The iPhone v2.0 firmware update has come and gone and a lot of people have now updated. There were some hiccups along the way, but - for me at least, and I imagine most people - it went pretty smoothly.

The biggest new feature is the inclusion of the App Store, finally allowing 3rd parties to sell and distribute their own software for the phone. The iPhone has gone from simply being one of the most intuitive (and expensive) cell phones around to being a computing platform in its own right.

On the whole, the App Store itself works great. It’s easy to browse for the software you want, download and use it. It’s remarkably well implemented, in my opinion.

I still have some concerns about the manner in which Apple is treating this as a closed platform, however. While I’ve written at some length about how Apple asserting itself as the platform’s gatekeeper severely limits the kinds of software people choose to install on their own phones, there is another aspect I had not considered: How it impacts 3rd party developers’ ability to write/test their own software and be responsive to end users.

In recent years, Mac software development has worked something like this:

You spec out a project and begin writing it, usually informed by a combination of perceived demand for certain functionality among potential end users as well as your own desires for a given type of program. You put feelers our from existing customers or early adopters asking for beta testers, eventually making these people an integral part of your development process. From time to time you may get stuck on a technical challenge or hurdle, at which point you visit on of the usual hangouts for Mac developers and ask for help. The Mac development community is pretty friendly, and odds are you’ll get sample code or suggestions in short order.

This is what Mac developers are used to.

iPhone developers - many of whom come from the Mac world and are used to the above process - are faced with an altogether different type of environment.

In order to develop software for the Mac you merely have to install a set of (free) development tools and start coding. By contrast, beginning iPhone development requires that you register with Apple and agree to abide by an NDA. This gives you access to the tools, but if you so much as even want to test your own software on your own phone (versus the emulator they provide, which works well but isn’t a substitute) you have to pay $99 and wait for someone at Apple to approve you as a developer.

Once approved - and, mind you, many people are still stuck at this point - you are still bound and gagged under that NDA. Want to ask for code snippets or assistance? Good luck with that. Want to gather a bunch of beta testers out of the more adventurous of your loyal customers? You can’t. You’re pretty much on your own.

Let’s assume, then, that you eventually finish developing your killer iPhone application and Apple approves it for addition to their App Store. While (if you’re lucky) the money will start flowing in, due to the lack of beta testing odds are that your customers will uncover a number of bugs that you missed or start suggesting all sorts of new features/changes that hadn’t occurred to you.

Most good developers are going to want to push out bug fixes and tweaks as soon as possible, but there is another barrier in place to be concerned about: Apple has an approval process for each and every single update you make, even if it just entails a single line of code.

A process that was presumably intended to ensure greater quality of software for the iPhone platform has - from the outset - resulted in more than a few buggy applications.

Take, for example, NetNewsWire, a RSS/Atom feed aggregator recently released for the iPhone. The desktop version for the Mac is a high quality piece of software engineering. The developer, Brent Simmons, has a long history of writing quality code and being responsive to end-users. He’s just the sort of developer any platform needs for long-term viability.

And, yes, the mobile version of NetNewsWire has only garnered a 3 star review in the App Store, with users complaining about performance and stability concerns. I’ve even held off on using it, though I’ve been consistently impressed with the desktop version.

In a recent post to his weblog, Brent laments his inability to get the latest versions of the application (which he says fix most of the problems) out to end users. He’s still waiting on v1.0.2 to complete its review, though he’s already finished v1.0.5 and has already begun work on v1.0.6. There’s no telling when these will be made available.

Even worse for Brent, once v1.0.2 gets pushed out he may well be working on much later versions. Stale bug reports and feedback from end-users will be of little utility, referring to issues he may have already resolved already.

This secrecy by Apple is even holding up books on iPhone development. What amounts to a public plea for help from Apple by The Pragmatic Programmers indicates that would could be a valuable resource to developers is being held back for no particularly good reason. One might be forgiven for thinking that Apple wants the iPhone platform to fail.

On a practical level the new App Store is well implemented, and in the last few days I’ve really enjoyed playing with the various tools people have written for my phone. But, even more than any technical glitches or blatant omissions, this “walled garden” philosophy that Apple is adhering to has the potential to kill the platform before it ever really takes off. There are, after all, other big players waiting in the wings with impressive technology of their own.