trevoro.ca | blogElegant Solutions to Complex Problems

This old list of the fallacies of grid computing needs revisiting in the new cloud computing space. I believe this list was originally written by Peter Deutsch, but I can’t find his original post.

Essentially everyone, when they first build a distributed application, makes the following eight assumptions. All prove to be false in the long run and all cause big trouble and painful learning experiences.

• The network is reliable
• Latency is zero
• Bandwidth is infinite
• The network is secure
• Topology doesn’t change
• There is one administrator
• Transport cost is zero
• The network is homogeneous

What is with all these online banks and billing systems telling me I can’t have a strong password?

The password I choose has to be between 6-8 characters and can only contain letters and numbers?

Twitter lets me have a superstrong password. It shouldn’t matter what the length is! People! What kind of old school platform are you running this shit on?

I can confirm that TD Canada Trust is among the culprits, as well as Shaw’s E-service center, and BC Hydro’s online billing system.

Has anyone else ran into a company that didn’t let you have a strong password?

People are scrambling to write applications for the new iPhone. The iPhone store model really appeals to a lot of people, and the platform for Devleoping seems really easy to use

The funny thing is, if I buy an instant messaging application for 5 dollars, will I pay another 5 dollars for one tiny new feature? Granted, there’s a lot of room to be creative, how else will people compete on feature sets and user-experience when the framework and more importantly – the amount of *useful* applications is going to be limited? If you’re the first person to roll out an application for a phone, how much will people want to pay to switch? How will they decide? The most popular applications will be the first ones in their category. Lets just take the Instant Messaging category as an example. If I can support 5 different IM services, and I’m the first to market I win. Because the next team is going to have to have 10 different services (that you WANT), a way better interface, be cheaper, and convinced you to switch based on those merits. There’s definitely a fine line between making money, and being ubiquitous – the best iPhone apps will do both.

Whenever I built lego projects I would obsess over finding the perfect piece -  The one single piece I knew in my mind I owned but could never find. Organizing and categorizing the lego wouldn’t help, because I was 7 – so every time I wanted a piece I raked through a bin of lego that at the time was roughly the size of my body.

Finding the perfect piece would take a really long time. So long in fact that I often gave up and searched for alternative pieces, and what happened has fascinated me ever since.

If I stopped looking for the piece of Lego I was originally searching for, I would find it. In fact, this worked so well that I started purposely looking for other pieces sooner to find the one I originally wanted – and it still worked.

Of course, this works for more than just Lego – you can find all sorts of different things. Try it!