Re: [arch-d] FYI: closure of the IAB Stack Evolution program

Le 23/08/2019 à 12:19, Brian Trammell (IETF) a écrit :
> hi Fernando,
> 
>> On 23 Aug 2019, at 12:15, Fernando Gont <fgont@si6networks.com> 
>> wrote:
>> 
>> Hi, Brian,
>> 
>> Thanks for the note! In-line...
>> 
>> On 23/8/19 12:07, Brian Trammell (IETF) wrote:
>>> Greetings, all,
>>> 
>>> During the IETF 105 meeting in Montreal, the IAB decided to
>>> close the Stack Evolution program in its present form.
>>> 
>>> The initial concept behind the IP Stack Evolution program was to 
>>> investigate how changes to the whole stack, especially to the 
>>> transport layer, could be deployed in the big-I Internet, and 
>>> explore the impact of encryption at the transport layer on 
>>> evolution of the stack. At the time the program started in 2014, 
>>> there was not much work yet in and around the IETF in this 
>>> space.
>>> 
>>> That has changed. QUIC has demonstrated that it is possible to 
>>> deploy a new transport protocol at scale, and the QUIC WG has 
>>> nearly finished an IETF standard version of that protocol.
>> 
>> I'm certainly not much of a QUIC expert, but... isn't QUIC
>> tunneled over UDP?
>> 
>> If so, then deployment of QUIC doesn't prove that it's possible to
>>  deploy new transport protocols in the big-I Internet. Actually, 
>> the decission to make QUIC run over UDP would seem to me quite the 
>> opposite: that you cannot deploy a new transport prootocol 
>> natively, and thus need to tunnel it on something that actually is 
>> known to work (TCP or UDP).
> 
> This is a well known discussion rabbithole, but.. no, QUIC is not 
> tunneled over UDP, it uses a UDP header as an encapsulation but is a 
> distinct transport protocol on its own.

Sorry - it makes little sense.

Or maybe we dont understand the same thing by 'being tunneled' over
something.

In my view, once there is an additional header that makes if for being
tunneled.

That additional header could be IP, UDP, TCP or even a Routing Header.

I replied to this email because it popped into my regular inbox, not
being catched by the normal filters.  I dont understand why.

> You're right that we can't deploy new *headers* in the Internet
> because of middleboxes, but a key insight behind QUIC is that you can
> run a new protocol over old headers.

That is nothing new.  There are so many people that run new protocols 
over old headers without even telling about this to other people.

> I would argue that thinking of UDP as a transport protocol is itself 
> a mistake; it is a UNIX sockets model friendly way to allow 
> applications to access an IP datagram service in a multiuser 
> environment. But I think I'm in the minority there.

It is an interesting view of thinking of UDP being more of a UNIX 
sockets model rather than a transport protocol.  In that sense, I also 
see the raw sockets to be more powerful than UDP.  And there is more.

Alex

> 
> Cheers,
> 
> Brian
> 
>> Thanks, -- Fernando Gont SI6 Networks e-mail: fgont@si6networks.com
>> PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492
>> 
>> 
>> 
>> 
> 
> _______________________________________________ Architecture-discuss 
> mailing list Architecture-discuss@ietf.org 
> https://www.ietf.org/mailman/listinfo/architecture-discuss
> 