Earlier this week I came across this interesting study by BroadbandChoices that suggested consumers care more about broadband reliability than speed.
While our national broadband plan does touch on the issue of broadband reliability, suggesting it's a topic worth researching further and one that's essential for applications like public safety and smart grid, it's pretty much silent on the issue of whether or not America's broadband infrastructure is currently reliable enough.
And yet, here are consumers saying that reliability is more important than capacity. So let's consider this issue a little further.
First, a fundamental truth of broadband reliability is that shared networks are less reliable than dedicated, and lower capacity networks are less reliable than higher capacity ones.
Put into practical terms, that means that DSL networks tend to be more reliable than cable or wireless networks as each customer essentially has their own pipe vs. sharing with all their neighbors. That also means that fiber's more reliable than any of these other networks as it has by far the most capacity.
Second, broadband reliability also includes issues related to the physical characteristics of networks. For example, DSL reliability is limited by distance, so if you're too far from the central office your connection won't work as well as if you're closer. Also, networks with active components in the field are less reliable than those with passive. Since copper networks require electronics throughout they're less reliable than passive fiber networks.
Third, broadband reliability encompasses the nexus of technical and business decisions made in connecting your local network out onto the Internet at large. So if your community doesn't have a big backhaul pipe to the Internet, or your provider decides to overprovision that backhaul connection to the extreme to try and drive greater profits your network will be less reliable.
What's so frustrating is that while our national broadband plan isn't completely silent about these issues, it also doesn't really address them head on within the context of insuring Americans have access to reliable broadband networks. Instead it almost seems to assume that broadband networks are and will continue to be reliable.
And yet the exact opposite is the case anecdotally all across America today. I can think of any time I've been at an outdoor event with tens of thousands of people and how my iPhone stops working. Or when I was visiting relatives in Cape Cod and learned that when schools let out in the afternoon Internet surfing slows to a crawl. Or when I try to use videoconferencing software at home and the call is dropped or the video quality deteriorates. Or any time there's a major national event and the Internet as a whole nearly shuts down, unable to handle the massive influx of traffic.
While I know this last point often has more to do with the reliability of websites and servers rather than broadband networks, it's another example of the reliability issue that we need to be finding solutions to if we intend to be able to rely on all the things broadband makes possible to improve our day-to-day lives.
It was bad enough to know that the FCC largely ignored issues of reliability when it was from the perspective of being a more academic exercise. But now that we're hearing that customers themselves demand networks that put reliability over capacity, it's even more galling that there's been little to no serious debate in DC around how we can improve the reliability of our country's broadband infrastructure.
So as we move forward, let's not forget that capacity alone is not the only issue that matters. That when push comes to shove, it's reliability that matters most as without it how can we rely on the Internet?
BroadbandChoices is a UK company. They describe their survey here:
http://www.broadbandchoices.co.uk/broadband-speed-tops-list-of-complaints-2906102.html .
Although reliability was "top priority" and the "most important factor when choosing a new package," "[a] massive 77% of people cited broadband speeds as their biggest issue." I find this description confusing.
I suppose the bottom line is that both are important.
I don't agree that it's a "fundamental truth" that "lower capacity networks are less reliable than higher capacity ones." But I do agree that fiber networks are typically more reliable than copper networks. Also, fiber networks are typically faster than copper networks.
You also say it's a "fundamental truth" that "shared networks are less reliable than dedicated" networks. Does this mean you think point-to-point active Ethernet fiber networks are more reliable than PON networks?