Mozilla has announced Prism, an effort to bridge the gap between the Web and the desktop.
Mozilla Labs is launching a series of experiments to bridge the divide in the user experience between web applications and desktop apps and to explore new usability models as the line between traditional desktop and new web applications continues to blur.
WebRunner has been picked up as the first candidate for the experiments. I have been using Webrunner for some time now, and it has been very useful. It is also quite useful in isolating certain applications from rest of your surfing.
The whole effort seems to be directed towards treating web applications different from web sites, and treat them along the lines of the desktop applications, if not the same. Of course this will give the desktop user a familiar start with using those applications, like double-click to start the application, a separate icon, integration with the start menu, the isolation, etc. One observation I have is that some sites depend on the browser’s back and forward buttons and are not efficient when run in WebRunner.
How does this compare with AIR and Silverlight? I think Prism will be more tightly coupled to the Web. At least today, Webrunner is just another way of deploying the web site, whereas AIR and Silverlight are technologies that have their own development and runtime environments, separate from the Web. With Prism, it is still developing for the Web. For the end user, there might not be much difference superficially, but it will be more apparent when the applications will be more widely available. As Ian Bicking says, the users might be able to make any web page into an application.
I think this will also lead to good technology for building smart clients. This is also similar to the webified v/s browsered applications in the sense that it deals with bridging desktop and the Web. I still side with the webified applications when we try to fit everything inside the browser. However, Prism’s approach is different by making web applications easily accessible from the desktop. It will be interesting to see how it will handle the exceptions, disconnects and downtimes. However, I have a feeling that just like the browser seems inefficient some times, this approach will be suitable only for certain class of web applications, not all.