Software, Interrupted

I switched to the iPhone 3GS the day it came out and I'm still waffling on how I feel about it. So far, the positives still outweigh the negatives, but my work style has changed and I am able to deal with a lot more mobile device flakiness than I was able to a few months or a year ago. If I was still traveling and running around all the time, the iPhone would be a total disaster.

From a software and cloud perspective, the iPhone represents an ideal world of development functionality mixed with an ability to use mobile services. However, the App Store approval process and AT&T's wonky network will still prevent us from reaching nirvana.

I tried to chronicle the issues I've had, but the truth is, the service (and therefore the device itself) ranges from excellent to sporadic to unusable, so I'll just list out the broad issues for those considering a move to the 3GS.

Battery life--the battery life is abysmal. I've gone on every forum, tweaked every setting, and done several tests to see what works best. The hacks that people suggest (turn off push, 3G, and Bluetooth) defeat the purpose of the device. Users shouldn't have to handicap themselves because of a lack of attention from the manufacturer.

If Apple really wants enterprise and business users, this is the most important issue that must be resolved.

Phone --I haven't been a fan of AT&T mobile service in the past (Verizon Wireless is my carrier of choice) and it continues to range from terrible to mediocre for me. Bizarrely, the worst call quality occurs when talking to another AT&T user. I did a conference call the other day with two other iPhone users and none of use could decipher the others' words.

I gave up on the 3G network both for the battery drain and the calls that dropped after 10 minutes (I took notes for 2 days) and every call dropped. A recent survey says that 34 percent of those polled won't buy the iPhone because of AT&T's network.

... Read more

Correction: This report misstated which BlackBerry version recently got a big marketing push from Verizon. It was the Storm.

New data from NPD Group suggests that RIM may have caught up with some of the iPhone marketing hype, taking the top spot in U.S. consumer smartphone sales for the first quarter of 2009. The BlackBerry Curve (of which there are several models across multiple carriers) bested the iPhone for the first quarter of the year, with RIM taking three of the top five spots.

We get a lot of Apple fanboy grief here in the CNET Blog Network, but I'm a BlackBerry user. Personally, I prefer the BlackBerry keyboard and form factor but feel that the iPhone interface and applications are superior.

But more important than the applications or the interface, I need my phone to work. I want it to be able to make calls, receive calls, send e-mail, etc. The iPhone, for all its glorious features, is at best a mediocre phone with occasionally terrible coverage.

AT&T, the lone iPhone carrier in the United States, has been slow to fix network issues and slow to respond to customer complaints, and it lacks a certain amount of customer service social grace. Most of the gadgety or techie types of people I know who don't use the iPhone avoid it entirely because of AT&T.

Realistically, there should always be more BlackBerrys sold than iPhones simply because of network diversity. While the iPhone may be acceptable--even good as a business smartphone, the spotty coverage and weak customer service makes the device a questionable choice for on-the-go business users.

The Blackberry Storm got a big marketing push from Verizon that no doubt helped grow the customer base, but the Storm is not an iPhone killer.

I'm looking forward to seeing what RIM has to offer in the future, as well as seeing if/when Verizon will finally get the iPhone. Until then, I'll stick with the BlackBerrys, which, despite the occasional random java error and simplistic user interface, have served me extremely well for the last five years.

Follow me on Twitter @daveofdoom

Just as Sun announced the latest rev of JavaFX, AT&T announced that they would drop Java-based phones, favoring Symbian.

Tom Krazit reported earlier today on the news, quoting Roger Smith, director of next-generation services at AT&T who said "Java has not been a success," Smith said. "It's not because Java is bad, but we didn't manage it effectively."

The basic premise of the argument against Java is that it became fragmented with every manufacturer using its own version and defeating much of the purpose of Java as a platform to begin with.

And as mobile phones have started to become more and more like mobile computers, the software on those phones needs to become more and more sophisticated to run intriguing applications, Smith said. Java doesn't reach down far enough into the lower levels of the phone to exploit hardware in the manner that full-fledged operating systems do, he said.

I find it a bit curious that Smith felt that Java wasn't capable of exploiting the full hardware functionality. That sounds like more of a development issue then it does a functional requirement. Arguably the operating system should be abstracted to the point where Java can do what developers need it to.

As Stephen Shankland wrote about JavaFX today:
JavaFX also comes with a slick feature, the ability to move running applications out of the browser and onto the desktop--and back, if desired. Essentially, they can change their nature and abilities according to where they're housed. And the same application also can run on JavaFX Mobile, holding the promise for programmers that they won't have to endlessly rewrite the same applications for different media.

JavaFX looks to be very powerful as witnessed in the video below by Redmonk analyst Michael Cote. But the question is one of relevance. It's not clear that developers are terribly interested in JavaFX to begin with and the dream of monetizing mobile phones is clearly out the window with the largest wireless carrier.

I was checking out the Gizmodo review of the new BlackBerry Curve 8900 (formerly Javelin) and wondering why Research In Motion insists on putting the majority its new devices into the clutches of AT&T, the exclusive iPhone provider, first.

RIM recently announced an application store, as well as the new Storm, which is the BlackBerry rival of Apple's smartphone, and plans to soon release this Curve 8900. The products (and soon services) have never been better, and yet it would seem logical to use one of the carriers that doesn't have the iPhone as the point of entry.

Wouldn't it make more sense to launch the majority of devices on Verizon Wireless or Sprint, where there is no iPhone competition? Or at a minimum create some exclusivity like AT&T has with the iPhone?

Verizon is getting the Blackberry Storm first, but that's still not enough. Network issues (GSM, CDMA, etc.) can all be resolved if RIM wants them to be. This seems to be more of a marketing and product strategy issue than a technical one.

I suppose the argument is that in order to compete with the iPhone, BlackBerrys have to also be available from AT&T, but I'm not sure that such a strategy makes sense.

Anyway, Giz likes the new Curve, but the Bold still seems like a better device.

The Bold is more substantial, exuding "executive" to the Curve 8900's "middle manager," or "normal person." It's clearly more powerful, and has a bigger screen. The keyboards are way different, too--personally, I prefer the Bold's larger, squishier keys to the Curve's smaller, stiffer ones. But obviously, the biggest thing is 3G. The Bold has it; the Curve 8900 doesn't, and we missed it sorely.

Storm aside, the Bold will be on AT&T first. And users will be stuck with GSM as their only option (which I've found to be surprisingly bad in the US.) Verizon announced the BlackBerry Storm, with global 3G, to be launched this fall, there is no word on when it will actually land.

Meanwhile, the iPhone continues to get better, and people care less and less about switching to AT&T just so they can have the device. When was the last time someone switched to Verizon or Sprint to get a BlackBerry? I am sure the Storm will do well, but I doubt Verizon will get the iPhone uplift that AT&T saw in the last quarter.

I went out and got myself a shiny new iPhone 3G yesterday from the AT&T store on Geary Street in San Francisco (which was only out of stock once, they said.) The cell phone buying process is typically akin to some circle of hell, but overall I had a very pleasant experience.

I've been accused of being an Apple fanboy a number of times, but I have stuck with the BlackBerry both for functionality and Verizon's network. I wanted to test some of the features (and play the games) on the iPhone and decided to give AT&T another shot.

I had been waiting for the 2.1 firmware upgrade before I got the phone because in the past I couldn't even get one bar in my house in San Francisco. All reports said the problems were solved so I went for it. When I got home I checked and I did have 3 to 4 bars pretty consistently. I made a test call which took about 10 seconds to initialize and all seemed well.

I plugged the iPhone into my MacBook Pro, iTunes launched and the set-up process began. And ended. Three times. Then suddenly it worked and I was registered with Apple and AT&T.

I decided to grab a few things from the App Store's free section. Every item failed multiple times. After my fifth attempt on Pandora, iTunes told me I already had it, and so I did.

Hard to believe it was an Apple product that was going this wonky on set-up.

My quick take? The iPhone is good, but not great. It doesn't replace a computer and has enough idiosyncrasies to deter the business user from choosing it over a BlackBerry. For consumers it's great, provided you can get AT&T to work for you.

The good stuff:
- The applications are fantastic and the user experience is truly unmatched
- Sound is great
- Integration with the Mac is great
- Camera is surprisingly good
- The onscreen keyboard is usable (though I still prefer the BB keypad)

The problems:
- AT&T network is shockingly bad. Verizon blows it away by a huge margin.
- Launching applications is slow (like three to four seconds)
- I have yet to get 3G coverage in San Francisco. It only worked for me in Burlingame, Calif., when I was down there for a meeting
- The time it takes to initialize a call is laughable. I did a test of the iPhone vs. my VZW BlackBerry and I had already finished the conversation before the iPhone had connected.
- For some reason Mail won't consistently log in into Gmail. I don't know which app is to blame.
- Shifting from portrait to landscape is highly inconsistent

I'll give the iPhone the full 30 days before AT&T punishes me for rejecting them. My hope is that I find the positive to outweigh the negative. So far I'm evenly split.

I wrote yesterday that there will be more than one Cloud vendor and today AT&T announced that they are the latest. Gloating aside, this proves that the market is still being shaped and that it won't necessarily be the domain of BigCo tech companies.

In reality it makes perfect sense that major telcos would become Cloud providers. They own the bandwidth, network and have data centers all over the world.

AT&T and Sprint have long offered hosting services including data center space, managed servers and services and of course network connectivity. Data centers are expensive and require ongoing maintenance. The expense of such build-outs will not be easily dealt with by an upstart provider.

This also suggests that the Cloud definition will be stretched further and that there will be more mutations of public and private Clouds depending on who the customer is.

NY Times is reporting that Comcast voice service sounds better than anything else, including AT&T landline and Vonage.

The differences between the two kinds of digital voice services were brought out sharply in a study released last month by Keynote Systems, an Internet testing business based in San Mateo, Calif. Placing 75,000 test calls among various phone services, Keynote found that not only were Comcast's audio scores superior to those of Vonage, but that Comcast also scored higher than even AT&T's landline service.

Of course, as Randall Stross, the author of the article notes, the service needs to be up for your phone to work.

A strange thing happened at my house, however. No more than five minutes after I typed "reliability hasn't been a significant issue ..." my cable modem's lights blinked out. No connection. No dial tone. I told my wife about the funny timing.

I'm still the only Mac guy I know who doesn't have an iPhone, primarily because AT&T service doesn't work in my house in San Francisco (it must be the fog.)

I still like my Blackberry 8870 (on VZW) despite the fact that it feels like I am operating in 1998 and all my friends are in 2008. It would be nice if RIM would modernize these things, especially for we addicts who prefer an outside keyboard. The new 9000 series seems to go a ways toward that until the carriers jump in and screw it up. Sorry VZW, your service is great but your user interface "enhancements" ruin all the work that the manufacturers do toward usability.

If you want to read more about the market dynamics around RIM and Apple, today's NY Times has what feels like an interminably obvious piece for your enjoyment.