Architects fly with angels but the devil is in the detail!

I’m not sure where I got this quote from (Google tells me to take the parentheses off so I now have a unique entry in Google!) but it so impressed me I put it in my CV. Fat help that was.....but you don’t care about that do you?

I think this sentence goes a long way to summing up why we go wrong so often with the delivery of solutions and it is one of the cornerstones of IT project failure. I’m going to follow this thread in the next few posts.

Architecture is one of the those words that is overused and under defined (top of my list is SIMPLE) and the thing that really gets my goat is that many people who profess to be architects rarely have any interest in the real implications of their decisions. 

This happens because there is generally too much detail to understand. And if you want to maintain your kudos in our world of technology the last thing you want to do is call on others to help you understand the detail.

So we ignore it. We hatch grandiose plans designed to ensure “executive buy-in” and “business approval” and toss them over the wall for the projects to deliver.

I was involved with a project a few years ago. It kicked off with a meeting which included all the technology providers. The Architect of the solution walked in and proudly stated that $5 million dollars worth of hardware was on its way and it was our job to build it into a functioning system. When asked exactly what was required of the solution he stated sternly the two primary objectives. As he said this there were furtive glances around the room from those who understood the technology and I gingerly spoke up and said “You realise that those two requirements are mutually exclusive for the platform you have chosen?” He looked startled, replied with “That’s not my problem it’s up to you guys to sort it out!” and promptly left the room.....

From that point on the “solution” compromises began and what they ended up with provided nowhere near the resilience they had required at the start. 

I really worry that there are not many people in the world who can take a high level view of complex IT systems and yet also have the skill and fortitude to  drill down into a level of detail that can preclude failure. When this happens in a project one of two things:

1. Plans are hatched at a high level and then when they are delivered they expose layers and layers of unthought-of implications. Outcome – under resourced projects failing to deliver.
   
2. Plans are hatched at a low level and when delivered they expose layers and layers of unthought-of implications. Outcome – under resourced projects failing to deliver.
   

Oh hang on......