Time Warner, Broadband Caps, Mark Cuban and ASIVS (That’s DVRs to You and Me)


As Time Warner cable this week begins their trial of tiered Internet pricing in Beaumont Texas, the blogs are aflutter over the various caps Time Warner has proposed.  Time Warner’s plans start with caps at 5 gigabytes and go up to 40 gigabytes. Going over the cap will cost $1 per gigabyte.  Time Warner is also bringing transparency to usage by giving customers a gauge that will allow them to monitor their bandwidth consumption the way cell providers allow you to track your minutes.

Despite all this complaints about Time Warner’s trial and claims its caps are way too low have been ringing around the Internet.

Exactly how much bandwidth do you consume?  It’s hard to say as the number various from user to user.  However, Plus.net put together a nice little graphic showing you what a single Gigabyte gives you – including 4 hours per day of web browsing, 10 song downloads per week, e-mail, Internet radio usage, etc.

What does all that equal?  Well, NCTA member BendBroadband operated with a tiered structure and found that 91% of their customers consumed less than 10GB per month.  BendBroadband found that 99.5% of their users consume less than 100GB per month and now uses that as their cap.

Somewhere above the 91% consuming 10GB per month, and the .5% consuming more than 100GB lies the heaviest Internet users.  Estimates in various studies suggest that 5-10% of Internet users consume half or more of all bandwidth.  Much of that traffic – though specific estimates vary greatly – consists of P2P (peer to peer) connections exchanging files.   A study by SafeNet, Inc. suggests the overwhelming majority of P2P traffic may also be illegal content:

But 90 percent of P2P downloads are still of illegally copied content, according to David Hahn, vice president of product management at SafeNet Inc., which tracks the networks.

Hahn said 12 million to 15 million people are file-sharing across the world at any one time, mainly on the BitTorrent and eDonkey networks. The attraction of file-sharing is not just that it’s free – there’s also content available that can’t be had by legal means, like TV shows that haven’t aired in Europe.

Absent an exact figure of P2P usage, and whether or not you accept SafeNet’s 90% estimate, one thing is undeniable – a small percentage of Internet users are placing a burden on other users.  That is one reason a number of P2P applications providers are working to identify ways to make P2P a better and more efficient means of distributing content.  We believe that is a worthwhile pursuit, which is why NCTA and various cable companies are participating in a “P2P Best Practices” effort led by the Distributed Computing Industry Association.

In many of the articles written about the Time Warner experiment, detractors point to the number of movies than can be downloaded as a specific reason the cap is too low.  An average movie downloaded legally from iTunes is around 1-1.5 GB.  A 40GB cap would allow you to download more than 30 movies per month (or one a day) if that’s all you did.  Most people, however, don’t consume one movie per day, let alone 30 per month.

Mark Cuban, one of the founders of Broadcast.com and a web pioneer, points out the folly of this argument in a post on his blog yesterday.

Its been amusing to read all the blog posts with the math telling all of us just how many standard def or high def movies tiered subscribers will be limited to. You can have 2 or 3 of your favorite SD TV shows per day, or X number of HD movies per month. Say what? 

I have news for all of you that want to dedicate their internet connections to downloading movies. There is a new and exciting development. Its called an Application Specific Integrated Video Service (ASIVS). What is an ASIVS ? Its a computer dedicated specifically to downloading and playing both standard definition and high definition video. You connect it to a network that is dedicated to delivering GIGABITS PER SECOND of high quality video with ZERO buffering. It’s amazing, it always works and connects right to your standard def or High Definition TV, easily. Most of the systems I have seen have a pretty good programming guide and scheduling system and they will let you download AS MUCH VIDEO AS YOU WANT, limited only by the size of its hard drive!!

If you haven’t heard of the ASIVS, its because most people call it a DVR.

If downloading TV shows is so important to you, add a DVR to your cable or satellite service for 5 bucks a month and download all you want. If you want to watch those shows on your laptop, connect the composite video out in your DVR to the composite in on your laptop. Same with movies.

Can’t download movies illegally, tough.

The internet is a great resource for unlimited quantities of video. Downloading video is an internet given right. Using the internet to fill up your PC turned DVR at the expense of the performance of every user around you is not.

Mark’s right on the money with this.  Using the Internet to download video is your right and prerogative.   Using your Internet connection to consume all the available bandwidth and degrade your neighbor’s Internet experience simply isn’t.

As for Time Warner’s caps, are they too low?  Time Warner will soon find out.  They have described this as a test and will determine whether the model works and whether the caps are sufficient.  Unlike many of their critics online, Time Warner is unwilling to pronounce something a failure before even giving it a chance to prove itself.

  • Paul Rodriguez

    Insight’s Michael Willner addressed this same issue this week on his blog and said that:

    …in the past month, 2.1% of our Broadband customers consumed more than 40GB.

  • Moose

    Imagine Comcast imposing a cap of 250 GB per month. That’s the equivalent of 125 SD movies. Either you’re very affluent, downloading illegal content, or running some kind of commercial business if you’re using that much. Any way you look at it, anybody using that much should be paying more. You’re just another person wanting something for nothing, as far as I’m concerned. Grow up and stop being a freeloader. And for Pete’s sake stop whining about it.

  • I am wondering if there has been any mention of download speed standards. What is the point of having X GB per month, if you only get a download speed of say 500Kbps or 1Mbps? In my view the move for paying per bit by consumers should accompany an equal trade off from the ISPs. That trade off is that ISPs can guarantee a certain level of performance, say 10Mbps download speeds.

    One of the underlying instigators for this experiment is the people downloading massive quantities. The massive quantities are hindering network management and ‘degrading neighbour’s Internet experience’. The tiered system is meant to negate these, and with that in mind shouldn’t the consumers be compensated for such a dramatic shift in policy?

    This is a positive step towards solving issues, even if it does not work out, at least one company is attempting to solve a well known issue. The download meter is also good, I had asked for such a tool simply because I have no idea how much data I move in a month.

  • Michael Turk

    Time Warner’s tiers do recognize a difference in speed. Your connection at the 5GB level is 768k down, while at the 40GB level the speed is 15mbps down.

    So there seems to be recognition that people consuming more bandwidth are likely engaged in more bandwidth intensive activities that would benefit from higher speeds.

    Given the rollout of DOCSIS 3.0, which enables top speeds many times faster than even the 15mpbs, it’s possible you may see companies exploring even greater speeds and corresponding higher caps.

  • That is great Michael. And here is the big ‘but’. But will ISPs guarantee a certain level of performance. Consumers are trading pretty much unlimited data downloads and per month prices, for pay-per-byte. What advantage is there for consumers of this new model over the old?


  • Michael Turk

    [W]ill ISPs guarantee a certain level of performance.

    That’s one of the problems this whole thing is meant to address. Unfortunately, with a shared network, it is difficult to guarantee performance because the actions of others using the infrastructure will impact your speed.

    Cable companies are trying to ensure that customers have the best possible experience. They are attempting to address a very small minority of users who currently have a negative impact on others.

    For instance, look at the network in your office. Let’s say you have 10 people on your office network connected through the same Internet connection.

    Ask your IT people what happens to network performance if one of the people in your office begins downloading 3 or 4 one GB files. He’ll tell you that download is going to slow the performance of everyone else in the office. That’s why many companies put restrictions on downloads, the size of file attachments on e-mails, the use of P2P applications and so on.

  • The entire question of caps is less of a QA issue than it is a protective measure to try and keep cablecos in the business of being cablecos. Once they connect you with unlimited bandwidth you can get all the video (and telephone) service you want through the TCP/IP pipe and so the triple-play disappears and they’re a commodity-brokering one-trick pony.

    Meaning, the value to a cableco was they provided lots of channels and we consumers simply couldn’t get lots of channels any other way. We all had like 3-10 channels over rabbit ears, so lots of channels without having to tune the ears/coathanger was great and it was basically “more channels” and “better reception” that sold us. Now that you have Internet TV, there’s no value to the local headend. The only value is in the broadband TCP/IP connection itself, as most all the video you’ll want to watch is scattered across thousands of servers all over the world, not just from your local headend.

    Moreover, traditional TV sucks. It’s 2008, man .. we’ve GOT to be able to fast-forward, rewind, record, zoom, link, index, tag and email the stuff we see. That’s InternetTV and that means the cablecos are simply the most current flavor of ISP. There will be a next wave, however .. look at WiMax and Clearwire. That tech is amazing, potentially a self-scalable mesh .. but it’s just another in a long (and continuing) series of progressive steps towards gigabit wireless atop a sea of 50c microchips.

    I mean, we established the value to the headend content is all but gone with InternetTV, in a post-wimax world, where’s the long-term value to the cable plant? Think about it – telco and cableco equipment is just the same as your PC – everything they buy and bury in the ground or hike up a pole is outdated in 18 months. Having to amortize purchases like that over 8, 10 years just isn’t a winning model in a freely competitive environment. Think about it – look at how much changes in 8 years.

    Bottom line – cablecos are enjoying the triple-play cash windfall now but in the end, they will provide no more than anyone else – TCP/IP, the dialtone of the web. All the content will be from elsewhere. They know this and want to throttle bandwidth NOT to guarantee you better service – no, they want to cap it because they want to keep the TV and telephone services on their OWN menu and not have their customers flocking instead to remote players like Hulu for TV, Vonage for phone, etc.

  • The usual idea is that you would use NFC to set up the link between the two devices and then do an automatic hand over to a different protocol for doing the actual transfer of data – eg Bluetooth,iphone 5