Will Netflix Destroy the Internet? | Hacker News Reader

Will Netflix Destroy the Internet?

slate.com

23 pointsedw51922 comments

It's more likely to save the Internet by forcing infrastructure upgrades rather than letting things stagnate forever with the justification that only pirate need more bandwidth.

So in your opinion cable companies (and phone companies that are now also offering competing television services) will upgrade their internet offerings so you can better enjoy Netflix, a service that makes their primary offering less valuable?

It's pretty clear from the incentive structure here that it's going to have the exact opposite effect of what you suggested. It's going to make them institute bandwidth caps and/or hike prices.

Cable companies will probably start first by caching items regionally, and using that regional cache. They already do this on some level, but I don't think on the 4gb movie level.

I agree though - they will be competing directly against companies like netflix. Maybe they'll come out with their own franchised netflix style streaming system?

It's called TV Anywhere.

Maybe Netflix will partner with Google to use some of its local storage.

Then some entity somewhere will decide to get off its ass and offer internet service the way PG&E supplies power -- with no conditions. If you're an ISP then you have no business competing with other internet services; it's time to spin off a new company.

I don't disagree with that in principle, but I'm confident it won't happen in the US. (I don't know enough about Canada to have an opinion.)

Cable companies and telcos won't willingly give up their internet business, especially since a spun off ISP with large amounts of bandwidth to their customers' homes could trivially compete with them. They have a lot of lobbying power and they'll use it to fight that. Whenever you're dealing with infrastructure like roads or utility lines, progress moves at the speed of half-frozen molasses even when the existing stakeholders are on board, and it takes very little for them to halt it entirely.

Instead what they'll likely do is find ways to align their interests with Netflix's. They'll institute bandwidth caps as a first measure. Then they'll perhaps sell you the Netflix service with bandwidth for some monthly fee that gives them both a cut.

The less optimistic part of me agrees with you but hopes that in the future this will somehow turn out to be wrong.

Yeah I hope so too, in fact I often find myself thinking something is true and hoping I'm wrong. But as a betting man, when it comes to the intersection of government and enterprise you have to give me pretty good odds for me to wager on the side of the consumer.

The good news is, in 5-10 years wireless internet will bring to ISPs the sort of competition that satellite TV has brought to cable I think.

That's fine, can we take back whatever government subsidies they've received as a result of this potential outcome?

You're correct. In Canada, Rogers appears to have lowered its bandwidth caps in anticipation of Netflix's entry into the Canadian Market: http://www.dslreports.com/shownews/Rogers-Lowers-Caps-Even-F...

Agreed. This is a sensationalist headline to a non-problem.

What surprises me though is that Netflix is pushing so much data given how poor their streaming library is. I'm a Netflix subscriber and I watch a few a month (the kids watch it much more), but just imagine how much of the Internet traffic they're going to account for when they are finally able to stream their entire catalog (which is what they want to do)...

The TV streaming library is pretty great though. And a season of TV is much longer than a single movie.

I still can not fathom the thinking that leads to this:

"This was Netflix's first venture outside of the United States, and because the company wasn't offering its traditional DVD-by-mail plan to Canadians, its prospects seemed questionable. How many people would pay $7.99 per month (Canadian) for the chance to watch Superbad whenever they wanted?

A lot, it turns out."

PROSPECTS SEEMED QUESTIONABLE?! Really, because people wouldn't be able to wait at their mailboxes for a whole day waiting for a tiny sip of media and instead can instantly watch a huge library... how does this make sense? It would be like if Google had 2 options, snail-mail you the results printed on a piece of paper or normal website google, and everyone was fretting that they couldn't get the snail-mail option.

Netflix's streaming library isn't very good and it was questionable whether people would pay $8/month for it. So far people in the US have been getting streaming "free" in addition to regular Netflix.

Netflix has some pretty decent tech-chops, so they probably took a good look at the number of discs subscribers were getting and comparing that to the amount of on-demand used to figure out the percentage of subscribers who were close to 100% on-demand only and going from there.

It isn't like it they just blindly said, "hey let's try this." It could also be the writer trying to add some drama to an otherwise straightforward part of the story.

I will add a little more detail. When Netflix starts sending enough data through standard connections that they start to cripple the internet, they will set up peering agreements and run direct line to the front door of every ISP. If you don't think that is possible look at Google who has peering agreements with everyone except tier 1. We will be fine, the sky isn't falling and no, this is not the end of the world. Not even close.

Peering reduces load on the backbone, but not on the middle or last mile which is where I think congestion is predicted to appear. Only the broadband ISPs can upgrade there.

Articles like these make me want to say, "No, of course not, next question." Economics works well in cases like this: people will pay more for working service, which makes it viable for companies to charge more for the service if they need to do so to keep it working. Or Netflix itself will pay by collocating caches near users and not using backbone bandwidth as much. Either way, it's certainly nothing to stress out about.

In articles like this we always hear how expensive and slow our broadband is.

Are there any good sources comparing the real cost of broadband internationally?

When I was a kid in Germany you had to pay a fee to have a phone and local calls were not free. You also had to pay a fee to have a TV or radio and there were no private stations. Although both the phone and broadcast industry have completely changed I'd like to know what the real total cost is once you add up fees and taxes for countries with larger government involvement in daily life than the US.

Of course our local monopolies act similar to the government run in other countries and sometimes do a poorer job of it.

I think what's interesting about this article is not the whole "Netflix will destroy the Internet" scenario, but the numbers about the relative size of Netflix vs. Bittorrent traffic (which is generally, though incorrectly, presumed to consist of pirated video). It certainly sheds a new light on what companies like Comcast have been shouting for so long about Bittorrent traffic. Even more interesting, from a business perspective, is that paid content can apparently overtake (supposedly) pirated content in no time.

Will Netflix Destroy the Internet? American broadband capacity might not be able to keep up with everyone who wants to stream movies.

Blogga, please. If it can't, then (as others have pointed out) the infrastructure will have to grow. It should have already -- it's just that companies are busy protecting their monopolies.

Reply on news.ycombinator.com