Re: [Nea] IETF67 NEA WG Meeting summary

[Keith Moore <moore at cs.utk.edu>]

> Would you be OK with the following amendment to the first requirement?
> 	NEA MUST NOT expose information about a host to any party other than
> 	the owner of that host or to parties authorized by the owner.

no, I'm not okay with that.  the problem is that people can be coerced 
into giving up their privacy if the protocol makes it easy for them to 
do that.

I'll give an example: HTTP cookies.  they weren't intended to be used as 
a means to track user behavior.  they ended up being used that way 
because (a) they were not designed with due concern for privacy and (b) 
many implementations didn't bother implementing the protections 
specified by the standards.  instead they gave users a yes or no switch: 
"do you want to enable cookies or not?"  and a lot of web sites also 
gave users a yes or no option: "either enable arbitrary cookies or don't 
use our web site".  and the web sites didn't adhere to the rules for 
using cookies either.

what I conclude from the cookie experience is that when the server and 
host don't have the same owner, the server (i.e. the network) can't be 
trusted to not abuse the ability to query hosts for fine-grained 
information if NEA gives it that ability, and NEA client (i.e. host) 
implementors can't be trusted to implement appropriate privacy 
safeguards for clients if the protocol itself isn't designed in such a 
way as to make it unlikely that networks will be able to demand such 
information from clients.

regarding NEA: what I might be okay with is giving authorized 
third-parties yes or no assurance that the host meets or does not meet 
their policies, without giving them fine-grained detail about what is 
installed on the host.  so the owner of a host could get details about 
why a host did or did not fit within a particular network's policy, but 
the network (if owned by another party than the owner of the host) could 
only get yes or no information.   I would like to see this option 
examined further.

Keith


_______________________________________________
Nea mailing list
Nea at ietf.org
https://www1.ietf.org/mailman/listinfo/nea