Re: Going forward with zero-config tunneling requirement

[Erik Nordmark <erik.nordmark@sun.com>]

Pekka Savola wrote:

> People who have commented on this seem to have reached (off-list)  
> consensus that draft-nielsen should be revised to be more explicitly
> 3GPP -specific, the 'simple' mode should be removed from assisted
> tunneling requirements, and a new document, discussing the the more
> general 'zero-config' problem space (ISP/unmanaged, possibly
> enterprise) should be written.

I don't understand what this would mean.
I don't think we want a 3GPP specific (simple) tunneling mechanism,
which is incapable to work across v4NATs, while the generic tunneling 
mechanism (which can work across NATs) is prohibited from having a 
simple mode i.e. a mode which doesn't require pre-registration.

But when reading the paragraph above that seems to be what you are 
saying :-(

Did you intend to say something else?


The other big picture thing I don't understand is this space is the 
issue about handling IPv4 NATs.

The 3GPP specific requirements do not require working across a NAT, 
which makes sense. But I see this being used as an argument that 3GPP 
needs a solution which is incapable of working across a NAT, which makes 
no sense at all; if there is a solution which satisfies the 3GPP 
requirements as stated, then whether that solution satisfies additional 
requirements should be ok.

   Erik