RE: Topic IPv6

["Tony Hain" <[e-mail]>]

> From: ietf [mailto:[e-mail]] On Behalf Of Alexander Nevalennyy
> Sent: Monday, November 21, 2016 7:45 AM
> To: [e-mail]
> Subject: Re: Topic IPv6
>
> Does anybody want to help me with writing Internet-draft?
> It saves emotions and time of a lot of engineers. 
> I will be glad if somebody help me. Maybe, my idea is not great, but have a chance as other ideas.

Alexander,

The reaction you have been getting on this list is due to people forgetting that not everyone has been working on IPv6 for 20+ years. If it were simple enough for you to dig through the archives of all the conversations on an array of lists over that time you would find that ideas on your list have been discussed multiple times in many different contexts. 

One issue you might be having with generating interest in creating a draft is that you have not clearly stated your goal. You listed some technical approaches as a set, which might imply certain goals, but depending on which country and set of laws that are being assumed, those are likely to be different for each person reading them.

To try and bring some focus to this, based on your list I will throw out a goal that might be where you were heading---
Goal:
Improve law enforcement's ability to track the economic terrorists that abuse the Internet and its related technologies to attack others.

That may not be what you had in mind, but it is an example of one possible goal. The feedback that will immediately arise for something like the above is that it is in direct conflict with someone else's goal of making sure that 'free speech' voices in countries with oppressive governments have a way to use the Internet without fear. 

The IETF has taken a position that its job is creating technologies and that it is someone else's job to deal with the political implications of those. 

All that said, you might be interested in an expired draft:
https://www.ietf.org/archive/id/draft-hain-ipv6-geo-addr-02.txt

While the goal of that doc was minimizing renumbering events for smaller organizations when they change providers, it has the side effect of providing ~6 meter targeting for the packet source (excluding use of ToR technologies). It also removes the political arguments about which organization 'controls' address prefix allocations by pre-allocating space in a consistent global standard. 

On your point about telephony mapping, most of the people on this list will tell you that despite the use of numbers rather than letters, the current implementation of telephone numbers acts more like names and the conversation will immediately redirect you to DNS. 

Finally, your point about SLAAC appears to assume that the device MAC address is used. While that was the original definition for implementation simplicity, it was always assumed that other sources would be used over time and that the only real requirement is that the IID is unique on the local media. To the point of privacy, it was immediately recognized that a fixed IID would allow nefarious web sites to track a device's movements throughout the day, so SLAAC includes a privacy extension which would appear to be in direct conflict with your other bullet points. Further, if you look at current implementations, SLAAC may result in what appear to be pseudo-random static values for the IID. These are generally based on things like the CPU or graphics chip serial number so they are somewhat consistent over time, but they  do not leak information about which nic vendor an organization has chosen. While those implementations may still appear to point back to a specific device, the OS vendor would have to provide the algorithm to reverse the IID. Even with that or nic derived MAC, the courts would likely throw out any attempt to use that as evidence because I could watch the LAN to see what IID your device was using, then use that IID somewhere else masquerading as you, and you would then have to prove you were not at the other location. See RFC 3971 & 2 as an example mechanism for increasing IID authenticity to limit MAC spoofing. The point is that if your goal is to allow law enforcement to track to a specific device, you need to state that. Then be prepared for the onslaught of privacy advocates that will argue their goals are more important than yours.


If you have a clear goal, and can show that it requires changes to the protocols, then I would recommend taking that to the ipv6 WG. If your goal simply requires operational practice documents, then take that to the v6ops WG. If your goal requires global treaty negotiations, the IETF is the wrong venue...

Tony