Re: [tsvwg] Adoption call for draft-white-tsvwg-l4sops - to conclude 24th March 2021

[Jonathan Morton <[e-mail]>]

> On 25 Mar, 2021, at 3:08 pm, Ingemar Johansson S <ingemar.s.johansson=[e-mail]> wrote:
> 
> One reflection around all this is that it really does not matter how many
> warning signs you put, information may still not reach the intended
> audience. 

Which, I think, just confirms that we have to consider what happens to operators and users who don't receive or act on that information.  How serious is the problem that is likely to result, how many people is it likely to affect, and are those people reasonably likely to be able to troubleshoot the problem and apply a suitable remedy?  Hence why I highlighted the principles of risk analysis in my talk.

That in turn is why it's much easier to standardise and deploy protocols that explicitly try to fit in with existing deployed specifications.  In that case, the answers to those questions are benign, and the protocol can be deployed without too much in the way of special safeguards - often, just an agreement to shut down experimental deployments if they prove to cause unexpected problems.

Unfortunately, the answers where L4S are concerned are *not* benign, because it is not designed to interoperate with a certain class of existing deployed infrastructure.  It therefore requires much more robust precautions surrounding any deployment - precautions which l4s-ops does not presently describe.

 - Jonathan Morton