Re: [RADIR] Last Call on problem statement -01

> Here are my comments.

> 1. In this definition:

> Control Plane Cost:  The overall cost associated with operating the
>        Control Plane.  The cost consists of capital costs (for  
> hardware),
>        bandwidth costs (for the control plane signalling) and any other
>        operational cost associated with operating and maintaining the
>        control plane.  In the present Internet, the primary current cost
>        concern is on the capital hardware."

> The final sentence, "In the present Internet, the primary current  
> cost concern is on the capital hardware", caught me by surprise until  
> I reread it focusing on the context of control plane cost.  Even  
> then, I'm not sure everyone would agree that capital hardware is  
> their greatest concern when it comes to control plane.  It would  
> probably not be hard to create a lively conversation in various  
> forums by putting up the strawman that the opex associated with the  
> current Internet control plane is insignificant.

Hmm.  I worded that sentence pretty careuflly.  If we look at the
whole route scaling problem, are folk raising the issue that "it's too
complex to operate?", and it costs too much in terms of (say) people
and that the cost is rising at an unreasonable or unsustainable cost?

Everything we've talked about so far has been really been about
hardware (even the bandwidth hasn't been cited as an issue).

I not strongly opposed to dropping the sentence. But the reason I put
it in was to make sure we weren't broadening the "cost" to focus in
other areas we haven't talked about yet (i.e, the non-capital costs).
I.e., if the "cost" includes stuff other than the capital, shouldn't
we have a statement in there that puts the concerns in context?

Thoughts?

> 2. Replace "more specific" with "more-specific" throughout except of  
> course if it's not being used to mean "more specific route", or where  
> it's fully written out as such.  (Also if grepping note that it's  
> broken across lines in some cases.)

Done.

> 3. Point 6 of the summary section seems to be redundant with point  
> 1.  Delete it?

done.

> 4. Speaking of section 6, can someone remind me why we didn't think  
> "provides end-to-end convergence/restoration of service at least  
> comparable to that provided by the current architecturexb" was worth  
> including?

Hmm. I recall talking about this point, but now don't remember the
details (or outcome). It may be that we did agree to add it and I
forgot to do so.

The only concern I might have with adding this sort of text is that
I suspect we could have long discussions/arguments about just how fast
things converge today. :-)

Unless someone objects, I'll add it in.

> 5. I made a few trivial proofreading edits to the document and have  
> attached a diff.

done. thanks.

Thomas

_______________________________________________
RADIR mailing list
RADIR@ietf.org
https://www1.ietf.org/mailman/listinfo/radir