What will T-Mobile Do?

So AT&T is going to give up on T-Mobile.  Most commentators seem to describe this as a bad outcome for T-Mobile.  What does the poor 4th place wireless vendor do?

The most important thing is to realize that T-Mobile is not going to grow its business into being the #2 or #1 network.  And if T-Mobile competes head on, it's going to fail: it can't offer the same products and services just with a worse network.

Instead, T-Mobile needs to skate to where the puck will be.  And where the puck will be is data.  All data, all the time.  GSM voice is a 90s technology.  People like Republic Wireless are leading the move away from segregated voice vs. data.  But Republic Wireless' weakness is that it doesn't own a 3G or 4G network or have WiFi hotspots.

T-Mobile, you're already big in the wireless hotspot game.  Keep it up.  Get in as many locations as possible.  Buy Republic Wireless with that $3 billion in go-away money and make their devices work seamlessly at your hotspots.  Figure out where a core clientele spends its time, provide WiFi there, bridge the gaps with 3G, and you have a truly unique offering.  People like unique.  People like avant-garde.

Pick up Twilio and a few other internet telephony start ups too -- and focus on making T-Mobile the platform of choice for inventors and entrepreneurs.  Soon, all sorts of cool apps will land on T-Mobile, and people will want to use your service for all the things it does.

So much better than being the #4 carrier losing at playing catch up.

Why virtualization will take over the desktop

One of the biggest problems that makers of desktop operating systems face is backwards compatibility.  You can either be a slave to it (Microsoft) or strategically break it (Apple), but either way a major new release of an OS is going to have to deal with it.

And it's not hard to maintain compatibility back one or two releases. But if you had some really beloved Windows 3.1 application, the chances of it working now are pretty slim.

The answer is virtualization.  What if, instead of having to maintain extensive backwards compatibility, you architected your system so that you had a low level OS, a VM on top of it (like so many existing VM stacks do) and then ran your OS in a VM?

When it came time to upgrade, the user could load a new OS and keep the old OS around for as long as needed.  Decades, maybe.  If you're Apple, and it's time for your 10 year shift in hardware, the VM could emulate the old hardware indefinitely, saving you from having to build something hacky.

In this world, the VMs don't have to provide complete isolation of the OSes.  The VM could, for example, support the underlying file system (since even modern OSes still support MS/DOS's FAT) and even potentially handle some windowing issues (like VMWare Fusion does on the Mac).

For those of us on a Mac who need Windows applications, we're already getting there. I have a virtualized Windows XP that I can run forever without worrying about XP specific programs breaking under new versions of Windows.  But for Windows users, this is still a new concept, but one that is coming, I'm sure.

Microsoft: Bundle your VM into Windows.

Apple: Buy Parallels and make it a core component of your next OS.

Linux: Start putting VM support into base distros.

You all can thank me later.

Using your phone as a boarding pass -- bad ergonomics

I've been meaning to write something about this for a while, but since yesterday I used my phone for a boarding pass again and ...

I'm not really sure it's a great idea.  Mostly because, well, when I'm traveling I'm clumsy.  I have a suit case, I have a brief case, and when I'm going through security I have a driver's license out and, either a boarding pass or my iPhone.  That's a lot of moving pieces.

The iPhone is getting moved around and waved around, and that makes me nervous.  It's an expensive piece of hardware to have in motion when I'm juggling other things.

A piece of paper and a driver's license are easy to handle in one hand, and they go to the same place, the TSA agent.  The iPhone and the license are going two different places.  That's just bad design.

As a side note, since they don't appear to look at the phone, and the scanner device doesn't appear to show my name (did I miss it), what's the point of showing the license? (Not that there ever really was a point).

The iPhone 4S -- why it's important

With Steve's passing, it's almost hard to remember that there was another round of Apple news this week, the introduction of the iPhone 4S.

Reactions to it seem to be split between pure disappointment and a nuanced assessment of the iPhone upgrade rhythm (major, minor, major, minor, ...).

But the real news is the upgrade of the processor.  We've seen that there's a been a very gradual and almost grudging introduction of multiprocessing in iOS devices, starting with none in iOS 3 (beyond a phone and iPod apps) and moving to a limited support in iOS 4 with some background processing and push notifications.

But multiprocessing is the key to the next generation of applications.  When you have always connected devices, you are going to want them to always be doing things.  Push notifications are find for infrequent events, but true background processing is what is needed for continual monitoring and processing.

You can say that you can jailbreak the device and get true multiprocessing, but people generally have bad performance experiences -- the hardware hasn't been powerful enough.

It's not clear if the A5 is going to get Apple all the way there, but it is certainly a big step in the process.

If I were an app developer, I would be thinking about what a program that runs persistently in the background could do that is new and compelling.  Then, when iOS 6 is released, I'd be ready to take the market by storm.

Ten Reasons Why the Kindle Fire is good for the iPad

If you follow the press after the Amazon Kindle Fire announcement, you might think that the only question is how bad it will be for the iPad.  Most seem to have settled on "it will take market share away, but it's not a killer."

I, on the other hand, think the Kindle Fire will be good for the iPad.  Why so?

1. It moves us from an iPad market to a Tablet market.  That might seem bad, but if the Fire sells well people will realize that it's not just a bunch of people mesmerized by Steve Jobs that are buying these things.  And so the conversation will move to what is the best in tablets, which is going to be the iPad.
2. It gives Apple something to compete against.  Not that Apple would rest on their laurels, but competition is always good for consumers.  The iPad will be better because of this.
3. It lets people put-off by the iPad (and its price) get into the market and move up latter where they might not jump in starting at an iPad.
4. It won't take away that many sales from the iPad.  Maybe the iPod touch, since the Kindle Fire is kind of a large type version of the iPod touch.
5. It pinches other competitors between Amazon's low price and Apple's high function.  That should discourage other businesses, not that other businesses have turned out to be much of a threat.
6. Fire is based on Android, but not marketed as such.  People will not naturally think they should have an Android phone to go with their Fire tablet (unless Amazon releases a Fire Phone).
7. Those of us with poorer eyesight (say, older folks) are not going to be thrilled with a 7 inch form factor for general purpose work.
8. Fire is going to have to pay a Microsoft patent troll tax.  But it is going to pay nothing to Google.  Enemy of my Enemy rule...
9. Apple is still going to dominate the world of apps.  People with Fires are going to envy iPads.  But not the other way around.
10. Fire's streamlined set of hardware is going to make people more impressed with what Apple packs in ... especially if the iPad 3 has a retina display.

There will be amazing sales of the device, and it will probably be the electronic gift of 2011.   But when the dust settles, Apple will still be selling iPads in huge quantities.