Re: Stephen Farrell's Discuss on draft-ietf-6man-stable-privacy-addresses-16: (with DISCUSS and COMMENT)

[Stephen Farrell <[e-mail]>]

On 01/23/2014 01:23 AM, Thomas Narten wrote:
> I have followed, but not commented on the perpass discussion. But I do
> share the concern raised by some that pushback as a result of perpass
> concerns can easily get out of hand. 
> 
> Is this a preview of what's to come?

No. Nothing is out of hand. I asked to discuss Brian's discuss.

> 
> At Wed, 22 Jan 2014 16:10:35 -0800,
> Stephen Farrell wrote:
>> ----------------------------------------------------------------------
>> DISCUSS:
>> ----------------------------------------------------------------------
>>
>>
>> (0) Just modifying my disucss to discuss a part of Brian's discuss:-)
>> I'm sure we'll sort it out easily enough though. I very much do not
>> think that it'd be a good plan to store every address that has been
>> generated using this algorithm. That would be making a
>> privacy-enhancing feature damage privacy. See [1] for an example.
> 
> And can you point me to the text in the draft that suggests doing so?
> (Hint: the draft says no such thing.)

No. And I clearly said that. "to discuss a part of Brian's discuss"
clearly indicates that I am not saying that the draft says that.

> Or for that mattter, who has suggested anyone do this? (Again, AFAIK,
> no one has suggested this.)

See Brian's discuss. If that's not what he meant (and I know
the draft author also thinks it was) then it won't be a problem.
If it is, we'll discuss it briefly and sort it out.

I'll skip the text below. I'm sorry you're upset, but frankly I
don't think your upset is at all justified.

S.

> 
> Are we in hyperbole land? Can we please get back to reality?
> 
> Are we now going to require that all internal logs be scrubbed of
> potentially sensitive material? If we think we are going to stop the
> storage and recording of IP addresses on the devices that are actually
> using those addresses, that is just nuts. Think this through for a
> minute!
> 
> I'm sure the IETF could say the IP stack (as part of trying to make
> network connectivity work better) shouldn't store/cache addresses it
> has used. And maybe as a result the stack will work a little less
> efficiently but it at least won't have stored an address it recently
> used...
> 
> But then we better hope that noone bothers to look in the browser
> cache, or /var/log/messages, etc. which will likely have the same
> information and much much more...
> 
> Thomas
>