Wireless Business & Technology welcomes veteran writer and technologist Charles Arehart as our newly appointed "Wireless Journeyman" columnist. Serving as "the developer's developer" he'll be examining wireless technologies and techniques, sometimes praising and sometimes challenging them. He may even change sides on a topic as things develop, but he'll always tell you why.

As you make your way into, or through, the world of wireless application development, you're going to face a lot of alternatives. You have to be careful to choose a solution that best fits your needs. The question is: Which wireless way?

The problem is that there are many interpretations of what a wireless application is all about: WAP? Java? Voice-XML? Palm apps? Pocket PCs? Bluetooth? The savvy among you will argue that these are apples and oranges - they're not the same. Bound in each are details of protocols, platforms, languages, and inherent hardware capability. But that's my very point. It's rarely an easy or obvious decision, considering all the alternatives.

I'd like to offer some insight into the underlying issues that should be considered as you debate which wireless horse to ride, whether it's:

• WAP, i-mode, Java, or some other platform
• Based on phones, palmtop devices, pocket PCs, or some other device
• Supporting content retrieval, messages, e-mail, or some other data
  Do you see that, already, we're moving the debate from any particular platform, device, or application? There's no clear choice in any of these matters, though for some applications there may be a stronger argument for one over another.

The Matter of Opinion
Still, you're going to see a lot of information and be presented with a plethora of opinions: "X is great," "X stinks," "X works there but not here." It can be really confusing, and not just for newcomers. I'd like to take some of these "sacred cows" and look at them more closely. I'll lean a little heavily on WAP as a context for the discussion, simply because it's something that a lot of people are at least familiar with. Whether you're a WAP proponent or prejudiced against it, considering the arguments more carefully can be valuable.

You're going to be relying on information from a variety of resources: articles, books, seminars, ads, sales calls, mailing lists, portal Web sites, and colleagues' opinions. Authors may or may not clearly distinguish the reasons behind their preferred approach. They may have a vested interest in the given technology, or they may sincerely be sharing what they've found to be a useful solution - or one they think makes no sense.

In the latter case, they may presume that you already understand the dilemma of why "X" is flawed. If you don't have that knowledge, you may be led to form an incomplete opinion. It may also be that you really don't share the concern that  drives them, in which case you have to be careful about making conclusions from their statements.

As an example, there's a lot of talk about WAP being crap. This isn't really an article intending to defend WAP, though in the interest of full disclosure I'll admit I am a coauthor of Professional WAP (Wrox Press), and I do still think the approach has its place. Still, there are those who will see that and immediately drop interest (in my opinion and in this article). It's almost like talk of religion or politics. People take strong positions. For many, WAP is a nonstarter. The question I think we should all ask is why.

When reading about WAP's doom, for instance, be sure to look carefully into the statements being made. The problem may not be with WAP at all.

Of course, it's now apparent that some of the slam against WAP is simple backlash from the overhyped assertions that it would provide access to the "wireless Web," with some taking from that that it would be a simple matter of using your phone to browse any Web site at all. Naturally, that was a mistake made by several parties in the industry.

Other similar "issues" include typically poor site design, "walled gardens" in phones, little use of Web customization for phone menus and registrations, and more. The bottom line is that none of these are inherent problems with WAP, but are instead a simple consequence of early adoption and the slow spread of best practices.

Even among those who see WAP services for what they're best suited to (access to targeted data anytime anywhere, possibly with location services mixed in on a wide scale someday), there's still broad argument against it. I wonder if it's always a fair attack. If we look closely, we can learn from this regardless of your favored platform.

The Network Is the Computer
Some of the most vocal arguments against WAP have been directed at the user experience. The connections are slow. It's expensive. There's broad incompatibility among the devices. Each is a valid point, but are they really WAP's fault?

When you read that people feel the networks are too slow, where is the author referring to? Europe? North America? Somewhere else? The situation is by no means uniform. Then there's the whole 3G debate, and the promise of still faster networks. Or is it that service is costly? Some phone providers offer unlimited service, while some charge very high rates. That's an issue of service provider differences. We can't really lump all phone-based wireless services into the same bin.

In some countries, wireless phones provide a far better experience than land line connections (because of costs to lay lines, unreliable service, etc.). And PCs may be very expensive. In those environments, it's suggested that people will take to wireless Web access much more readily because they have no desktop alternative to compare it to.

Some will argue that i-mode is far better than WAP. Sure, everyone knows it's got better graphics, and to some, C-HTML seems better than WML, but is that the real basis for its success? When you learn that in Japan the network is packet-switched, so the connection is always on and they don't pay per minute charges, no wonder it's taking off. That's more about the network, isn't it? If WAP phones used an "always on" network with no cost per minute, I think its success would be a very different story.

Further, what about the devices? In Europe, there was by and large mass adoption of a very limited number of phones early on, so mass opinion was formed to a great degree by that experience. Now that more devices are coming along, old opinions must be challenged.

In the U.S., there has been such a plethora of networks and standards that there are a ridiculous number of phones that may visit a site. A developer faces significant challenges supporting visitors from all possible phones. Yet in Japan there's much tighter control by NTT DoCoMo over those specifications. Again, isn't that about something other than WAP, itself?

It's the Device, Stupid
Some will argue that WAP is fundamentally flawed because of its reliance on a mobile phone as its delivery vehicle. They bemoan the miniscule display, the atrocity of entering data via a numeric keypad, and the lack of processing power to name a few. Is that a problem of WAP? No, not really. WAP's a protocol for the transmission of data. It doesn't care about the device. It's just that early devices have tended to be phones.

Naturally, those who would promote Palm apps or Java-based devices will be able to point to much more "intelligence" in their applications, and indeed in the ability to provide a local database in the client device. Will that make a tremendous difference? Indeed!

Could the WAP model, which works nearly identically to HTTP as a protocol, be used to create more powerful applications if the application could store and retrieve data from the device? You can see the foundation for such a possibility with the ability in some phones to dial or integrate with the phone book via the WTAI interface. Certainly it could be possible for an interface to be defined to allow integration with a local database.

But that may not happen anytime soon. In that case, you could argue very strongly that an approach that leverages a device like a Palm organizer, or even newer phones supporting the Java platform, will be inherently more capable. Then again, phones are morphing into PDAs and PDAs into phones.

This leads to a more significant question: What's the intended use of the device/application?

Why Are We Here?
Further dividing our camp of wireless development compatriots is the matter of intended use of the devices and platforms. Who is your audience, and what do they want to do? What kind of information do they want to get? Or provide?

That's a fundamental issue, and one that I don't think is looked at carefully to evaluate alternative solutions. If we're talking about creating a solution where the main goal is to read and respond to e-mail, I'd agree a WAP phone is a challenged choice. It's just so hard to type a response (a phone with predictive input helps, but only a little).

For that requirement, a RIM BlackBerry device (a dedicated wireless e-mail device with a full keyboard) makes a lot more sense. But that device isn't particularly well suited to running applications locally on the device. It's pretty much dedicated (albeit very effectively) to its intended purpose.

If you want to run real programs on your device, you need something with more processing power and possibly even a local data store. Again, a phone may not be the best choice (today).

That leads many to consider the Pocket PC platform, which of course leverages the tremendous ubiquity of Microsoft Windows and related applications. You know that Microsoft is working very hard to be a player in the wireless space. Indeed, their upcoming .NET initiative is wrapped entirely around ubiquitous networking and pervasive computing. Sounds like a natural fit for wireless.

Still others, wanting to fight off the Goliath, prefer a Java solution. Sun's JMEE platform speaks directly to the mobile device platform, and brings with it all the characteristics (pro and con) of an "open" architecture. Also, Symbian on the Epoch OS offers similar extensive programmability on the device.

But do you need all that power to enable simple content retrieval? A WAP phone, for all its detractors, can still be used very effectively for accessing targeted content. It works even better with effective interface design and a Web-based interface for customizing the retrieval and presentation.

Then there's the whole area of messaging, both one way ("there's an accident on your route to work") and two way (responding with a single click to a query about "are you on your way home?"), not to mention integrating messaging into business applications, and of course the ever-popular instant messaging that's taken off (to the wonder of many) among the younger crowd.

The Bottom Line
We've covered a fair bit of ground here, showing how the "commonly held conceptions" about wireless devices really do require much deeper consideration than is often the case in many resources (my article has surely left out some valuable points).

Remember, when reading the opinions of others - even in books, articles, and seminars purporting to state fact - there may be bias. More important, there may be a tendency to presume commonality that may not be there.

Experience with different networks, devices, and applications will have a major impact on the perception of the user community, the press, and the development community.

As you consider the alternatives, keep an open mind, ask questions, probe for insights, and question even popular opinions. There's often more to the story than meets the eye.

• Figure 2
• Figure 1