Programming in the Next Decade: II

We need a new kind of browser. Today's bunch just don't cut it for developing web applications. As average bandwidth increases and the likes of Google pour PhDs into development, web applications are getting richer all the time, but forcing what is essentially a document reader to deliver close to the same user experience you can on a stand-alone application still hurts. Clever frameworks, wizards and third party controls often insulate the developer too much from what is going on.

• I shouldn't have to write my presentation layer in JavaScript when the business and data layers are written in C#.
• I shouldn't have to Ajax data from the client to the server in order to avoid writing validation logic in two languages: I want to write a validation library that the business layer can call and have the same library installed on the client.
• I want to choose the right language for my projects and know that it will be compiled, JITted or interpreted in every layer, from the client facing pages to the stored procedures.

What we need is a common language runtime in the browser with new JIT compilers / interpreters delivered as browser plug-ins. It's not exactly a new idea.