Catching up on this thread (message times are based on NZST, UTC+12):

On 13/06/2014 00:45, Markus Stenberg wrote:

> This sounds _way_ too specific to me. 

I agree. Discussion of adding metrics together, although
it seems like Routing 101, just seems out of place.

On 13/06/2014 01:59, Ted Lemon wrote:

> the understanding I came away with was that they felt that the 
> document as written was quite open-ended and left available the 
> possibility of doing some things that are Not A Good Idea.

That would be a very valid criticism of WG charter goals. I am
not at all convinced that it's a valid criticism of an Informational
architecture document.

On 13/06/2014 02:34, Juliusz Chroboczek wrote:

>    The Arch document MUST NOT specify what kind of data the metric
>    computation algorithm is allowed to take as input.

and, I think gave authoritative reasons for that assertion. Basically,
we shouldn't squeeze these very complex arguments into a general
document.

On 13/06/2014 02:55, Acee Lindem wrote:

> I was involved in this discussion and the statement was merely an attempt
> to capture the fact that the existing unicast IGP routing protocols would
> meet the homenet requirements with the addition of routing based on
> source-address.

That's a very fair statement, but that isn't at all what the new paragraph
says.

On 13/06/2014 03:27, Ted Lemon wrote:

> "Existing routing protocols" covers a lot of protocols that I think 
> we clearly don't want to use.

Yes, but that's very different from defining a rigid approach to metrics.

On 13/06/2014 09:20, Ray Hunter wrote:

> Apart from the use of the word "graph" which could be overloaded
> or suggest a preference for link state protocols.

Why? It's the term of art for both Bellman-Ford and Dijkstra algorithms.

   Brian






_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to