Level 3 blames Internet slowdowns on ISPs' refusal to upgrade networks | Ars Technica

Sun Mar 23 19:04:30 UTC 2014

On Sun, Mar 23, 2014 at 8:06 AM, Blake Hudson <blake at ispn.net> wrote:

> This is exactly my point. If a subscriber can use the service for 30
> consecutive days and never achieve the "8Mbps" because the network is
> incapable by design, or by virtue of its over subscription is statistically
> impossible of delivering it, then I believe this is false advertising. I,
> and most others, accept that when a service is marketed as "up to", the
> service may not always deliver the "up to" number. But if the service is
> marketed as "up to" any number, then the service should at least be capable
> of delivering that advertised number some reasonable fraction of the time;
> Never is not a reasonable fraction of the time.
>
> --Blake
>

So, you want something like EPA MPG ratings, where
empirical, standardized testing is done to validate
manufacturer/vendor claims, rather than just taking
their word for it that the claimed speeds might once
in a blue moon be achievable. with updates to the
claimed performance if subsequent testing fails to
validate the initial claims, such as with the ford c-max
hybrid:
http://yosemite.epa.gov/opa/admpress.nsf/bd4379a92ceceeac8525735900400c27/8a00bfd7633d548f85257bc800637d37!OpenDocument
http://epa.gov/otaq/documents/fueleconomy/420f13044.pdf

Doesn't sound too outlandish.  Mind you, I'm sure
it would raise costs, as that testing and validation
wouldn't be free.  But I'm sure we'd all be willing to
pay an additional $10/month on our service to be
sure it could deliver what was promised, or at least
to ensure that what was promised was scaled down
to match what could actually be delivered.

Thanks!

Matt