[178526] in North American Network Operators' Group
Re: Verizon Policy Statement on Net Neutrality
daemon@ATHENA.MIT.EDU (Owen DeLong)
Sat Feb 28 04:34:48 2015
X-Original-To: nanog@nanog.org
From: Owen DeLong <owen@delong.com>
In-Reply-To: <54F188D3.5090205@seacom.mu>
Date: Sat, 28 Feb 2015 01:29:39 -0800
To: Mark Tinka <mark.tinka@seacom.mu>
Cc: nanog@nanog.org
Errors-To: nanog-bounces@nanog.org
> On Feb 28, 2015, at 01:22 , Mark Tinka <mark.tinka@seacom.mu> wrote:
>=20
>=20
>=20
> On 28/Feb/15 10:51, Owen DeLong wrote:
>> Competition? What competition? I realize you=92re not in the US,...
>=20
> Yes, I know competition in the U.S. is not where it ought to be :-).
>=20
> My comment was more global, as we all use the same technologies around
> the world, even though you do get varying levels of market conditions =
as
> such.
>=20
>> so perhaps there is some form of meaningful competition in Mauritius.
>=20
> I am based in South Africa, which isn't saying much.
>=20
> The .mu domain throws everyone off :-).
>=20
>>=20
>> There is no such thing in the US. It=92s oligopolies at best and =
monopolies at worst.
>>=20
>> We have, unfortunately, allowed the natural monopoly that exists in =
infrastructure (layer 1) to be leveraged by private enterprise to form =
an effective monopoly on services.
>=20
> I'll continue to postpone my immigration to those unions :-).
>=20
>> The point here is that adequate up and adequate down are not =
necessarily defined by having them be equal. Yes, you get better uplink =
speeds on symmetrical technologies. That=92s sort of inherent in the =
fact that asymmetrical technologies are all built for higher downstream =
speeds and lower upstream speeds.
>=20
> I agree.
>=20
>>=20
>> My point is that in the vast majority of cases, a hardware limitation =
where the downstream is faster than the upstream is not inappropriate =
for the vast majority of content consumers. The problem is that in most =
cases, consumers are not given adequate upstream bandwidth, regardless =
of the size of their downstream bandwidth.
>=20
> This is where I disagree, because we are making the case for (the vast
> majority of) customers based on the technologies they/we have always =
used.
This is where I disagree with you.
Look at it this way=85 I bet even you consume far more content than you =
produce. Everyone does. It is the nature of any one to many =
relationship.
We consume content from many sources. We are but one source of content.
Even if everyone produced the same amount of content, mathematically, =
you=92d be consuming more than you are producing if everyone consumed =
everything.
If you have an example of any concept of an application where an =
end-user is likely to need the same amount of bandwidth upstream as they =
do downstream, I=92m all ears. Your first example utterly failed. Do you =
have a better example to offer?
> We have seen what can happen to GSM networks when you put a smartphone
> in the hands of an ordinary Jane. Not even the mobile operators saw =
that
> one coming.
Even phones consume asymmetrically and almost entirely down-stream.
> Let us open up the uplink pipes and see what happens. If we keep on
> thinking that the patterns will always be the way they are today, the
> patterns will always be the way they are today.
I=92m all for bigger uplink pipes, but insisting on symmetry is absurd.
>=20
>>=20
>> If you had a good solid 256Mbps up and 1Gbps down, I=92m betting you =
would be a lot less upset about the asymmetrical nature of the circuit. =
Even if you continued to complain, I think you will admit that the vast =
majority of users would be quite happy. I know I would and I=92m pretty =
upstream-heavy for the average residential user.
>=20
> Yes! I would be very happy with that if it were reasonably reliable, =
or
> degraded in a way that would at least leave me reasonably happy.
Not sure what you mean by =93degraded in a way that would make you =
happy=94.
>=20
> Symmetric circuits significantly reduce the likelihood of degradation =
on
> the uplink more than asymmetric circuits do. So an asymmetric service =
on
> a symmetric network is more likely to perform better than any service =
on
> an asymmetric network. Ultimately, that is my point.
This makes no sense whatsoever.
Owen
