> true, and require some means of hosts being able to change > identifiers at some point in time.
Is that a must have? E.g. one could think of many useful applications where an identifier never changes during a life time of the id bearer. Isn't it the whole purpose of an id after all? Thanks, Peter --- On Tue, 7/8/08, Tony Li <[EMAIL PROTECTED]> wrote: > From: Tony Li <[EMAIL PROTECTED]> > Subject: RE: [RRG] perceived privacy issue > To: "'RJ Atkinson'" <[EMAIL PROTECTED]>, "'IRTF Routing RG'" <[email protected]> > Date: Tuesday, July 8, 2008, 4:03 PM > |% That's been widely discussed during > |% the first iteration of GSE and pretty generally viewed > |% as a Bad Idea. At the very least, there needs to be a > |% mechanism to escape from the MAC address and jump > |% to a separately assigned space. > | > |This concern is misplaced, and there is nothing magic > |here about a MAC as the Identifier. Any Identifier > |will have the same essential properties. > > > Any identifier space is necessarily going to confront the > same issues, it's > true, and require some means of hosts being able to change > identifiers at > some point in time. Obviously, anything associated with > the old identifier > is lost in a change, so there is a non-zero cost involved. > > > |Traffic analysis techniques have been employed > commercially > |since at least the middle 1990s to track users -- even > users > |that change IP addresses often. I understand that such > methods > |continue to be used (and continue to be effective) by a > number > |of firms on the network. Note that these methods can > track > |"users", and are not limited to just tracking > "nodes". > | > |While I don't have a URL to hand just this minute, I > understand > |that recent work at U. Cambridge of late has put > additional > |network traffic analysis methods into the published > literature. > | > |I don't object per se to an "escape > mechanism", but I do NOT > |believe that there is any real privacy benefit to such a > mechanism. > |I feel similarly about the IPv6 Privacy Extensions (sic), > |which have similar levels of (in)effectiveness. > > > While I cannot disagree with you technically, it does seem > to me that those > advocating privacy will definitely raise the issue and that > any eventual > engineering solution will need to provide some mechanism > for addressing > those concerns, misplaced or not. > > Tony > > > > -- > to unsubscribe send a message to [EMAIL PROTECTED] with > the > word 'unsubscribe' in a single line as the message > text body. > archive: <http://psg.com/lists/rrg/> & > ftp://psg.com/pub/lists/rrg -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
