Feature-less is more

300+
300+ New Features – why? I would rather have one new feature: Stability. What about you?

In September 1998 Wired magazine ran an article titled “Lord of the Toasters”. I remember reading the article with a smile on my face, while hoping that a desktop operation system that didn’t crash would emerge soon. It’s now been almost ten years, and it doesn’t seem to happen in the near future.

Why?
It’s a matter of priority, making money, and what customers want (without knowing it). Or rather; what trickers customers to buy, which is not equal to what they really want.

The 4th of February Microsoft released SP1 for Windows Vista proudly telling that stability was strongly increased, from 17 to 34 hours between disruptions. No one can argue that 34 hours is better than 17 hours, but to me it’s completely unacceptable.

Somehow Microsoft and other software companies have managed to convince people that software without bugs simply isn’t possibly, and to a high extend they have succeded. Without thinking much about it, it seems to be accepted as the truth about software design, and therefore we have to settle with it.

Brakes that only fail every second month
Allow me to give an example that makes it clear that it is in fact possible to make 100% stable software.

Let me ask you a question. If the car dealer shows you two cars, priced equally, some engine, some leather seats. The only difference is that the car to the left have a nicer color and more attractive curves. Which one would you choose? The one to the left of course.

The car dealer looks at you and says: “Oh, there is one little problem that you should be aware of – the breaks doens’t always work. But it’s a beautiful car, please sign here”.

I guess you would hesitate, am I right?

Vista in your brakes
Let’s imagine that we install Windows Vista in your car to control the brakes. Most people drive about 20 hours per month, so 34 hours gives you 2 months between “break disruptions”. That’s only 6 accidents per year, nothing to talk about.

Surely, Windows Vista is a more complex piece of software than embedded operations systems (like the ones that control the breaks in your car), but it is possible to development a stable desktop OS. It’s only a matter of priority, and the priority for Microsoft and Apple is not to make a stable OS, but rather a profitable OS, and in the end it’s really not their fault. They are businesses, and businesses have to make money to please shareholder etc. Therefore they develop software that people want to buy, and that means flashy, funky, and cool new features, not stability.

Pacemakers and spaceships
Other examples are pacemakers and spaceships. Imagine a pacemaker running Windows Vista, skipping a beat or crashing completely every 34 hours. If that was the case, there would be no point in installing a pacemaker in the first place.

So who is to blame? We are. Everyone that buys, and this way financially support the companies that sell these “do not use for traffic light regulation” operation systems – and other programs that would fit in the same category.

The “buying gene”
It’s funny how we crave for design, new funky features, and a graphical user interface that looks sparkling. This is what turns us on emotionally when we see the new product demonstrated. What turns us off is when the software crash or doesn’t respond.

You can’t have both (yet)
“But I want both”, you might say. “I want a stable system that is sparkling and have 300+ new features”. Of course you do, but like everything in life it’s a matter of priority. No matter how many software developers that are employed, the number is not infinitive, and there is a limit on what can be done within a release cycle.

And this is what it all boils down to: Priority.

If Microsoft or Apple could make an operation system that was both sparkling and stable, maybe they would, but time and developer resources doesn’t permit this. And even if it was possible, they might have reasons for not doing so, since they would no longer be able to earn money on support – but this is another story.

You are to blame
So, since software houses are companies we can’t blame them for doing what they do; conduct business. It might be nice to have someone else to blame, but not being able to do it is even better.

Mark these words: Companies try to produce what people buy, not what people want (or what would make the world a better place for that matter).

So, if we really want a stable OS, what we need to do is contain our compulsive “buying gene”, and instead demand stability. My might need to do this for a couple of years or more, since it takes time to reprogram a habit, and the companies have fallen into the habit of making feature-rich and stability-poor software.

Let’s do something about it – please.