Building Mobile Applications - What's the Right Strategy?

What's the right way to build apps for handheld devices? I've contended for some time that most enterprise apps must eventually move to the Web services model in order to minimize the cost of developing apps (write once, run everywhere) and because so much corporate data cannot be cached effectively on local devices. But I've also acknowledged that the lack of wireless connectivity (in terms of coverage, throughput, and reliability) still with us prevents the successful implementation of an IT strategy based on continuous connectivity. So, worst case, we're still going to be developing native apps for handhelds for some time into the future, and, best case, we're going to be using a hybrid local/Web strategy for most apps requiring access to remote (and, increasingly, Web-based) data.

Yesterday I spoke with senior executives from Rhomobile, a company that has a really interesting idea for developing apps for mobile devices. The key element here is Rhodes, a framework that supports the Ruby programming language on the iPhone (MacOS), Windows Mobile, BlackBerry, Symbian and Android-based smartphones. The complex stuff (UI and backend) are handled by HTML (very clever, for reasons of both portability and support of Web-based apps) and another product, Rhosync, respectively. So what's happening here is a novel (and likely very easy-to-use, although I've not tried it yet myself) way to build native apps, but with the hybrid execution model I noted above in mind. This is a very cool idea, and, oh, did I mention, all of this is open source? Extensions are encouraged. Developers pay a small royalty to Rhomobile for the use of their technology in non-open-source deliverables, but this looks, anyway, like the best of all worlds for developers of mobile apps.

• Wireless / Mobile
• mobile application development tools
• mobile applications
• Rhodes
• Rhomobile
• Rhosync