On The Importance Of Unfettered Broadband
from the it-goes-beyond-just-movies dept
A little over a year ago, I tried to explain why metered broadband or capped broadband slows down innovation. However, many have responded that as long as the basic caps are reasonably high, it shouldn’t be a big deal, because, “how much bandwidth do you really need?” People note that the only ones who might be using up the 250GB caps announced from Comcast can only be downloading non-stop movies. But that’s not really the point. The real question is what innovations are we not seeing because of limited bandwidth. Not so long ago, the very idea of something like YouTube was preposterous, but thanks to abundant bandwidth it became possible. But something like YouTube is just a tiny way down the path to what’s possible.
Tim Lee has a great post making the point that it goes well beyond just “how much bandwidth does a single person need” or even looking at what specifically they’re downloading, to recognizing the change in tradeoffs for creating applications if bandwidth is effectively unlimited. As he notes, in any engineering situation, there are resource tradeoffs. If you’re building an application, there are tradeoffs to making a client side app vs. a web-based app, for example. However, if bandwidth is truly abundant, the very nature of those tradeoffs change and it allows for entirely different types of development, often in ways that are difficult to fathom right now.
Lee gives a few random examples of what unlimited bandwidth might allow as the tradeoffs change:
People with cable or satellite TV service are used to near-instantaneous, flawless video content, which is difficult to stream reliably over a packet-switched network. So the television of the future is likely to be a peer-to-peer client that downloads anything it thinks its owner might want to see and caches it for later viewing. This isn’t strictly necessary, but it would improve the user experience. Likewise, there may be circumstances where users want to quickly load up their portable devices with several gigabytes of data for later offline viewing.
Finally, and probably most importantly, higher bandwidth allows us to economize on the time of the engineers building online applications. One of the consistent trends in the computer industry has been towards greater abstraction. There was a time when everyone wrote software in machine language. Now, a lot of software is written in high-level languages like Java, Perl, or Python that run slower but make life a lot easier for programmers. A decade ago, people trying to build rich web applications had to waste a lot of time optimizing their web applications to achieve acceptable performance on the slow hardware of the day. Today, computers are fast enough that developers can use high-level frameworks that are much more powerful but consume a lot more resources. Developers spend more time adding new features and less time trying to squeeze better performance out of the features they already have. Which means users get more and better applications.
The same principle is likely to apply to increased bandwidth, even beyond the point where we all have enough bandwidth to stream high-def video. Right now, web developers need to pay a fair amount of attention to whether data is stored on the client or the server and how to efficiently transmit it from one place to another. A world of abundant bandwidth will allow developers to do whatever makes the most sense computationally without worrying about the bandwidth constraints.
So the question is not how much bandwidth does any person really need. It’s how will the entire ecosystem of what we can do change when bandwidth is completely abundant?
Filed Under: broadband, broadband caps, tradeoffs
Comments on “On The Importance Of Unfettered Broadband”
The real purpose of the caps
I still think the real purpose of the caps is to reserve high bandwidth applications like HD movies to the broadband supplier. Download your movie from someone else and you hit the cap. Download the movie from us (or use the cable company owned by us) and we will wave the cap.
Re: The real purpose of the caps
“wave the cap”
You’re a right funny one, you are.
Agreed...
I think Anonymous Coward hit the nail on the head… if, for example, Comcast did not have a PPV service, I bet you would not see the imposition of caps. It’s clearly a conflict of interest and should be challenged.
Comcast is the predominant bully in the cable & internet arena; being huge, they figure they can do as they want (like when they throttled P2P traffic and got caught). Letting a cable company control internet traffic is a mistake; let them provide the pipeline (the wires) and let someone else manage them… someone who does not stand to lose money when more people download or stream media.
By the way, Comcast has also wormed its way into Clearwire, the new Sprint/Clearwire venture that will handle WiMax rollouts, so they’ll impose the same limits on that technology as well.
Tim Lee's argument won't sway the SPs
The Service Providers are looking at their bottom line for the next quarter and possibly next year. What TL doesn’t do is connect the unlimited bandwidth argument to increased profits 2 or 3 years down the road. Until the SPs see that the long term strategy of unlimited bandwidth will ultimately improved their profits in the longer term, SPs will continue their capping policy
I don’t think innovation is necessarily the right argument to use against bandwidth caps. While a major drawback will be the stifling of innovation in technology, one could also argue that the limits imposed could also be a boon to technology as developers, while spending more time developing more efficient programs, can pass on that knowledge and help develop faster, more powerful applications that take up less bandwidth and less time as an end result. So, in essence, either method would not be better or worse than the other.
Rather, what I think would be the better argument to focus on primarily is that by allowing ISPs to have unfettered control of bandwidth would be giving ISPs the ability to have unfettered control over many other technologies, as they could hold bandwidth at ransom until those that need the bandwidth give in to the demands that ISPs would certainly take advantage of. Stifled innovation just being a side effect of greed, but it’s not necessarily the case that that would happen at all.
Shift to a Demand-side economic view
It’s how will the entire ecosystem of what we can do change when bandwidth is completely abundant?
As the internet is continually expanded upon for things Vincent Cerf imagined, people will expect it to be held together with more than magic, pixie dust and handshakes.
If the providers can’t provide, it will create new pressures and probably become regarded as a true utility- on the lines of water, sewage and electric.
How does this happen?
Well, as people move to commercial VoIP solutions, there will be a spike in unfettered bandwidth with QoS. Unfortunately, SIP is a real complex turd of a protocol that was never meant to be used in a residential setting, but hey, it’s here. So by enacting bandwidth caps, all it will take is one person not able to call for emergency services on their VoIP phone to get some sort of action.
But none of this will happen if it’s approached from a demand-side economic view, and the powers that be start laying more fiber to increase bandwidth and stop investing in toilet drain projects like that God awful TIA, which apparently does little beyond humoring linguistics officers and fighting sniffles.
The money spent on that would be better spent in laying fiber, adding redundancy, and just overall– infrastructure.
Michael's arguement only looks at it from the consumers's POV
I don’t disagree with Mike’s points, but to take ejs’ point a little further, nothing in Mike’s argument makes the case for why an ISP would want to continue to invest in increasing their network capacity when there’s no clear model to improve their revenue and/or profitability.
If the role of the ISP becomes similar to that of other public utilities, then they in effect become a commodity supplier. (Note: one definition of a commodity is anything that is primarily differentiated by its price.) As a marketing person, I can tell you that is NOT where any good marketing person wants to land up. You want to bundle other value-added services into packages that hopefully give you a unique and compelling offering in the market. Just offering a big fat pipe with no caps at a fixed monthly price is NOT appealing to the Comcasts of the world, even though I as a consumer would love to have just that.
I am not trying to defend the ISPs, just trying to point out how it looks from their point of view. I use Comcast as my ISP and would love to drop them given their recent decision to add caps, but the only other option available to my home is Qwest at a much slower (too slow) speed at nearly the same price. I use Vonage for both of our voice lines. I am probably not one of Comcast’s more profitable customers.
Caps
Maybe Netflix and companies alike should ban together to support a new ISP that will never impose caps. As long as enough companies join in support, the bandwidth shall remain limitless!! Okay… I have awaken from my dream now. Caps suck.
Gates' own example
“how much bandwidth do you really need?”
“no one will need more than 640k of memory. . .”
Unlimited vs. efficient
A world of abundant bandwidth will allow developers to do whatever makes the most sense computationally without worrying about the bandwidth constraints.
Unfortunately, that’s how you end up with bloated pigs instead of useful, efficient software. All you have to do is look at your own computer — does all the new bloatware, requiring 4GB of memory and a terabyte of disk, really do that much more than the applications of the past, that just a few years ago required only a few hundred MB of memory (or less) and a 100MB disk? The applications became bloated because developers no longer paid attention to the efficient use of resources (memory, disk and CPU). Microsoft codes for the next generation of hardware, so that their developers don’t have to think about fast or efficient. App too slow? Get a faster processor. Get more memory. Get a bigger hard drive.
If you allow that line of thinking with Internet apps (“So what if our application transfers 2 petabytes of data — bandwidth is unlimited!”), we’ll be right back where we are now — inefficient, poorly-implemented bandwidth-sucking applications, and ISPs that will again find the need to implement caps.
Unlimited bandwidth is not the answer. Efficient use of bandwidth is.
This is a direct parallel to the auto industry. Just because gas was cheap and plentiful was not a good reason to completely ignore fuel efficiency, but that’s what the auto makers did, and look where that got them.
Re: Unlimited vs. efficient
We are not talking about software efficiency here… we’re talking about content. Digital content can only be compressed so much before it loses a lot of quality. If you download an HD movie from iTunes, do you want it so compressed that it will be pixelated on your TV? You can only do so much to streamline content downloads.
At what point does Internet Access become a public utility?
Other public utilities such as water, gas, electricity are deemed to be essential to our lives, require large capital investments, and often only 1 provider is available to the average consumer. For those and other reasons these utilities are regulated.
While I am usually not a fan of letting the government make technology decisions, as internet access continues to become more essential to our basic lives and also becomes more of a commodity, at what point would it make sense to treat it as a public utility? Or is internet access somehow fundamentally different than the the other utilities and hence should not be regulated?
Re: At what point does Internet Access become a public utility?
The issue, at least with Comcast’s cap, is that they have no provision whatsoever for “overages”. You go over the cap, they cut off your service for a year. In any other “measured” utility, you get “x” amount of minutes, for example, and pay so much for overage. Not in Comcast’s case; so even if you want to use more, you can’t.
Also at issue is their former use of the term “unlimited”. If you purchase a VoIP phone service, you get unlimited local and long distance, meaning that you can make as many calls as you want at any time you want. Comcast interprets the term “unlimited” as 24/7 access, not unlimited usage. So, in their jaded thinking, they are still offering “unlimited” access, when they are not.
unlimited resources causes bloatware
i prefer designing embedded systems rather than computer software because i have/get to be efficient and elegant in my code. when you only have 640k you can’t afford bloat.
part of why windoze has such a big footprint i think is because the devs know they have the space to spread out. everyone has 3ghz processors and 500gb harddrives. the devs don’t have to write tight efficient code because the builders will just throw enough hardware at the slug to make it fast.
Re: unlimited resources causes bloatware
I think we’ve strayed from the topic… AC has only used Bill Gates’ original statement as a euphemism for shortsightedness. We all know today that Vista will not run well with less than 2 GB of RAM, yet Gates’ original prediction was based on MS-DOS systems. Likewise, the ISPs do not foresee any growth in content providing (or view it as a threat) and are putting artificial limits on that growth.
Innovation not limited only by bandwidth
Some of these innovations that we are not seeing for the consumers have other limitations besides bandwidth.
“People with cable or satellite TV service are used to near-instantaneous, flawless video content, which is difficult to stream reliably over a packet-switched network.”
Some of this innovation needs to take place at the ISP or higher level. The innovations are limited by protocols and infrastructure in addition to bandwidth. Truly innovative solutions would not necessarily need more bandwidth if they just use it more efficiently.
Free Gas, Free Water...Why Not?
Mike,
We’d also see much more innovation in cars and other internal combustion vehicles if gasoline were flat rate. Should we force gas to be free to release the latent innovation?
My lawn would look better if the damned city just made water flat rate. I could water constantly, and have elaborate fountains. Just think of the innovations in fountain design!
There are constraints in this world. As an economist, you should appreciate that resources are not unlimited. Innovation, thus, must take place within these constraints.
Data transport, though not expensive, is also not free. Thus, there are cost constraints to moving data. Carriers should offer reasonable terms, and innovators should invent within the realm of economically viable use of limited resources.
The 600 lb gorilla in the room: This issue has nothing to do with efficiencies, innovation or any other “human-based” value. This issue is all about control. And control is all about money. More control, more money.
Give ANY industry the opportunity and they’ll charge for it; either directly or indirectly. They’ll find a way to charge for it. Cable and satellite and their ilk are not innovators; they’re leeches.
My suggestion: Open it up, full throttle; wide open.
Oh yeah, one other thing: every municipality supported by a tax base should provide free, high-speed, “unfettered” wireless access to EVERYONE. Comcast, Mediacom, etc. can find some other way to make a buck other than sucking the users dry.
It's Part Of The Ploy To Rob From You
The caps are there to get people used to and excepting of them. Then they will be lowered to hamper IPTV and other media that interferes with the cable company selling you TV.
Unless you buy your approved multimedia from the cable company you will be paying more for your internet connection.
Now is the time to break with the telcos and cable companies (if you can) and get your internet connection from someplace else.
I hope wimax hits it big. But not in the hands of the cable or telco companies.
Re: It's Part Of The Ploy To Rob From You
As I said before, Comcast already has its dirty little hands in a large WiMax venture (Clearwire). I wish we could all have T1 or OC3 lines into our homes, but it’s cost prohibitive. The cable and telcos bribe municipalities to allow them to operate, then slowly choke their customer base. They start out with unlimited, then squeeze!
Unfettered Broadband
Any Cap will hurt future software development. I have been teaching my self how to write code and build apps over the last 3 years. Without unlimited access I couldn’t. Has anybody looked at how big SDK’s are? Or how big a Linux Distro is? I use Windows and Linux both. Should say multiple versions of Linux. If I had a Cap that would go out the window. Much less what it takes to get a Clean Windows install current. There is no way out of millions of just Americans I am the only one doing this. If there needs to be a limit put it on how much these media sites can send not on how much the end user can download. There is still a thing called a TV and a Stereo for media. You can’t develop software on either.
5GB cap in 2009
That is what Frontier is planning to do according to press reports. I work from home, jacking into my companies internal global network via an ADSL VPN connection. Last week on one single day my 1.5mb “broadband” connection was real busy when work auto refreshed a database on my hard drive with qty 12,386 files update. Only took 6 hours to transfer. Later that evening I was working on a movie clip I made for a local animal rescue group. I needed to upgrade Pinnacle Studio. That was about a 3GB download, then another 200mb update patch, then upload the flick to the rescue group web site. In one day I easily exceeded that 5GB monthly cap in one day.
Caps are all about money, greed, and making seniors execs at ISP’s wealthy.
Bandwidth is a measure of bits per second. Caps are are measurement of total transfer amount. Which are you talking about Mike? Tim is talking about bandwidth, not caps.
Caps are not a problem
Nor do they limit innovation.
It simply the pricing structure.
If broadband is $10 plus $.10/GB, 250GB comes to $35/month.
Cable is looking at pricing bandwidth like cellphone plans, with minimums, buckets, floors, overages, and “nights and weekends”.
If, instead, it was truly metered, the usage rates were not exorbitant, and we could actually monitor our usage/bill, I don’t see why it would be a problem for the consumer. I have a measured phone bill, a measured cellphone, a measured gas, electric, water and sewer bill, and I’d be happier if my trash was by-the-pound.
Why not the same for broadband, and while you’re at it, let me buy only the channels I want on the TV?
Time to chime in here.
I believe there’s a confusion between business broadband with consumer broadband. These are NOT the same, as I’ve yet to see any caps placed on business broadband.
The average consumer does not need 250G/mo of broadband at this time. Even if someone where to download 8 HD movies a day, they’d not touch the cap.
Consumers who utilize more than this per month are probably being assumed to be a business, which has a much different pricing structure than a consumer, often with larger benefits of utilizing the business account. I’ve yet to see anyone challenge this cap when they were affected to a degree in which the ISP didn’t work with them to remove the cap.
So far, these blogs have been speculation on what could be but doesn’t touch what is actual.
I do agree that as the web continues to grow, the need for more broadband will be apparent, at which time most ISPs would then readjust their cap. But 250G/mo now seems pretty damn reasonable for the average consumer.
Now I want to focus something said in the initial blog:
Today, computers are fast enough that developers can use high-level frameworks that are much more powerful but consume a lot more resources. Developers spend more time adding new features and less time trying to squeeze better performance out of the features they already have. Which means users get more and better applications.
This is not entirely accurate. In fact, over the past 20 years, consumers have been arguing more is not often better as programs get more difficult to use, rather than easier.
A great example of this is Microsoft products. I remember a time in which these were absolutely easy to use and did quite well for what they were intended. In looking at products today from Microsoft, not a single one is easy to use and trying to use them are challenging for many users, especially those who’ve been using the product for quite some time.
I remember when the web’s largest complaint was page load times. Website’s with large images and complicated database transactions over a limited (truly capped by the FCC here, folks) phone line weren’t tolerated much. This wasn’t too long ago.
Now, more and more “applications” are getting so big, websites place in “loading” screens, forcing visitors to wait. For what? A dancing fish singing the blues?
I’ve been teaching people for years the KISS* method is the ONLY rule to define how the application should be developed. To think each line of code costs you $1 to write, so save money when developing.
This has lead some of my students to rethink the way to actually develop code and to utilize the full potential of available resources rather than do what most do today: Bundle it all up.
I can only say if that the argument for non-capped accounts is such that we have to endure longer loading screens just to see dancing fish singing the blues, I say keep the caps.
Maybe this will force developers to rethink their design and retain the KISS method, rather than overloading their products with features that less than 5% of users take advantage of.
Just because something can be done does not necessarily mean it should be done.
*KISS = Keep It Simple, Stupid
The big problem with the whole idea that people are using too much bandwidth ignores one small fact;
Every broadband account is already capped by the fact that each user’s account is speed capped at whatever service tier they pay for. If I have a 6Mb account, I can’t magically download data at 20Mb. This automatically limits how much data I can transfer.
So what the ISPs are saying is that they’ve sold everyone accounts with caps that they can’t support. Of course they knew right from the start that they couldn’t support the speeds that they were selling. So now, rather than upgrade to support the service tiers that they have, they’re imposing usage caps so that they can use any upgrades to offer even higher tiers of service that they can’t support.
What they should do is figure out how much they’d need to charge to actually support the service tiers they’re offering and charge people that with no usage cap. It might be higher, but at least people wouldn’t have to worry about getting nickel & dimed to death in overage charges.