I’ve been having conversation recently on what the future of enterprise architecture should be in the context of a world of microservices. Or more specifically, is there a future for it?
In a world in which all the capabilities that an organisation needs are captured in a set of microservices, and where the microservices teams themselves are responsible for making the right technology choices, what’s the value of enterprise architecture? For those of you who haven’t bumped into microservices, here’s a great talk by Martin Fowler.
Let me start by painting some context. About a year ago I did a talk to a group of architects where I blamed them for many of corporate IT’s woes. Although that was unfair, my aim was to get them to think differently about how they really added value to the work of other teams. The reason being that I think the problem may be with architects, not architecture. Or more particularly how architects go about the business of architecture.
My main point was that too many architects still have an obsession with technology homogeneity. Once upon a time, this might have been helpful, but these days it’s not. If Tim Berners-Lee had worried about homogeneity, we wouldn’t have the World Wide Web today. He realised that you could achieve a nirvana of interoperability by adopting a minimum set of standards, HTTP, HTML and URL, and allowing everything else to be locally defined. In a microservices world, you’re likely to get much further, much faster, by embracing this type of simplicity and diversity.
So is there any value in enterprise architecture in a post digital world?
Of course there is. Individual teams naturally think individually. To take another example, it’s not hard to create one agile team, and indeed it’s not hard to create thirty agile teams. But those who have attempted to scale and integrate agile in their organisations know, thirty independent teams won’t naturally come together into a coherent framework that makes the whole greater than the sum of the parts. Far from it.
The value in enterprise architecture comes from having a breadth of view across the whole landscape, and not just to the edges of the organisation, but beyond them too. And not just a view of today, but a view of the changing technology landscape over time.
But the last thing we then need is reams of technology standards, issued centrally. Local teams need to be empowered to make the right technology choices. For their part, these local teams need to make those choices within a context where diversity is driven by actual need, not by the latest fashion. Enterprise architecture adds value by ensuring that we have the thinking that allows scaling, interoperability and efficiency, both in what we do and how we do it.
But that’s EA 101 for the enterprise architecture team. The real value is in helping the whole organisation make sense of a rapidly changing landscape, where the pace of change is accelerating. The real challenge for enterprise architects is to speed things up, not slow them down.