Enterprise Architecture seeks to divvy up another useless role, responsibilities, process, methodology, metrics, etc, etc, etc. It's another over-paid high level role for failed humans to fill-in.

C'mon, really?

Perhaps it's because we continue to fail so often and so gloriously that we feel the need for more process? Truth is, that just makes failure more expensive. The very same people which caused earlier failures should not be put into Ivory Towers and empowered to repeat themselves. We're just giving them new titles and new ways to do it.

REBOOT

A better approach to succeed is this: meet in the middle.

Let's face it - going all ivory tower is no good for anyone or anything. At the core, you have developers building the very foundation of your enterprise. We need to be close to them, understand what they are doing, and guide them properly. If you don't know what concrete (cement) is, how can you design the foundations of your enterprise? Thus I propose building the foundation whilst your BA's are gathering requirements, and let everyone do their job.

The key is: EVERYTHING IS GOING TO CHANGE

Make sure your design is adaptable, the foundation is flexible, the business and management understand that you can succeed without fully scoping out requirements, 'cause hell, they're just going to change anyway.

Have all your people doing their jobs, communication, and, you guessed it, they all meet in the middle; the high level folks, the developers, etc, they all just meet somewhere in the middle.

Of course, you need a core set of competent folk to make this work. Without that, perhaps you do need that Enterprise Architect after all? If you're going to fail, do it big!

To avoid fail, take note:

• It's all going to change - design for it
• Put good people in charge
• Stay out of their way
• Put away the project plan - just set target completion dates and have faith