Re: [tae] [tsv-area] Transport negotiation

Stuart Cheshire wrote:
> On 9 Dec, 2008, at 19:01, james woodyatt wrote:

<snip>

> It's a testament to the IETF's extreme antipathy to NAT that we refused 
> to layer new transport protocols like SCTP and DCCP over UDP, as if we 
> imagined that creating new protocols that didn't work with NAT would be 
> the thing to convince the rest of the world that NAT was evil. It was as 
> if we imagined that in the resulting battle for consumer mindshare our 
> new transports would win and NAT would lose. We sacrificed SCTP and DCCP 
> on the altar of anti-NAT fanaticism.
> 
Not so. At least for DCCP, this seems to be re-writing history:

DCCP made a design decision (in an IETF design review, I think, on the 
basis of NAT experience) to not use a stronger checksum and to position 
the 16-bit DCCP checksum and port fields in places where they would 
EXACTLY match those of UDP.

The decision to use the UDP checksum algorithm was largely motivated by 
the desire to allow incremental header checksum update.

SCTP made different design choices, but at least for DCCP, the 
NAT-traversal is not much different to those protocols that already 
traverse NATs.

> If SCTP and DCCP were just user-level library code that any application 
> writer could use freely, without waiting for support from the OS vendor, 
> NAT gateways, or network, they'd be much more widely used today. This 
> strange notion that layering transport protocols over UDP is a gross 
> hack has got to go.
> 
I disagree. This was a conscious decision, see DCCP problem statement 
(RFC4336).

> Layering new transport protocols over UDP is not a hack. It's actually 
> the right architecture AND (because of that) it also happens to work 
> though NAT gateways too.
> 
I am not convinced this is a good long-term decision.

> Stuart Cheshire <cheshire@apple.com>
> * Wizard Without Portfolio, Apple Inc.
> * Internet Architecture Board
> * www.stuartcheshire.org
> 
> _______________________________________________
> tae mailing list
> tae@ietf.org
> https://www.ietf.org/mailman/listinfo/tae
> 
Gorry Fairhurst

_______________________________________________
tae mailing list
tae@ietf.org
https://www.ietf.org/mailman/listinfo/tae