Re: ULE Security Requirements : NiTs on text

[Gorry Fairhurst <gorry@erg.abdn.ac.uk>]

Thanks Sunny for doing all those edits, and responding to my questions.

You specifically asked for clarification on:

<snip>

Section 3.2
-- perhaps a forward pointer to a later section that ³masquerading and
modification² are more difficult on a specific link, than they are in the
general internet, and will be described more in section 3.

//Sunny

any suggestions ?????

GF - I was thinking here that a key driver in the design of IPsec, and 
one that is often stated by our security ADs is that one must protect 
from masquerading and modification of IP packets, this being 
comparatively easy in an Internet environment. So, my suggestion was to 
agree with such arguments, but then to state for link-layer security fro 
broadcast links that you note such attacks are in fact MUCH harder.
- This could for instance motivate the use of sequence numbers in IPsec, 
but not the mandatory use of them on synchronous links.


<snip>

and on:

<snip>
4) I would like to see more references to methods used by DVB/MPEG-2 
network architectures ...
* How is the proposal positioned against technologies from say the IEEE 
(for 802.14; 802.16).

//Sunny

//Any suggestions ?????

<snip>

- I don't know the answer... does anyone else?


I now suggest that you make a new revision of the draft and upload to 
the web site (I can do this for you), so we can see the interim version, 
and see if Michael and others can see if they have anything more to add 
at this stage.

Gorry