Super Bowl, Awful Cable – and a reason LTE will be about data not voice

Posted by Mike Roberts Monday, February 2nd, 2009

Add a comment

I’m watching the fourth quarter of Super Bowl 43 and the excitement is building, with the Arizona Cardinals coming back through a great drive and touchdown – and then my cable box crashes. Screams, disbelief, anger, incomprehension – how could such a mature service crash at such a critical time?

Well the box came back on and went through its lengthy reboot, with hardware checks, software applications loading, etc etc. I grabbed the phone—luckily it was a landline so didn’t rely on the cable line to work—and called the cable company, and got through to someone fairly quickly. They were efficient and professional and arranged to have a technician come around the next day at 8am. So not a bad fault handling service.

And by the time I was off the phone the TV was working again – so I only missed about two minutes by the game clock, which made me thankful for once for all the commercials. So maybe my title above is a bit harsh, given that the service recovered quickly and I saw the vast majority of the game without a problem.

But to paraphrase the game commentator John Madden, you can talk all you want to about statistics, but what it comes down to is do they do what they have to when it really counts? And that applies to telecoms operators and services as much as it does to football teams and players.

In the case of the Pittsburgh Steelers, the answer is definitively yes. They led most of the game, and then lost the lead in the fourth quarter, and came back in a dramatic drive culminating in a picture-perfect touchdown pass in the last seconds of the game. The stuff of schoolboy dreams making for the best Super Bowl I’ve seen.

In the case of my cable provider, the answer is no. Yes their customer service rep was responsive and addressed the problem as best she could, and yes the service came back online after five to ten minutes. But the fact that it crashed at all during the biggest US sporting event of the year was almost too much to bear. My confidence in the service took a serious hit, and I’m not sure it will recover. At the very least I’ll be doing some serious shopping around to see what the competition has to offer.

It doesn’t help that the technician came the next day and fixed the TV but broke broadband, at least for a few hours until he could come back a second time to get it working again. But to be fair, the technician was very knowledgeable and helpful, and to some extent was unlucky that a previous technician (this wasn’t the first time I’ve had problems) had left things in a muddle.

That’s all very well and good, you may say, but what does it have to do with LTE? Well my cable box is a computer, and computers crash, and maybe I’ll have to get used to it. Or will I? Will the next-generation services really get to the point where they have the five-nines reliability that we’ve learned to expect from analogue TV and phone services? Or will we simply learn to take a few steps back in quality in exchange for a wave of new features and services?

I’ve raised a similar point with telecoms experts at many industry conferences, and the answer is always that the new services will have similar or even better quality than the old analogue services. The only problem is that has not been my experience so far—particularly when it really counted. Like my cable box crashing during the Super Bowl, or my old fixed-line VoIP service flaking out during more than one key call.

So to come around more directly to the topic I’m working on at the moment, LTE. Our mobile phones are already computers, and will be even moreso with the move to LTE. So are we all ready to take a chance with our mobile services, to hope they’ll work but not to the point we can really count on them?

Not really, the industry has decided - at least not for voice services. That’s one of the reasons VoIP did not make it into LTE as standardized in 3GPP Release 8. Of course it’s not really necessary for LTE to include VoIP given how mature and cost-effective traditional circuit-switched mobile voice services are today, and given that LTE will be backward-compatible with WCDMA/HSPA and CDMA2000. But the fact that it’s not even in the standard says a lot.

First and most obviously, it says LTE VoIP is not ready. Second, it says that mobile broadband may be booming, but mobile voice services still pay the bills, and the industry is understandably very cautious about messing with the mature and battle-hardened technologies underlying its main breadwinner. Even if that means delaying the transition to all-IP networks and services.

It also says to me that most of us won’t accept a loss of quality in exchange for new features or services. In a way that’s a truism in our industry, but I’m always struck by how that doesn’t fit with my own experience. Maybe that’s just me.

Or maybe it’s more representative than we’d like to admit, and helps to explain why LTE will initially be data-only, with voice services coming along much later. Maybe robust VoIP will become part of LTE in 3GPP Release 9, as currently planned, and so will be completed along with the standard by December 2009. Or maybe it won’t.

In any case, when LTE VoIP services are eventually standardized and come to market, they better work when it counts or we’ll all be shopping for alternatives.