Thursday, February 4, 2010

New definition of enterprise architecture emphasizes 'fit for purpose' across IT undertakings

Listen to the podcast. Find it on iTunes/iPod and Podcast.com. Read a full transcript or download a copy. Sponsor: The Open Group.

This live event podcast discussion comes to you from The Open Group’s Enterprise Architecture Practitioners Conference in Seattle, the week of Feb. 1, 2010.

We examine the definition of enterprise architecture (EA), the role of the architect and how that might be shifting with an expert from the Open Group, Len Fehskens, Vice President of Skills and Capabilities. The interview is moderated by Dana Gardner, principal analyst at Interarbor Solutions.

Here are some excerpts:
Gardner: I was really intrigued by your presentation, talking, with a great deal of forethought obviously, about the whole notion of EA, the role of the architect, this notion of "fit for purpose." We want to have the fit-for-purpose discussion about EA. What are the essential characteristics of this new definition?

Fehskens: You'll remember that one of the things I hoped to do with this definition was understand the architecture of architecture, and that the definition would basically be the architecture of architecture. The meme, so to speak, for this definition is the idea that architecture is about three things: mission, solution, and environment. Both the mission and the solution exist in the environment, and the purpose of the architecture is to specify essentials that address fitness for purpose.

There are basically five words or phrases; mission, solution, environment, fitness for purpose, and essentials. Those capture all the ideas behind the definition of architecture.
Also from the conference: Learn how The Open Group's Cloud Work Group is progressing.
Gardner: The whole notion of EA has been in works for 30 years, as you pointed out. What is it about right now in the maturity of IT and the importance of IT in modern business that makes this concept of enterprise architect so important?

Fehskens: A lot of practicing enterprise architects have realized that they can't do enterprise IT architecture in isolation anymore. The constant mantra is "business-IT alignment." In order to achieve business-IT alignment, architects need some way of understanding what the business is really about. So, coming from an architectural perspective, it becomes natural to think of specifying the business in architectural terms.

We need to talk to business people to understand what the business architecture is, but the business people don't want to talk tech-speak.



Enterprise architects are now talking more frequently about the idea of "business architecture." The question becomes, "What do we really mean by business architecture?" We keep saying that it's the stakeholders who really define what's going on. We need to talk to business people to understand what the business architecture is, but the business people don't want to talk tech-speak.

We need to be able to talk to them in their language, but addressing an architectural end. What I tried to do was come up with a definition of architecture and EA that wasn't in tech-speak. That would allow business people to relate to concepts that make sense in their domain. At the same time, it would provide the kind of information that architects are looking for in understanding what the architecture of the business is, so that they can develop an EA that supports the needs of the business.

Gardner: So, in addition to defining EA properly for this time and place, and with the hindsight of the legacy, development, and history of IT and now business, what is the special sauce for a person to be able to fill that role? It’s not just about the definition, but it's also about the pragmatic analog world, day-in and day-out skills and capabilities.

Borrowed skills

Fehskens: That's a really good question. I've had this conversation with a lot of architects, and we all pretty much agree that maybe 90 percent of what an architect does involves skills that are borrowed from other disciplines -- program management, project management, governance, risk management, all the technology stuff, social skills, consulting skills, presentation skills, communication skills, and all of that stuff.

But, even if you’ve assembled all of those skills in a single individual, there is still something that an architect has to be able to do to take advantage of those capabilities and actually do architecture and deliver on the needs of their clients or their stakeholders.

I don't think we really understand yet exactly what that thing is. We’ve been okay so far, because people who entered the discipline have been largely self-selecting. I got into it because I wanted to solve problems bigger than I could solve myself by writing all code. I was interested in having a larger impact then I could just writing a single program or doing something that was something that I could do all by myself.

That way, we filter out people who try to become architects. Then, there's a second filter that applies: if you don't do it well, people don't let you do it. We're now at the point where people are saying, "That model for finding, selecting, and growing architects isn't going to work anymore, and we need to be more proactive in producing and grooming architects." So, what is it that distinguishes the people who have that skill from the people who don't?

An architect also has to be almost Sherlock Holmes-like in his ability to infer from all kinds of subtle signals about what really matters.



If you go back to the definition of architecture that I articulated in this talk, one of the things that becomes clear is that an architect not only has to have good design skills. An architect also has to be almost Sherlock Holmes-like in his ability to infer from all kinds of subtle signals about what really matters, what's really important to the stakeholders, and how to balance all of these different things in a way that ends up focusing on an answer to this very squishily, ill-defined statement of the problem.

This person, this individual, needs to have that sense of the big picture -- all of the moving parts -- but also needs to be able to drill in both at the technical detail and the human detail.

In fact, this notion of fitness for purpose comes back in. As I said before, an architect has to be able to figure out what matters, not only in the development of an architectural solution to a problem, but in the process of discerning that architecture. There's an old saw about a sculptor. Somebody asked him, "How did you design this beautiful sculpture," and he says, "I didn't. I just released it from the stone."

What a good architect does is very similar to that. The answer is in there. All you have to is find it. In some respects, it's not so much a creative discipline, as much as it's an exploratory or searching kind of discipline. You have to know where to look. You have to know which questions to ask and how to interpret the answers to them.

Rarely done

Gardner: One of the things that came out early in your presentation was this notion that architecture is talked about and focused on, but very rarely actually done. If it's the case in the real world that there is less architecture being done than we would think is necessary, why do it at all?

Fehskens: There's a lot of stuff being done that is called architecture. A lot of that work, even if it's not purely architecture in the sense that I've defined architecture, is still a good enough approximation so that people are getting their problems solved.

What we're looking for now, as we aspire to professionalize the discipline, is to get to the point where we can do that more efficiently, more effectively, get there faster, and not waste time on stuff that doesn't really matter.

I'm reminded of the place medicine was 100 or 150 years ago. I hate to give leeches a bad name, because we’ve actually discovered that they're really useful in some medical situations. But, there was trepanning, where they cut holes in a person's skull to release vapors, and things like that. A lot of what we are doing in architecture is similar.

What we want to do is get better at that, so that we pick the right things to do in the right situations, and the odds of them actually working are much higher than better than chance.



We do stuff because it's the state of the art and other people have tried it. Sometimes, it works and sometimes, it doesn't. What we want to do is get better at that, so that we pick the right things to do in the right situations, and the odds of them actually working are much higher than better than chance.

Gardner: Okay, a last question. Is there anything about this economic environment and the interest in cloud computing and various sourcing options and alternatives that make the architecture role all the more important?

Fehskens: I hate to give you the typical architect signature which is, "Yes, but." Yes, but I don't think that's a causal a relationship. It's sort of a coincidence. In many respects, architecture is the last frontier. It's the thing that's ultimately going to determine whether or not an organization will survive in an extremely dynamic environment. New technologies like cloud are just the latest example of that environment changing radically.

It isn't so much that cloud computing makes good EA necessary, as much as cloud computing is just the latest example of changes in the external environment that require organizations to have enterprise architects to make sure that the organization is always fit for purpose in an extremely dynamically changing environment.
Listen to the podcast. Find it on iTunes/iPod and Podcast.com. Read a full transcript or download a copy. Sponsor: The Open Group.

You may also be interested in:

No comments:

Post a Comment