How Do You Solve A Problem Like Amtrak’s WiFi? Ctd

Andrew Sullivan —  Jan 29 2013 @ 9:26am

A reader warns:

Your reader probably shouldn’t “just jailbreak my iPhone and use it as a hotspot.”  As of this past Saturday, such activities are now illegal.

Update from a reader below correcting that claim. Another reader:

If they can do it on high speed rural trains in France and mountainous and tunneled areas of Germany, surely Amtrak has no excuse. Wifi router on board connected to a satellite system and a contract with a wireless provider for the few tunnels they have and – boom – it’s done. Ok, maybe they need to start with a customer base, but that may be a chicken/egg argument.

Another goes into detail:

One of your readers wrote: “If it’s really so hard to provide good wifi, how is it that airplanes can do it? Not too many cell towers at 30,000 feet and 500mph, last I could see.” That’s obviously written by someone who has no idea how wireless signals work.

If you were to turn on your cell phone during a flight you’d most likely have no trouble getting a signal. Wireless signals will go a long way unless something blocks them. The reason you need so many towers at ground level is that there are things like buildings and hills getting in the way of the signal. When you’re in an airplane, you have direct line of site to many more cell towers than you ever would on the ground.

This is actually what the problem is. The cellular network is designed to use the power of the signal received to figure out which tower has the best signal to you and prepare for hand offs as you move from one tower to the next. When you have direct line of site to multiple towers and you’re distance between them is fairly uniform, the network has a hard time figuring out which tower to use and how to pass you from one to the other. The is compounded by the speed at which you are travelling and need to be passed. What actually happens is that the network is reserving extra space for you on multiple towers, eating up spectrum that would be better used by people on the ground.

Because of that, the trick of wifi or cellular calling on the plane is to make sure that your device connects to the receiver on the plane and not one on the ground. Once you connect to that receiver, you receive your data from the plane’s network. The plane manages the entire connection to the rest of the world. And again this is made easier due to the fact that you are in the air. You have direct line of site to locations on the ground for hundreds of miles in each direction and easy line of site to satellites. Compare this to a train. It’s going up and over hills, into cities with tall buildings, through tunnels, and dealing with all the problems of wireless communication on the ground. It’s not an impossible problem to solve, but it’s a harder one. And since most us already have personal internet devices that work just fine on the ground and aren’t banned (like they are in the air), the expense to make a ground based network isn’t worth it.

Another goes similarly in depth:

Gogo uses a series of cellular towers that point into the sky and then transmit data to the aircraft from the ground.  This solution would not work for Amtrak because trains are on the ground and it loses line of site to the towers – the problem they have right now with Verizon and AT&T.  Gogo, on the other hand, is able to cover the entire US with just over 100 towers.

So Gogo is a no go. The only solution will be to get Verizon and AT&T to build more towers that provide coverage to the rail lines.  Perhaps charging for the service on the trains could provide revenue but this has a chicken and egg problem.  Until you have consistent reliable service you can’t get people to pay for it.  You can’t build out the reliable service until you have the revenue to cover the costs.

The problem though is even getting beyond that, the revenue from the wifi service likely can’t cover the costs involved.  Consider Gogo’s business model where they have ~100 towers, are charging $14/day and presumably making a profit. In 2011, there were 720 million airline passengers and only 30 million Amtrak passengers.  Many fewer passengers means many fewer wifi users and a lot less revenue.

If you assume that the same proportion of passengers use wifi in both scenarios, that means those 30 million passengers would have to pay $336/day to bring in the same amount of revenue as what Gogo gets.  Furthermore, I’m assuming here that you could cover Amtrak’s dead spots with a mere 100 towers which seems unlikely.  Sure there’s other factors to consider here like the fact that Amtrak could operate the service for no profit, and that an Amtrak passenger may spend multiple days on a train, unlike an airline passenger.  But that’s a pretty big gap to make up for.

Seems to me that, given the numbers, the free model is the only one that really makes sense.  You provide a service that’s inconsistent but largely works.  When it doesn’t work people can’t really complain because they didn’t pay for it.  You don’t have to worry about billing or technical support.  But at the end of the day you can advertise that your trains have wifi and mostly be correct.

Another looks ahead:

If you haven’t seen, JetBlue is launching near-streaming quality Internet access on flights in coming months. They’ll start with 30 planes and then expand to the entire fleet. The key to the approach is a fundamentally different type of technology. How do you make Internet connection faster when you’re using dial-up modems? Well, you need to stop using dial-up. Today’s ground-based services like the ones on Amtrak are about to be made obsolete through Ka-band satellite technology.

Most interesting, rather than jacking up prices like Gogo is, they’re using the crack dealer approach: they’re so confident people will come back for it, they’re making the first taste of the service free (at least for a period of time).

That update from a reader:

A reader wrote in that jailbreaking an iPhone was now illegal, and provided a link to a Daily Beast article (sourced from the NYT) which stated as such. This is incorrect.

Jailbreaking and unlocking are actually two different things, and one has been made illegal, but not the other. What’s now illegal is “unlocking” – this is when you alter the software on your phone to make it usable on other carrier’s networks, so your AT&T phone starts working on, say, T-mobile’s network. This is now against the DMCA as of a few days ago.

Jailbreaking – removing the software controls that restrict access to the apps on the phone, so you can install software that the manufacturer doesn’t support – is still protected, and should last through at least 2015 according to the EFF. Doing this is likely to break your phone’s warranty or violate your carrier’s terms of service, and thus might result in your carrier blocking your service, but it is not against the law.

The specific issue in question – jailbreaking the phone to install software that will use it as a wifi hotspot – is still protected under the law.