Richard Bennett, NANOG posting, and Integrity

"We'll allow you to keep these connections
in place as a legacy favour, but as far as the
rest of the world is concerned, they don't
exist; we don't pass routes from it along
to others, and neither will you. They get
used for internal traffic only."

Those types of situations are why traffic
flow data tends to be kept very, very secret.
Every network has its dark corners, its dirty
little secrets that shouldn't see the light of
day. It's easy to make sure those aren't
drawn on the maps released to the public.
It's a lot harder to make sure the presence
of those edges doesn't become visible if
you export actual flow data.

Matt

In fact Netflix is asking to connect to eyeball networks for free:

Yeah, because when I pay UPS on my corporate account to pick up a
package in California and deliver it to me in Virginia, the guy at the
pickup in California is asking UPS to deliver it for free.

Your claim is twisted man. Twisted. I pay Verizon to connect me to
Netflix and the rest of the Internet at substantial speed. Netflix
demands only that Verizon give me what I paid for.

This isn't the traditional understanding of net neutrality, but this is the
beauty of murky notions: they can be redefined as the fashions change:

There is no "traditional" understanding of net neutrality. The term
was co-opted to mean many different things the moment it entered
political awareness, before any tradition could develop.

"You've designed your network to handle the traffic demands of web browsing?
That's cute, now rebuild it to handle 40 times more traffic while I sit back
and call you a crook for not anticipating my innovation."

Right, because how could anyone anticipate that more than a handful of
folks might want to use 5 or 6 mbps of traffic on a 25mbps flat-rate
product for hours at a time. How rude to suggest that an allegedly
high speed network designed only to handle the traffic demands of web
browsing is little different than that age old confidence scheme, the
pig in a poke.

Regards,
Bill Herrin

It's hard to see a revolution when you're in the middle of it. As consumers transition from watching multicast TV on the networks' schedule past time-shifting and on to VoD, the traffic demands on the infrastructure will grow by 25 - 40 times. Similarly, the Internet will shift from a tool for reading web sites and watching occasional cat videos to a system whose main job (from the perspective of traffic) is video streaming. The magnitude of the change will necessarily cause a re-evaluation of the norms for interconnection, aggregation, content placement, and protocol design.

I think it's a mistake to approach this transformation in a "nothing to see here, move along" manner. It's reality that packet networks are statistical, especially at the level of aggregation and middle-mile distribution. The Internet's traditional financial model is one in which infrastructure providers make the most serious investments and edge services extract the highest profits. This model may not be the most sustainable one, and it may not be consistent with supporting the upgrades the infrastructure needs for adaptation to this new application. Alternative models - such as Europe's open access regime - fare even worse in this regard than the vertical integration model that's the norm in North America and East Asia.

I don't claim to have all the answers here, or even any of them, but I think it's important to keep an open mind and pay attention to what works. I'm also not enthusiastic about relying on government programs to upgrade infrastructure to fiber of some random spec, because the entry of government into this market suppresses investments by independent fiber contractors and doesn't necessarily lead to optimal placement of new fiber routes. The First Net experience is proving that to be the case, I believe.

In other words, the Internet that we have today isn't the best of all possible networks, it's just the devil we know.

RB

Astroturfing doesn’t require a fake organization, just fraudulent use of an organization claiming to be grass roots.

I guarantee you that the majority of the communities represented by those organizations probably don’t even understand the issue. Of those that do, I suspect that if you polled them, you’d find most of the not backing the position contained in the document.

Somehow, the anti-internet-freedom collection of monopoly/oligopoly interests managed to coopt the leadership of those organizations into this astroturf.

Owen

So we're supposed to believe that NAACP and LULAC are phony organizations but pro-neutrality groups like Free Press and Public Knowledge that admit to collaborating with Netflix and Cogent are legit? Given their long history, I think this is a bit of a stretch.

It's more plausible that NAACP and LULAC have correctly deduced that net neutrality is a de facto subsidy program that transfers money from the pockets of the poor and disadvantaged into the pockets of super-heavy Internet users and some of the richest and most profitable companies in America, the content resellers, on-line retailers, and advertising networks.

Recall what happened to entry-level broadband plans in Chile when that nation's net neutrality law was just applied: the ISPs who provided free broadband starter plans that allowed access to Facebook and Wikipedia were required to charge the poor:

"A surprising decision in Chile shows what happens when policies of neutrality are applied without nuance. This week, Santiago put an end to the practice, widespread in developing countries <http://techcrunch.com/2014/05/29/twitters-emerging-market-strategy-includes-its-own-version-of-a-facebook-zero-like-service-called-twitter-access/&gt;, of big companies “zero-rating” access to their services. As Quartz has reported <Italian leather iPAQ case, companies such as Facebook, Google, Twitter and Wikipedia strike up deals <http://qz.com/69163/the-one-reason-a-facebook-phone-would-make-sense/&gt; with mobile operators around the world to offer a bare-bones version of their service without charging customers for the data.

"It is not clear whether operators receive a fee <http://techcrunch.com/2014/05/29/twitters-emerging-market-strategy-includes-its-own-version-of-a-facebook-zero-like-service-called-twitter-access/&gt; from big companies, but it is clear why these deals are widespread. Internet giants like it because it encourages use of their services in places where consumers shy away from hefty data charges. Carriers like it because Facebook or Twitter serve as a gateway to the wider internet, introducing users to the wonders of the web and encouraging them to explore further afield—and to pay for data. And it’s not just commercial services that use the practice: Wikipedia has been an enthusiastic adopter of zero-rating as a way to spread its free, non-profit encyclopedia."

When net neutrality backfires: Chile just killed free access to Wikipedia and Facebook

Actually, I don’t see this ruling as such a bad thing.

Internet Freedom? Not so much.

We can agree to disagree. I don’t think leveraging one semi-captive audience to build a captive audience for other companies is a good thing. It reduces the potential for new entrants to compete on an even footing. (Not that there aren’t already plenty of barriers to competing with Facebook and/or Google, but adding cross-subsidies from TPC shouldn’t be an additional one.

Owen

I don't think it's conflation, Joly, since the essence of NN is for the eyeballs to pay for the entire cost of the network and for edge providers to use it for free; isn't that what Netflix is asking the FCC to impose under the guise of "strong net neutrality?" Professor van Schewick is pretty clear about making the users pay for the edge providers in her tome on Internet architecture and innovation.

This is as absurd as the people you shill^wpoopy-head (per your request) for.

The users pay either way.

Either the content provider(s) pay the carriers and then bill the users (at a mark up) or the users pay directly (hopefully without the markup).

We are, after all, not talking about data that Netflix wants to inflict on the unsuspecting user. We are talking about data that the user REQUESTED from Netflix.

Saying “Content providers should pay” sounds great, because it sounds like it gives the end-user a free ride, but the reality is a little different.
Let’s have a look at the unintended consequences of such a policy:

  1. End users get billed more by the content providers to cover this additional cost.
  2. Content providers have to mark up what they are charged by the end-user’s ISPs, and they want to charge a uniform
    rate to all customers, so the most likely result is that they bill end users based on a marked up rate from the most
    expensive eyeball ISP they are forced to pay.
  3. As a result of these additional charges, you create barriers to competition in the content space which begins to turn
    content into more of an oligopoly like access currently is. Its a giant step in the exact opposite direction of good.

Frankly, I give Netflix a lot of credit for fighting this instead of taking the benefits it could provide and screwing over their customers and
their competition.

Competition is a wonderful thing where it can work, but it's not a panacea, especially for the poor and for high-cost, rural areas. Communication policy has pretty much always relied on some form of subsidy for these situations, that's the universal service fee we pay on our phone bills.

How would you know… Let’s _TRY_ it and see what happens? Subsidy for those situations is probably necessary, but so far, subsidy has always been structured to subsidize monopolies and block competition (at the request(demand) of the very people you shill^wpoopy-head for).

If we changed the subsidies a tiny bit so that all subsidized infrastructure was built in a manner open to multiple higher-level service providers (e.g. subsidized open fiber builds to serving wire centers with colocation capabilities) and made those facilities available to all service providers on an equal footing (same cost, same ToS, same SLA, same ticket priority, etc.) I bet you’d see a very different situation develop rather quickly.

Susan Crawford explicitly complains that American ISPs "gouge the rich" by charging more than the OECD norm for high-speed (50 Mbps and above) service, but she fails to point out that they also charge less than the norm for low-speed (15 Mbps and below) service.

Whatever… The bottom line is that overall, throughout the US, even in the most densely populated areas, we are far behind what you can get in places like NL, KR, SG, SE, etc. and paying generally more for it.

I think it's easy to create unintended consequences if you don't look at how specific regulations affect real people, no matter how high-minded and principled they may appear at the surface.

OK, so please tell me what are the horrible unintended consequences of making layer 1 an open platform available on an equal footing to all competing L2+ providers that want to compete? As you point out, most L1 has been built with taxpayer money and/or subsidy, so what’s the horrible downside to letting it actually work or the taxpayers instead of the oligopolistic law firms masquerading as communications companies?

Owen

I don't have much to add to this discussion, but...

Richard Bennett <richard@bennett.com> writes:

I'm also not enthusiastic about relying on government programs
to upgrade infrastructure to fiber of some random spec, because the
entry of government into this market suppresses investments by
independent fiber contractors and doesn't necessarily lead to optimal
placement of new fiber routes. The First Net experience is proving
that to be the case, I believe.

People will eventually come to rely on the Internet as a critical piece
of infrastructure. And many already do. Provisioning service and
routing packets needs to be separated from provisioning physical access
in any form. If the governments need to step in to do the latter, I'm
happy for them to do so as long as it falls under some lattice of
framework similar to the public utilities commission. So that the
localities responsible for maintaining the infrastructure are compelled
to act responsibly.

Or if you *really* want to be in the business of owning infrastructure
on a commercial basis, your business should be wavelengths, not packets.

In other words, the Internet that we have today isn't the best of all
possible networks, it's just the devil we know.

-Daniel

In fact Netflix is asking to connect to eyeball networks for free:

http://blog.netflix.com/2014/03/internet-tolls-and-case-for-strong-net.html

" Strong net neutrality additionally prevents ISPs from charging a toll for interconnection to services like Netflix, YouTube, or Skype, or intermediaries such as Cogent, Akamai or Level 3, to deliver the services and data requested by ISP residential subscribers. Instead, they must provide sufficient access to their network without charge.”

Which is as it should be… There’s no reason $EYEBALL_ISP should get to double-bill both their customer and Netflix et. al. for the same traffic.

There are a few possible cases…

<USER><EYEBALL_ISP><CONTENT_PROVIDER>

In this case, USER is paying Eyeball ISP and the costs are minimized.

<USER><EYEBALL_ISP><PUBLIC_EXCHANGE><CONTENT_PROVIDER>

In this case, USER pays Eyeball ISP and EYEBALL_ISP and CONTENT_PROVIDER pay PUBLIC_EXCHANGE (minimal fee usually) and
costs are still relatively small.

<USER><EYEBALL_ISP><TRANSIT_ISP><CONTENT_PROVIDER>

In this case, USER pays Eyeball ISP and CONTENT_PROVIDER pays TRANSIT_ISP. Since both ISPs have been paid by their respective customers, there shouldn’t be any need for money to change hands between TRANSIT_ISP and EYEBALL_ISP.

This is the most expensive case for CONTENT_PROVIDER and possibly USER.

In all of the above scenarios, EYEBALL_ISPs costs are very similar. There’s really no valid reason for EYEBALL_ISP to attempt to extort money from CONTENT_PROVIDER in order to deliver packets requested by USER who already pays them.

No matter how much you spin this or how many times you try to contort it to argue that CONTENT_PROVIDER should be forced to subsidize USER’s service from EYEBALL_ISP, the argument just doesn’t hold water if you actually analyze it.

This isn't the traditional understanding of net neutrality, but this is the beauty of murky notions: they can be redefined as the fashions change: "You've designed your network to handle the traffic demands of web browsing? That's cute, now rebuild it to handle 40 times more traffic while I sit back and call you a crook for not anticipating my innovation.”

It seems pretty close to the traditional understanding of net neutrality to me. A neutral network requires that Network A doesn’t try to jack Network B for payment to deliver packets requested by users paying Network A.

However, I realize that these facts interfere with your role as a shill^wpoopy-head, so obviously you can’t accept them as in any way legitimate.

Owen

Richard,

Before Netflix it was Bittorrent. Before Bittorrent it was Usenet.
Before the Internet, history records no shortage of companies willing
to falsely advertise a product that did less than was claimed. Nor is
fraudulent double-billing a recent invention.

There is nothing new under the sun, no matter how much you may protest
otherwise, and every one of these eyeball networks sold products
which, on paper, were consistent with the use of Netflix. Without
requiring additional payment beyond the customers' subscriber fee.

And continued selling the product as described, long beyond any
reasonable doubt their customers expected it to work with Netflix.
Right through this very minute and beyond.

Regards,
Bill Herrin

It would be amusing to see Netflix just call their bluff. And maybe donate some
lawyers for the inevitable class action lawsuit for false advertising against the eyeball
networks. I imagine other self-interested 900lb gorillas might join the fun too.

Mike

Owen, your mother should have told you that you need to play nice if you want the other children to play with you.

This is a self-fulfilling prophecy that reflects the intense conservatism of a certain part of the Internet establishment. I'm inclined to go for new services, new norms, and progress. But that's just my personal bias, not a law of nature.

RB

Second verse, same as the first. A little bit louder and a little bit worse.

Regards,
Bill Herrin

I pay for (x) bits/sec up/down. From/to any eyecandysource. If said
eyecandy origination can't handle the traffic, then I see a slowdown,
that's life. But if <$IP_PROVIDER> throttles it specifically, rather
than throttling me to (x),I consider that fraud.

I didn't pay for (x) bits/sec from some whitelist of sources only.

I think we've seen the first shots of this battle fired already -- Netflix
was putting up notices saying "your video is crap because $ISP is congested"
for a little while. I expect that wasn't the last we'll see of that kind of
tactic.

- Matt

Hey, just wait until the eyeball networks decide
they can charge different amounts depending
upon their view of the morality of the content
being sent...

#engage_fly_on_wall_of_boardroom_mode

"OK, let's see...Netflix traffic, they get charged
$2/mb extra, because they show adult situations
and brief nudity. Pornhub show explicit material,
but it's mostly boobs and butts, so we'll look the
other way, and only charge them $4/mb to get
past the choke point (because there's no such
thing as a fast lane with QoS, there's only normal
and "be glad we didn't throw it *all* on the floor").
Oh my...doublefistingdudes.com...we don't like
the idea of naked dudes getting it on over our
wires...for them, it's $100/mb if they want their
bits to make it to our users. Guess they'll have
to jack the price of their content *waaay* up.
*sound of high fives all around* "

#end_fly_mode

Hey, if they don't have to be neutral about it,
why not enforce their morality through differential
pricing, while they're at it?

We could even have differential pricing based
on days of the week.

"Oh, you want to send your movies to our users
on the holy day, when they should be praying?
For that privilege, it will cost you 10x what it
does on any other day, for you are luring our
users into vice and depravity."

That "whitelist" must be sounding pretty darn
tempting to some executives right about now.
Forget about censoring content on the internet
that they don't like...they can just bill arbitrarily
high rates to let it get through. Price it high
enough, and nobody will watch it anymore, and
they can go to bed happy.

Matt
getting ready to start a mail-order DVD service
that doesn't charge extra based on what you want
to watch...

Remembering the things (which had to do with network operations) that I go banned for.

One wonders why I felt bad about it,

NANOG = NANAE us a slur on NANAE.

> > It's more plausible that NAACP and LULAC have correctly deduced that
> > net neutrality is a de facto subsidy program that transfers money
> > from the pockets of the poor and disadvantaged into the pockets of
> > super-heavy Internet users and some of the richest and most
> > profitable companies in America, the content resellers, on-line
> > retailers, and advertising networks.
>
> I've got to say, this is the first time I've heard Verizon and Comcast
> described as "poor and disadvantaged".
>
> > Recall what happened to entry-level broadband plans in Chile when
> > that nation's net neutrality law was just applied: the ISPs who
> > provided free broadband starter plans that allowed access to
> > Facebook and Wikipedia were required to charge the poor:
>
> [...]
>
> > Internet Freedom? Not so much.
>
> I totally agree. You can't have Internet Freedom when some of the
> richest and most profitable companies in America, the content resellers,
> on-line retailers, and advertising networks, are paying to have eyeballs
> locked into their services. Far better that users be given an
> opportunity to browse the Internet free of restriction, by providing
> reasonable cost services through robust and healthy competition.
>
> Or is that perhaps not what you meant?

I think he meant the actual poor people that broadband subsidies and free
walled garden internet to access only fb and Wikipedia are supposed to
benefit, but I could be wrong

I've got a whopping great big privilege that's possibly obscuring my view,
but I fail to see how only providing access to Facebook and Wikipedia is (a)
actual *Internet* access, or (b) actually beneficial, in the long run, to
anyone other than Facebook and Wikipedia. I suppose it could benefit the
(no doubt incumbent) telco which is providing the service, since it makes it
much more difficult for competition to flourish. I can't see any lasting
benefit to the end user (or should I say "product"?).

FYI it's Bharti-Airtel, not an incumbent, but a multinational GSM operator.