The Messenger project - some thoughts on building the perfect thing

If you've been reading The Daily Upload for awhile, you'll know that I have a soft spot for technology, which includes cool stuff that happens in space.

So that's why I'm pointing you to this cool collection of photos in The Big Picture about NASA's Messenger project, which is a mission to Mercury, the closest planet to the sun in our solar system.

It was visited by the Mariner 10 spacecraft twice in the 1970s, and about 45% of the surface was mapped. On August 3rd, 2004, NASA launched a new mission to Mercury, the MErcury Surface, Space ENvironment, GEochemistry and Ranging probe (or MESSENGER). MESSENGER is now in the last stages of multiple gravity-assist flybys of Earth, Venus and Mercury, en route to an insertion into orbit around Mercury in March of 2011. In just two flyby encounters, MESSENGER has already greatly increased our knowledge about Mercury's surface features. As you look at Mercury in the new images below, keep in mind that it has minimal atmosphere, gravity about 1/3 of Earth's, and surface temperatures ranging from -183 C (-297 F) in some polar craters to 427 C (801 F) at high noon (Mercury's solar day lasting 176 Earth days)

Here's the link to the pictures:

As I was looking through this collection, I remembered a conversation I'd had about the inherent nature of creating computer code and whether you could build a bug-free program.

Bugs in software code are a fact of life. Think of all the security problems that we've had over the year with bugs in computer software being exploited by hackers. It's not surprising that people think that errors are inevitable.

But the truth is, if you have enough time, money and patience, you can create code without errors. And these space projects that NASA does are a good example. So are satellites that orbit the earth. They are built with amazing hardware, but the missions run with computer code. And there is no way that you can just fix the bugs that might pop up. They can't have any bugs - period. And they don't. When something has to be absolutely bullet-proof, they can build it. It just takes longer and it's really, really expensive.

The problem is that when you build something to be perfect, it is, by definition, only perfect when you build it. It may do exactly what it's supposed to, but it will not do anything else. So in a few months, or a few years, you will still have something that does exactly what it's supposed to do, but it will also be out-dated and unlikely to meet your current needs.

That's OK if you're a spacecraft floating through the galaxy, but it sucks if you're a business trying to keep your customers satisfied.

What technology companies have come to realize is that getting a product out to the customers, and letting them have a go at it (as with Google's BETA projects) is the way that things will work from here on out. No one really knows how a product will be used until people are out there using it. And the manufacturer needs to be able to adapt to those uses or face losing out to another product that can do it better.

That's the new mantra of the consumer economy. Companies will have to adapt and realize that their customers call the shots - not their engineers.

I still like things that were built to do one thing well - like typewriters. But I'm glad that we've got new toys to play with that can adapt to our needs, not force us to adapt to theirs.

I'm not sure how this line of thought links back to the Messenger project. But I'll work on it and if I come up with something, I'll let you know.