Why do some ISP's have bandwidth quotas?

Mailman List Mirror (Read Only)
1

> Thought experiment: With $250 per megabit per month transit and $30 -
> $50 per month tail costs, what would _you_ do to create the perfect
> internet industry?

I would fix the problem, ie get more competition into these two areas
where the prices are obvisouly way higher than in most parts of the
civilised world, much higher than is motivated by the placement there in
the middle of an ocean.

Perhaps it's hard to get the transoceanic cost down to european levels,
but a 25 time difference, that's just too much.

That's approximately correct. The true answer to the thought experiment
is "address those problems, don't continue to blindly pay those costs and
complain about how unique your problems are." Because the problems are
neither unique nor new - merely ingrained. People have solved them
before.

And about the local tail, that's also 5-10 times higher than normal in the
western world, I don't see that being motivated by some fundamental
difference.

The fundamental difference is that it's owned by a monopoly.

Here in the US, we wrestled with Mark's problems around a decade ago,
when transit was about that expensive, and copper cost big bucks. There
was a lot of fear and paranoia about selling DSL lines for a fraction of
what the cost of the circuit if provided with committed bandwidth would
cost.

The whole Info Superhighway thing was supposed to result in a national
infrastructure that provided residential users with 45Mbps to-the-home
capabilities on a carrier-neutral network built by the telcos. These
promises by the telcos were carefully and incrementally revoked, while
the incentives we provided to the telcos remained. As a result, we're
now in a situation where the serious players are really the ILEC and
the cable companies, and they've shut out the CLEC's from any reasonable
path forward.

Despite this, wholesale prices did continue to drop. Somehow, amazingly,
the ILEC found it possible to provide DSL at extremely competitive
prices. Annoyingly, a bit lower than "wholesale" costs... $14.99/mo
for 768K DSL, $19.99/mo for 1.5M, etc. They're currently feeling the
heat from Road Runner, whose prices tend towards being a bit more
expensive, but speeds tend towards better too. :slight_smile:

Anyways, as displeased as I may be with the state of affairs here in the
US, it is worth noting that the speeds continue to improve, and projects
such as U-verse and FIOS are promising to deliver higher bandwidth to
the user, and maintain pressure on the cable companies for them to do
better as well.

US providers do not seem to be doing significant amounts of DPI or other
policy to manage bandwidth consumption. That doesn't mean that there's
no overcommit crisis, but right now, limits on upload speeds appear to
combine with a lack of killer centralized content distribution apps and
as a result, the situation is stable.

My interest in this mainly relates to how these things will impact the
Internet in the future, and I see some possible problems developing. I
do believe that video-over-IP is a coming thing, and I see a very scary
(for network operators) scenario of needing to sustain much greater levels
of traffic, as podcast-like video delivery is something that would be a
major impact. Right now, both the ILEC and the cable company appear to
be betting that they'll continue to drive the content viewing of their
customers through broadcast, and certainly that's the most efficient
model we've found, but it only works for "popular" stuff. That still
leaves a wildly large void for a new service model. The question of
whether or not such a thing can actually be sustained by the Internet is
fascinating, and whether or not it'll crush current network designs.

With respect to the AU thing, it would be interesting to know whether or
not the quotas in AU have acted to limit the popularity of services such
as YouTube (my guess would be an emphatic yes), as I see YouTube as being
a precursor to video things-to-come. Looking at whether or not AU has
stifled new uses for the Internet, or has otherwise impacted the way users
use the Internet, could be interesting and potentially valuable
information to futurists and even other operators.

... JG

2

.. or try to attract those sorts of content delivery networks into Australia
to serve said content locally and bypass the whole US transit issue.

Of course, the Australian market is so god damned small compared to the American
one, let alone trying to get content providers over to the West Coast when
there's two million people here, and 15+ million over east.

Some ISPs played around with Youtube caching. I won't name names, but there
were more than two of them. Its whats inspiring me to start getting some numbers
on bandwidth saved. What they found is that caching Youtube under test
conditions gave -immediate- traffic savings, but it skewed the TX/RX ratios.
Their inbound dropped but their outbound stayed the same, so their links were
"just as utilised." They then judged it not to be worth it at this time.
Of course, I'd say "can't you invent something to put in place of your now
lower RX utilisation?" but I'm just an Arts student, what do I know about
content? :slight_smile:

Adrian
(Why don't I move overseas again?)

3

> > Thought experiment: With $250 per megabit per month transit and $30 -
> > $50 per month tail costs, what would _you_ do to create the perfect
> > internet industry?
>
> I would fix the problem, ie get more competition into these two areas
> where the prices are obvisouly way higher than in most parts of the
> civilised world, much higher than is motivated by the placement there in
> the middle of an ocean.
>
> Perhaps it's hard to get the transoceanic cost down to european levels,
> but a 25 time difference, that's just too much.

That's approximately correct. The true answer to the thought experiment
is "address those problems, don't continue to blindly pay those costs and
complain about how unique your problems are." Because the problems are
neither unique nor new - merely ingrained. People have solved them
before.

"Address those problems" sounds quite a bit like an old Sam Kinnison routine, paraphrased as "move to where the broadband is! You live in a %*^&* expensive place." Sorry, but your statement comes across as arrogant, at least to me.

> And about the local tail, that's also 5-10 times higher than normal in the
> western world, I don't see that being motivated by some fundamental
> difference.

The fundamental difference is that it's owned by a monopoly.

Bingo. So, how do you propose an ISP in Australia fix the political structure, and do it in a timescale that fits your expectations?

Here in the US, we wrestled with Mark's problems around a decade ago,
when transit was about that expensive, and copper cost big bucks. There
was a lot of fear and paranoia about selling DSL lines for a fraction of
what the cost of the circuit if provided with committed bandwidth would
cost.

The whole Info Superhighway thing was supposed to result in a national
infrastructure that provided residential users with 45Mbps to-the-home
capabilities on a carrier-neutral network built by the telcos. These
promises by the telcos were carefully and incrementally revoked, while
the incentives we provided to the telcos remained. As a result, we're
now in a situation where the serious players are really the ILEC and
the cable companies, and they've shut out the CLEC's from any reasonable
path forward.

Despite this, wholesale prices did continue to drop. Somehow, amazingly,
the ILEC found it possible to provide DSL at extremely competitive
prices. Annoyingly, a bit lower than "wholesale" costs... $14.99/mo
for 768K DSL, $19.99/mo for 1.5M, etc. They're currently feeling the
heat from Road Runner, whose prices tend towards being a bit more
expensive, but speeds tend towards better too. :slight_smile:

I should note that this applies only where the ILEC (or cable company, for that matter) has bothered to deploy service. Unlike telephone service, there has been no "universal service" approach. There are large areas without service other than dialup.

Verizon, it's particularly sad, charges $19.95/month for dialup that'll also tie up a POTS line, where it'll offer the lowest DSL speeds at $14.95. And Verizon "cherry picks" the places where it offers DSL (and moreso for FiOS) so the affluent towns get high speed service, while the rural and poorer places only have available dialup (and that dialup is more expensive).

I would be curious if any of the places in the world with higher-cost high-speed service also have any sort of requirement of coverage?

Anyways, as displeased as I may be with the state of affairs here in the
US, it is worth noting that the speeds continue to improve, and projects
such as U-verse and FIOS are promising to deliver higher bandwidth to
the user, and maintain pressure on the cable companies for them to do
better as well.

Of course this only applies if you live in an inner city or wealthy suburb.

US providers do not seem to be doing significant amounts of DPI or other
policy to manage bandwidth consumption. That doesn't mean that there's
no overcommit crisis, but right now, limits on upload speeds appear to
combine with a lack of killer centralized content distribution apps and
as a result, the situation is stable.

My interest in this mainly relates to how these things will impact the
Internet in the future, and I see some possible problems developing.

Do you believe there is any reason for the "Internet of the Future" to be everywhere? You're concerned about video over IP delivery and other advanced applications, but do you expect to make a video call to your cousin who owns a farm outside of town? This question is largely ignored in discussions about cranking the 'net to ever faster speeds, at least in the US. I'd be interested to know how it's addressed elsewhere in the world.

4

Verizon, it's particularly sad, charges $19.95/month for dialup
that'll also tie up a POTS line, where it'll offer the lowest DSL
speeds at $14.95. And Verizon "cherry picks" the places where it
offers DSL (and moreso for FiOS) so the affluent towns get high speed
service, while the rural and poorer places only have available dialup
(and that dialup is more expensive).

In my experience, the support cost of DSL is significantly cheaper than
dial-up in terms of helpdesk calls. DSL/Cable/FiOS is typically a plug and
play, where as dialup can be quite a bit more troublesome, involving more
tech time in the long run.

5

I occasionally see providers offering pay-per-incident tech support,
particularly for their lower-priced service offerings, as a way to offset the higher support costs as a percentage of overall revenue from those users. I'm not saying it's right or wrong, but it is out there.

jms