Re: [Teas] WG adoption - draft-nsdt-teas-transport-slice-definition

[Adrian Farrel <[e-mail]>]

Hi all,

 

I spent a good chunk of time yesterday and today catching up on the threads about this document. I believe there are several important things to say:

 

1. Many thanks are due to the Design Team which I know has invested a lot of time and thought into working on their small set of documents. Such effort should not go unrecognised.

 

2. Several comments have referred to the consensus that the Design Team reached on the content of the document. It's good that the DT was able to reach consensus - I haven't gone back to determine whether it was rough and how it was measured, and that doesn't matter if they agree it was reached. However, the consensus, of itself is only informational as we consider the document and doesn't carry weight against arguments put forward in this discussion (especially by people who were not active participants in the DT). Let's hear all voices and opinions and not squash them with the DT's collective view.

 

3. I am puzzled by the debate about what to do with text that might be removed from this document. We are debating adopting this draft and don't (at this stage) have control over what goes in other DT documents (or any other non-WG document). If the WG decides that text does not belong in this document, then that is all we decided and people are free to place that text elsewhere: in some other non-WG draft if their co-authors agree, or in a new draft of their own. So let's debate what text we want in *this* document on its own merits.

 

4. The discussion of "isolation" seems to get significantly hung up on the difference between "isolation as part of a service objective" and "isolation as a delivery mechanism". Service objectives are not necessarily directly or easily observable by a client. For example, you might ask your service provider to guarantee that your traffic is not routed via <insert name of country you don't trust>, but when packets are tunnelled and the provider doesn't give you good access to their internal diagnostic traces, it can be less than simple to verify that they are meeting this objective. Thus, absolute requests for isolation are probably meaningless at the service interface and in some network technologies they cannot really be delivered, anyway. What is more important is that we deconstruct "isolation" into the concepts that the client does actually care about: no misdelivery, no observation of traffic by another client, no interference on traffic quality/delivery owing to the activity of other clients. In draft-king-teas-applicability-actn-slicing we tried to tease out these distinctions and believed that the way in which these service features might be delivered are an entirely different matter - if someone wants to write a document called "Delivering Service Isolation Using SRv6" then I think that would be a fine place to discuss such issues.

 

5. Document review and opinion on adoption. That's next on my list 😊

 

Best,

Adrian

 

From: Teas <[e-mail]> On Behalf Of Vishnu Pavan Beeram
Sent: 19 August 2020 16:50
To: TEAS WG <[e-mail]>
Cc: TEAS WG Chairs <[e-mail]>
Subject: [Teas] WG adoption - draft-nsdt-teas-transport-slice-definition

 

All,

This is start of a *three* week poll on making
draft-nsdt-teas-transport-slice-definition-03 a TEAS working group document.
Please send email to the list indicating "yes/support" or "no/do not
support". If indicating no, please state your reservations with the
document. If yes, please also feel free to provide comments you'd
like to see addressed once the document is a WG document.

The poll ends September 9th (extra week to account for vacation season).

Thanks,
Pavan and Lou