Why is this page text-only?
 

« All Un/Underserved Communities Should Be Aggregating Demand | Main | When Is A Community Fully Served By Broadband? »

August 5, 2009 10:07 AM

Death To T1s (And Maybe DSL, Too)

I've been meaning to say this for a while, so here it goes: Death to T1s!

Before going any further let me pause to give them some credit. They've been a great way to get reliable access to the Internet for years.

But their time must come to an end ASAP. Quite simply: we have too many people paying too much for T1s that deliver too little bandwidth all across this country.

I cannot tell you how many times I've come to find entire hotels running off of a single T1, schools getting by with only a couple of T1s for hundreds of students, government buildings that are lucky to even have T1s.

I have a hard time fathoming how anyone can do anything with this little bandwidth. A single T1 delivers only 1.5Mbps of bandwidth. While this is guaranteed bandwidth, unlike cable or wireless networks, it simply isn't enough capacity to support the dozens of people in a typical school that may be trying to access the Internet at the same time. And I can tell you from experience, that I'm not sure any number of T1s are sufficient to handle a hotel's traffic once everyone's back in their rooms at night, trying to check their email before going to bed.

But what's even more flabbergasting about the prevalence of T1s are how expensive they are. Businesses and hospitals are paying hundreds of dollars a month for a single T1 line. A big part of the reason why they don't have more bandwidth is they can't afford to pay even more money just to add bandwidth 1.5Mbps at a time.

T1s were great a few years ago, but the fact that they're the only business-class service available to many buildings is a clear sign of the inadequacy of America's broadband infrastructure.

Every building that houses any number of people needs to have access to a fiber line, which can not only deliver exponentially more capacity but at a price that's often roughly the same as they pay now for T1s. I've heard some of my fiber providing friends offering 100Mbps service at similar prices to what incumbent providers charge today for a few T1s that offer less than 10Mbps.

It's time to take our heads out of the sand and recognize that when it comes to supporting the bandwidth needs of large buildings, we need fiber to be ubiquitously available. T1s just aren't up to the task of supporting the demands of the 21st century digital economy.

Which brings me to DSL. I'm a little more cautious about calling for the death of DSL as I fully acknowledge that at least when it comes to upgraded DSL networks, today they are delivering the kind of bandwidth most users need. And there is the potential for them to evolve further to deliver speeds of 100Mbps, though whether or not they'll be able to support that much throughput to all users or just those located geographically close to the central office is another matter entirely.

But at the same time, I can't help but be frustrated by DSL's impact on discussions regarding our national broadband policy, namely how many people want to limit the bandwidth goals we set to numbers that DSL can reasonably manage.

If you have a thorough conversation about how much bandwidth a typical house of the near future will need, the numbers quickly add up to at least 100Mbps.

But when DSL defenders talk about how much bandwidth we can reasonably get to most people, they don't think universal 100Mbps is realistic.

Don't get me wrong, I don't blame them for this. How can they be asked to support something that makes their technology irrelevant?

Yet at the same time, I don't see how we're allowing ourselves to limit the goals we set that are supposed to be in the public's best interest just to protect a technology that most everyone I know, even some DSL providers, acknowledges will be inadequate in the not too distant future. That doesn't make any sense to me.

That said, I wouldn't go so far as to call for the death of DSL as it does have a lot of utility to provide for the foreseeable future. Instead, what I'd ask is that if policymakers want to remain technology neutral, then let's set goals that are truly technology neutral, let's set benchmarks based on the public's interest and not on protecting DSL's relevancy.

Then by doing so, perhaps we can spur further evolution of DSL technology to be able to support universal symmetrical 100Mbps across its entire footprint. Just because we set a more aspirational goal doesn't mean we're ruling out technologies that are currently inadequate. We're just clearly stating what level the public needs and that these technologies should be striving to achieve.

So let us all say together, "Death to T1s! And death to DSL's current technological limitations as the benchmark for our national broadband goals!"

Del.icio.us Digg Yahoo! My Web Seed Newsvine reddit Technorati

TrackBack

TrackBack URL for this entry:
http://www.app-rising.com/cgi-bin/mt/mt-tb.cgi/1439

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)