In the early 1990s, client-server computing was all the rage. But it sucked, because networks were too slow and because updating client software was unreliable. Then the web happened, and soon, applications were we written for web browsers. It was a great time, for a while.
Client server is back, and it’s now – arguably – the dominant computing model today.
I’m talking about smartphones (iPhone/iPad, Android, Blackberry, etc) with their app stores.
Smartphones are outselling laptops. And while web surfing is popular among smartphone users, app stores is where it’s at.
Smartphone apps are the new client server model. The protocols are better (POX – plain old XML) and more efficient, HTTPS for security, and bandwidth is better. The entire mechanism for updating smartphone apps is reliable, semi-automatic, bandwidth friendly, and easy to use.
I’m not knocking web browsers, really. They are great and getting better. But the differences between them is making the development of web applications that work across all of the web platforms and versions increasingly difficult. The web is great for lightweight application interaction, but it’s difficult to get it right in complex applications. Making web apps work across the popular browsers, versions, and OSs is not unlike the unenviable job Microsoft has of making Windows work on everyone’s Intel-based system. In the early days of the web, you wrote HTML and it worked everywhere. Not so any more. The bloom is off the rose.
So, what about app stores for laptops / desktops? Since app stores are accepted by smartphone users, it makes sense that we’ll see them on laptop and desktop operating systems (Windows, Mac, Linux). If you use OSX (Mac), it’s already here. Microsoft is late to the party. Again.
I believe we will see a resurgence of client-server computing in the form of app stores for all major computing platforms, and that serious business applications that were previously web based will be app-based. Just like in the old days, only better.
Peter H. Gregory 