[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Resolution of my discuss comments for draft-ietf-v6ops-nap-02.txt



Jari Arkko wrote:
tone/language
I re-read the specific examples. Lets go through them one by one.

  Indeed, product marketing departments have
  effectively driven a perception that some connectivity and security
  concerns can only be solved by using a NAT device, without any
  mention of the negative impacts on applications.  This is amplified
  through the widespread sharing of vendor best practice documents and
  sample configurations that do not differentiate the translation
  function of address expansion from the state function of limiting
  connectivity.

I wholeheartedly agree with the second sentence. But the first sentence
focuses on the role of the product marketing whereas in reality I believe
the situation was more complicated. The desires of equipment vendors
got mixed with the desires of the service providers, and with real-world
issues with addresses etc. How about this: "Indeed, it is often claimed
that some connectivity and ..."

I would think that changing the pharase "product marketing" to "hardware companies" without pointing at marketing as the speciffic group responsible. This should maintain the message while doing away with the word marketing that has been causing so much discussion.

  Product marketing departments have widely
  sold IPv4 NAT as a security tool and suppliers have been implementing
  address translation functionality in their firewalls, though the
  misleading nature of those claims has been previously documented in
  [2] and [4].

I don't have a strong opinion on this, but I would use "NAT has been
sold as a security tool ..."

I can live with this suggestion.

<snip>

You appear to be assuming desktop/server OS's. Many/most cell-phone/pda OS's
and virtually-none of the embedded appliance implementations include IPsec
for IPv4. Even when they do they don't include the nat traversal pieces.I
did add the comment about the helper services not working in all situations.


It is true that such devices are less likely to have IPsec, or
other advanced features. (Features tend to be added to
devices when there is a clear, specific function they enable.
Perhaps more so than IETF mandates, even for IPv6.
For instance, corporate VPN access has been a driver for
IPsec inclusion in higher-end cell phone / pda devices.)

Would you be happier with the text with s/IPv4/some IPv4/?

I aticiapte that in the near future the IP excryption options will cause problems as DPI products (and the associtated service featuers like protocol based QoS are demanded) and enforcable wire tapping aer mandated then the concepts of IPSec in mobile will not be such a widley used feature. - Eric