Why I Use T-Mobile

Lots of people have piped up about abstract issues in the AT&T acquisition of T-Mobile. I think it's a horrible idea, and my reasons are pretty concrete.

Several years ago, I became a Cingular (now AT&T Wireless) for 10 days. I wasn't happy with their service, and I went back to T-Mobile. I cancelled my $20/month unlimited data plan within their 14-day cancellation period, and the monthly data fees should have been pro-rated. Instead, I got billed for over $3,700 of off-contract data usage. I had done a handset exchange on day two, and the salesperson didn't update their database properly to show that the new handset was now on the contract.

I spoke to a sequence of people in Cingular customer service, and every single one agreed that the charge was incorrect. In spite of this it took nine months, a long-standing ding on my credit report, repeated contact with their executive complaint group, and ultimately the threat of a lawsuit before they got the matter fully resolved, even though it was never in question that the error was theirs.

I'll leave aside whether their off-contract usage was sensibly priced, and whether such a pricing structure wasn't inherently usurious at a time when smart handsets were new and often didn't offer any way to fully disable their data usage if the customer did not need that. AT&T, Inc.: We Scare Because We Care.

Today I'm once again a very reluctant AT&T customer for my iPad, but all of our handsets remain on T-Mobile. Setting aside my minor billing mishap, the reasons are pretty simple:

• T-Mobile customer support answers the phone at hours when I can actually call them. AT&T doesn't.
• AT&T's customer-service has improved since my last run-in with them.
• In my experience as a user, AT&T doesn't reliably honor their roaming agreements. I've only found this to be true where T-Mobile service was dicey and AT&T offered four bars. Thankfully, it's only a problem at Disney World and my parents' house, and only for the last decade or so.
• T-Mobile offers handset and device options that AT&T is slow to implement. No current-generation Android tablets seem to be contemplated over at AT&T. It reflects a different attitude about their respective customers.
• T-Mobile's international roaming in Europe is priced very reasonably. That's useful when I travel.

The last point also explains why I'm not a Verizon customer - their phones don't work in Europe at all.

As a reasonably happy T-Mobile customer, my concern with the proposed merger is that as they consolidate operations between the two organizations, AT&T will retain the wrong customer service group, the wrong management team, and the wrong product management team. Other than those small inconveniences, the merger sounds like a fine idea.

Where's Lilly Tomlin when we need her? One Ringy-Dingy...

Tethering and the Cost of Wireless

I'm a developer, which means that I look at a lot of options that aren't necessarily appropriate for consumers. I'm also an entrepreneur, which means that I care about the cost of essential services, including data.

In the last few weeks (March 2011), I found myself looking at the current and upcoming Android tablet devices. Android tablets aren't quite ready for prime time unless you are willing to install a few add-ons and put up with some teething pains. I'll take that up in another blog entry. The surprise is that certain features of Android Gingerbread (phone) and Honeycomb (tablet) are game changing in ways that haven't been examined in mainstream reviews.

Most users who buy a current-generation tablet device already own a cell phone. I currently own an iPad (Apple/AT&T), a Xoom tablet (Android/Verizon) and a Samsung Galaxy-S phone (Samsung/T-Mobile, the original Vibrant handset). The iPad is an AT&T device, and I purchased it in spite of bitter negative experiences with AT&T. The Vibrant is with T-Mobile, my long-term cellular carrier, because I travel internationally. The Xoom is with Verizon. In my opinion, David Pogue (NYT) and Walt Mossberg (WSJ) overlooked significant issues in their comparisons of the iPad vs. the Android Tablets.

If you compare the one-year cost (including data plans) of these devices, you will find that the costs are identical. The Android tablets are unquestionably rougher, and they need a few, cheap, applications installed to achieve parity, but once you do this a direct comparison comes out closer than the Mossberg/Pogue reviews suggest. If you need a device that works right now, get the iPad. Otherwise, remember that the reviews compare a pre-release Xoom to a year-old iPad. But the Android devices offer WiFi tethering, and that changes the game. Depending on how immediate your needs may be, you may also want to consider that Android handset numbers overtook iPhone handsets in less than a year.

From a cost perspective, the important question is: how much data do you need to pay for over the first year of operation? Also: which device should you use as your source of connectivity? The Gingerbread (Android 2.2) and Honeycomb (Android 3.0, tablet) devices all have the ability to serve as access points for other devices.

I am a heavy data user. I have measured my usage since the day the iPad released on April 3, 2010 (today is March 22, 2011). Since the day I purchased the device, I have used a grand total of 5GB of download capacity. Some caveats:

• I don't download movies when I am on the road. Movies eat capacity. I transfer movies at home, in advance of my trips.
• My iPad is first-generation, which means that it isn't capable of video calls. Video calls require less capacity than movies, and I don't (yet) have data on the impact of this.

Subject to these two caveats, my total usage over nearly a year is 5GB. On average, that is much less than 1GB per month. WiFi at home covers the rest. Your usage may be different, but data usage varies greatly from one month to the next. It is probably worth your effort to look.

In my case, the data usage numbers mean that if I pay for 1GB (or more) on each of my devices, I'm paying for redundant service. If one device could serve the others, I'ld be much better off paying for cellular access on just one. On the current generation Android devices this capability is standard.

Today, I am paying for three data services:

• The AT&T "unlimited" plan at $35/month. This is no longer available.
• The T-Mobile "unlimited" data plan at ~$20/month. On my phone this delivers 7Mbits. On current phone it delvers 21 MBits.
• The Verizon 3G plan at $35 per month for 5Gbytes/month. Delivered speed not calibrated by me.

I have tested both the iPad and the Xoom using the T-Mobile phone as an access point, with two results:

• There is no noticeable loss of speed when these devices operate through my T-Mobile phone.
• There is no noticeable increase in latency when these devices operate through my T-Mobile phone. In English: video calls work fine.

Given this, there are only a few reasons to maintain multiple cellular data plans:

• You own an Apple or a Microsoft device, which do not provide tethering functionality.
• You want the ability (as I do) to test responsiveness on different networks.
• You are insensitive to usage-driven cost metrics.
• You find the complexity cost of enabling the access point functionality higher than the cost of paying for multiple plans. For many users this is a legitemate concern!

There is also a more subtle consideration. Tethering burns power, which raises the question: Where can you recharge? Current tablet device chargers require a 120V outlet. This means that they cannot be recharged in your car. Most cell phones today can be recharged from USB power. There are a wide range of car adapters for cell phones. Unless your car provides 120V A/C outlets, you can't recharge a tablet.

So if you are interested in consolidating your data plans, you need to consider the problem of power. When using your tablet device in your car, it's relatively easy to plug your cell phone in to a USB adaptor and use it as an access point. In a hotel, or at home, you can reasonably expect to have direct access to both power and WiFi. But the unlimited data plan on your cell phone is something that you probably pay for already, while the limited datra plan on your tablet device is - at least by comparison - expensive. The currently advatageous pricing and charging point is to tether your tablet through your cell phone.

It seems evident that we are looking at a pricing regime that cannot last. Third parties will make on-demand activation of access points possible, and competitive forces will make it difficult for cellular carriers to intercede. Either the Google branding effort (and consequently tethering) will fail, or the cost of service for multiple devices will drop. Which one seems up for grabs.

What does seem clear is that an investment in tethering as a baseline capability - and consequently an investment in Android handsets and tablets - is likely to reduce the de practico cost of data services. The more people who buy Android devices (of any flavor) today, the more downward pressure will exist on data bandwidth pricing.

Speaking for myself, I find that the battery capacity of my Vibrant, coupled with a USB adapter in my car, is more than adequate to serve my mobile data demands. Your mileage may vary, but if nothing else, try the experiment.

Edit (3/22): Peter Sahlstrom points out that tethering with a USB cable has been possible for some time. He is right, of course. It has also been available over bluetooth, though at fairly slow speeds. Both are fairly awkward solutions, and in most cases they violate your wireless agreement. The difference here is that the mechanism is both "official" and convenient. I can turn on WiFi tethering, stuff the phone back on my belt, and then use the other devices conveniently and unobtrusively, and without any cables getting in the way.