Re: [E-impact] [OPSAWG] side meeting #119: Power Metrics: concrete usage example

[George Michaelson <[e-mail]>]

On Tue, Mar 26, 2024 at 1:07 PM Hesham ElBakoury <[e-mail]> wrote:
>
> The solution of the sustainability problems in ICT is just as much sociology and marketing as it is engineering.
> Without also addressing the sociology of the use of e-services we may not be able to maintain or improve overall ICT energy efficiency in the long run when we exhaust all the techniques to improve the energy efficiency of ICT products and services.
>
> Hesham

Not disagreeing Hesham, but I would re-iterate: Where is the activity
which necessarily must be in the IETF to pursue this problem? If the
problem is society at large and marketing, then should we not ask the
sociologists and marketers to look into it? I don't see that the IETF,
or the IAB, have much voice in this.

"wasting energy is bad" -says the IAB, in non-normative language.

Normative MUST, I want to understand where we have the sense that
affects on-the-wire.  A position like "future protocols MUST be
prepared for startup delay to bring high speed carrier live, to
address energy burdens" is way way off. I cannot but think a bunch of
applications space people would be saying "why did we spend all this
time reducing the 53ms perceptual delay fetching the first packet in
VOIP if you are going to impose a 30 second bring-live delay on my
application"

I just don't think we need to solve the societal, or marketing/PR
problem. We need to help people solve substantive sustainability
problems. Finding out if the energy cost to ship bits is better or
worse than the DC cost to house them, the full life cycle cost of
routing and switching and computing cradle-to-grave is way beyond
IETF. I would think we had a slice of this in instrumentation of the
device and protocol. I don't see strong drive into protocol design.

cheers

-George