A few years ago, gigabit broadband service was still considered science-fiction: only a few places had it, like Japan, Korea or Sweden, and to most of the world it seemed so insanely better than whatever broadband speeds they had available that no one dared to dream that it would become mainstream one day.

And yet, in just a few years, gigabit has become a common offering. It’s far from universally available yet, but there are many more markets where one can subscribe to a gigabit of broadband at affordable prices: France, Turkey, Portugal, and many US cities.

This week, Japanese ISP Nuro announced the launch of their 10Gbit/s residential broadband offer. The service is going to be 10/2.5, based on NG-PON1, as far as I can tell, and will cost 6480 yen (about $64 per month) for a two-year contract for broadband only. Does that mean we're entering the 10 gig era? Will this make gigabit offers obsolete in just a few years?

Of course, you can bet that as soon as a newspaper in Europe or the US writes about this, incumbent operators and their frontmen will come up with their favorite line ever (come on, you know you can finish this one with me): "Nobody needs 10 gig!" In fact I'm willing to bet they will go one step further and say: "Nobody will EVER need 10 gig!" It's easy to understand why they will say that; they're trying to convince most of their existing customer base that 50Mbit/s is more than enough.

And as usual, when they're not being disingenuous, they'll be missing the point. Big time.

Japanese operators first deployed fiber over a decade ago. That infrastructure is deployed, it's in the ground, and for the most part it's been paid for. Furthermore, the first wave of PON equipment that was installed to light that fiber and deliver gigabit service is close to obsolete now (these active components generally last 5-7 years on average.) Which means that the ISPs who have deployed this fiber 10 years ago are in the process of renewing their active equipment. 

If you're going to do that, you might as well go for the latest (or at least one of the latest) technologies and deliver more and better service to your customers. The marginal cost of doing that is minimal: NG-PON1 may cost a fraction more per subscriber than GPON, but the difference isn't massive.

So the real question is: If you have fiber deployed already, and if you need to reinvest in active equipment anyway, why wouldn't you offer 10 gig to residential customers if you can?

And the only reasonable answer to that is: Because customers won't be willing to pay a little extra to get it. Except gigabit has shown us that there is a sizeable proportion of customers willing to pay a little extra to get a better throughput. Will this be true of 10 gig service?

I can't say. But I know that a whole lot of people are willing to pay a whole lot more money for 24MP cameras (not to mention 50MP) even though the likelihood of them ever printing a billboard size photo (which is what so many pixels allow you to do) is close to zero. Customers are willing to pay a lot more money for cars that will peak at speeds of 300kph or more despite never ever having the opportunity to even get close to 200 (plus it's illegal). 

Marketing is not an exact science, but neither is it only about need. In fact, it's about want a lot more than it is about need. And it's about differentiation. And once all your competitors have the same technology you have, one of the ways you can differentiate is to deliver service over a better technology. 

So I for one will be following Nuro's first ever affordable worldwide gigabit offering (PCCW has an offer available in Hong Kong, but affordable it ain't) with great interest. Who knows, it could be that Japanese customers actually like being at the cutting edge.