The beauty of launching an application like Walletproof on the web is its cross-platform interoperability. As long as a user has a web browser, he’s set.

In a recent post titled Facebook is the new AOL, Jason Kottke makes an interesting point about Facebook’s development platform and how its “walled garden” approach might represent a negative trend: If every big player on the web offers their own closed proprietary platform for developers to launch their products on, it becomes inefficient and expensive.

It’s difficult enough to develop for OS X, Windows, and Linux simultaneously … imagine if you had 30 different platforms to develop for.

Erick Schonfeld, editor of the excellent Business 2.0, discusses this in The Race to Become the Next AOL and he links to a story in the Financial Times where one of MySpace’s founders says they’ll probably follow Facebook’s lead and introduce their own development platform.

An example of a more constructive approach is the platform Netvibes launched recently for developing widgets: Netvibes UWA. They emphasise openness and if you write a widget for Netvibes, it will work in iGoogle, the Apple Dashboard and “many more”.

Creating walled gardens like the Facebook development platform may well make sense as a business decision. One thing is certain though, it won’t benefit the consumer.

Simon Branford, who studied high performance computing with me in Reading and is now a PhD student there, posted a link to the list of the world’s Top 500 Supercomputers. The computer they have in Reading ranked number 36 and is the fastest academic computer in the UK.

On a related, yet slightly ridiculous note, Computer World recently published a list of The top 10 dead (or dying) computer skills (via Kottke). Up there in 6th place they put C-programming:

C++ and C Sharp are still alive and kicking, but try to find a basic C-only programmer today, and you’ll likely find a guy that’s unemployed and/or training for a new skill.

C is dead (or dying)! Who’s going to break the bad news to Linus, Guido and Matz?

If you want to utilize any of the top 500 supercomputers for some hard-core number crunching, you will probably stick with C.

Here’s another tip: If you find a programmer who knows only one programming language, don’t hire him.

Cal Henderson’s Building Scalable Web Sites is a must read for anyone in the business of developing for the web.

The author is a PHP hacker and it was interesting to see that according to him, PHP has had it’s fair share of the same question people keep asking about Rails: “Does it scale?”

The distinction between performance and scalability is made very clearly in the book and a scalable system is defined as one that has thee characteristics:

• It can accommodate increased usage.
• It can accommodate an increased dataset.
• It is maintainable.

Rails passes all these tests for the same reasons PHP does. An increased dataset is accommodated in the database layer, an increased user-base is accommodated though the “share nothing” architecture and as for maintainability, Rails passes with flying colours.

My hackday contribution is finally bug free enough to be launched onto the world.

Introducting: Nabazhack.com

The application uses BBC’s APIs to know what shows are on and when the user indicates which shows he likes, his nabaztag bunny will alert him when his favorite show is about to start.

This is work in progress and will be submitted to the hackday judges at 13.00 today (however that submission is suposed to take place).

This Saturday and Sunday I’ll be at the London Hackday along with 499 other developers, trying to cook up something funky with Rails.

Appropriately I’ve added my name to the wiki, signed up to the backnetwork and updated my backnetwork profile.

Along with the developers, a bunch of interesting companies will be showing up, like O’Reilly, IBM, Moo and last but not least, Nabaztag!