If you’re going to land on the Moon, at some point you have to land on the Moon

The recording of the Apollo 11 Moon mission, just before the Eagle lands in the Sea of Tranquility, never fails to make my hair stand up. The tone of the astronauts is matter of fact, calm and professional: if you didn’t know the context then you would never imagine that they were engaged in one of humanity’s greatest endeavours.

And the more of the context you know, the more astonishing that air of calm becomes. Those last minutes before the landing were filled with alarms, unexpected behaviour from the lunar module, an overshoot of the planned landing site, and a search for a favourable place to land. By the time the module touched down, the craft had less than 5% of its fuel left, and was less than thirty seconds away from abandoning the mission. (This page gives a full transcript of the last 13 minutes before landing, and a great commentary on what was going on behind the words.)

If that wasn’t enough, we must also remember that, prior to landing on the Moon’s surface, there was no way of knowing exactly what the surface was like, or whether it was even capable of supporting a safe landing. Despite the speed with which the Apollo missions were developed, pretty much all other aspects of Apollo 11 had been tested: the craft had been launched from the Earth’s surface to orbit, the lunar module had been extracted, and the moon had even been orbited in a crewed mission. The Apollo 11 crew had been through countless simulations. But the only test for landing on the Moon was to land on the Moon.

I doubt that anybody reading this article will have the opportunity to achieve the same types of historic firsts as the Apollo astronauts (and if you have had such an opportunity, please leave a comment!). But those of us working in enterprise technology will all have the opportunity for firsts in our career.

We will deploy a new technology for the first time. We will deploy a system for the first time. We will migrate data for the first time. We will expose our system to end users for the first time. We will publish a site, an application - or an article - to the Internet for the first time. We can - and should - test all of these things extensively. But the first time will always be different from the tests, and, it is likely that we will learn more from that first time than we learn from all the tests. Every first is a chance to learn by doing.

As I work for a Cloud company (although the views in this article are my own), I have the privilege of seeing many companies do things for the first time. The public Cloud industry is still very young, platforms are evolving all of the time, and most companies are still in the early stages of Cloud adoption.The field is full of first steps. In taking those first steps, it is right to look at what others have done, to plan, to design and to test - but we also learn most from actually taking those first steps. 

Enterprise technology is not as exciting as space exploration, and does not contain the same monumental achievements as the Moon landings. But it does have the similarity, that, however much you test, however much you prepare, at some point you have to take that first step, and learn by doing.