I’ve just read an interesting article from Richard of SpreadingScience: The Post=PC era will be a multi-platform era. The author’s premise is that the ease of regenerating the same application for multiple platforms will result in no single platform having dominance of the post-PC market.

While abstraction layers do allow the easy generation of native applications for multiple platforms, I believe the next era is platform agnostic. With HTML5 we are moving into an era where web applications will come into their own as the primary software vehicle.

I have the native Facebook application on my Android phone. Yesterday I clicked an email link which opened Facebook in my phone’s web browser. The experience was similar both in terms of quality and performance – I found I could happily live without the native Facebook application. This is the future.

The trend has begun to gather momentum with Microsoft following Google’s lead into on-line office applications. Office applications – text processors, spreadsheets, databases – are the core functional modules for non-entertainment computer use.

It doesn’t follow, however, that by adopting web applications for common tasks data must be stored exclusively on-line. While Google’s Chrome operating system is an exciting development in many respects, the ability to store data locally and work off-line with transparent synchronisation is a key requirement for many users.

The SpreadingScience article is entirely correct about the importance of a rapid, iterative development cycle. Publishing first cuts of web applications, even with limited functionality, provides quick feedback: either the product will be adopted by users or it will be ignored. That feedback should guide the next development step which, for web applications that can’t find an audience, may well be to start again with a fresh idea. The cycle from inception to delivery to payback becomes shorter, and involvement between producer and consumer becomes closer, to the benefit of publisher and user alike.

With these thoughts in mind, there is substantial room for improvement in available tool sets for web applications. Taking the most common platform for web development – the LAMP stack (PHP and MySQL served up by an Apache web server on a Linux box) – far too much development is still hand-coded in text editors. There are substantial and useful libraries, there are web application frameworks of increasing merit; but as yet I’m unaware of any effective two-way tools for graphically constructing applications from reusable off the shelf components. I believe this is an area of significant opportunity. We need products that deliver for web development what Delphi and VB delivered for Windows development nearly two decades ago.

I’m prepared to predict that, when my trusty Vaio laptop finally gets put out to pasture, the operating system of its replacement will be a matter of preference rather than necessity. Whatever my choice, I expect it will be capable of delivering all the functionality and applications I need via the web. That’s the future.