Re: [TLS] Consensus Call on MTI Algorithms

[Nico Williams <[e-mail]>]

On Fri, Apr 03, 2015 at 12:28:09AM +0300, Yoav Nir wrote:
> I thought the whole point of IoT (as opposed to proprietary protocols)
> was that it *did* interact with the Internet. But that aside, “the
> web” as represented by the HTTP/2 document made its own profile. For
> applications other than “the web”, UTA made their BCP document. For
> the IoT DICE made their profile. 

I'm not entirely sure that I want my things interacting with the web.  I
may want to interact with my things across the Internet.

IoT needs something of a definition, yeah?

> So everyone has their own MTI algorithms, some of them contradictory
> to what we are proposing to recommend. So who takes their MUSTs and
> SHOULDs from TLS as opposed to their own niche document?

IoT seems rather special.  Ignoring IoT, how many profiles of TLS do you
think we'd need for the protocols and non-IoT devices we have now?

IoT's definition, scope, requirements, ... -- we're going to need to
pin these down.  As for *this* thread, I think we should factor IoT out.

Nico
--