I very much doubt that the spammers have given their consent, implied or otherwise, under section 2.
Anyone help me square this circle?