Few weeks ago, I was on business trip, dining alone at my hotel restaurant in Gothenburg (sad story isn’t it ;)) I was using my favorite device: iphone 4 to read interesting Enterprise Architecture articles & papers, when, suddenly, my attention was caught by a direct reply on one of my tweets from my respected architect colleague: Roger Sessions. Roger asked me:
Which at that time, I hadn’t read yet. So, I decided to read it carefully, as it deserved to be and give some of my feedback / thoughts to Roger and you through my blog, since a tweet would not be enough.
So, to start with, not to mention that you should spend the valuable time to read Roger’s paper, which I don’t want to re-write here. Let me introduce it to you quickly. The title this paper is: “Comparison of the Top Four Enterprise Architecture Methodologies” To be honest, this title is sufficient by itself to summarize the purpose of the document.
Through this paper, Roger gives us four very good overview of the top four framework/process/methodology/practice: Zachman, TOGAF, FEA & Gartner, but in addition to this he also gives us some clues of what does each is good at (and not that good at, as well).
This said, here are my points:
When taking each of these frameworks/process… separately, I always felt uncomfortable. While reading, I remembered when I was seating in TOGAF 9 training, having the feeling that there were things missing. I couldn’t explained it at that time, then I was missing experience in Enterprise architecture field and couldn’t step back enough.
As an example:
To my point of view, Zachman framework is more a reference to which you should compare with. What for? To benchmark in which Zachman cells you are currently missing documented knowledge in your EA work. Of course, this is not enough, once you did this first step; you need to set your priority accordingly to your strategic business objectives.
So if you take Zachman only and try to use it, first thing you will get hit by is that you are missing a process to do it… This is of course, where TOGAF is coming into the picture, bringing the process… So TOGAF is completing Zachman, good… but not enough – that would have been too easy –
Then comes FEA which brings a methods, yes, ok… but… still not enough. So there it comes, the big one: Gartner! Hurrah! We finally get it all, right? Of course not! But why? will you ask me! We have a reference model, a process to get the architecture up and running, and methodologies and then top of the world EA specialists… Well, there are different reasons why, let me gives you the main ones I foresee, with the help of Roger’s paper.
Why does each of the Top Four methodologies are not enough (taken separately)?
Roger Sessions: “TOGAF merely describes how to generate an enterprise architecture, not necessarily how to generate a good enterprise architecture.”
This said, there is nothing much to add about TOGAF.
To continue, a general comment on FEA, Zachman. These 2 are IT oriented frameworks / methodology (there is also a debate about TOGAF, but let’s no opening it here now). So, OK, we know that, but the issue I see is not that they are IT oriented, but the issue is that none of them are fitting with 201x enterprises’ challenges.
Zachman and FEA were designed to answer 1980’s problems and challenges. When it comes to TOGAF, as we’ve shown it above, it doesn’t answer to any other challenge than: create the architecture. Not saying that creating the architecture is not useful, but it is architect matter that is addressed, not the CEO’s challenges, such as: “enterprise profitable growth”.
So… you will tell me: “We need another framework / process / methods…” Well, here is a debate that deserves to have its own post (coming soon)… We already have plenty of these (remember that here that Roger took only the top four used ones) and I don’t want to re-invent the wheel again, but obviously, based on what I just described taken separately each of these 4 attempts is not enough. So, we need something to help us to manage the complexity of the “thing” (enterprise in our case), to fullfill the current challenges our enterpises are facing today and to get the Enterprise Architecture moving forward.
“Get the Enterprise moving forward (with the help of the EA)”
How should Enterprise Architects help their CEOs to get their enterprise moving forward? To start with:
Then, once it is done, let’s us come back to frameworks, process, methods, best practices… when it comes the time to choose, you will have difficulties to pick one of these since they are always missing one aspect. Then… what to do? In a first step, what is important is to know these methodologies, understand what they are capable to offer you. Then, the second step is to find your own way.
“Find your own way”
I know that for a structured mind as an architect is usually provided with, this statement will not sound “satisfying”. So let me bring you few additional points here:
- The “best practice” is always your practice (because it’s yours!) Who else than you should know better than you what you need? Of course, you might need help to express it, we all need such help from time to time, but at the end, you must be the one knowing what you want to do, right?
- When it comes to choice and getting thing done. This is where we should stop (for a while) to structure things. Remember Gartner quote: “Just enough Enterprise Architecture, just in time”. To me, this also means that we need to keep space for “not structured thinking” (cf. my post about non-linear thinking) in order to keep freedom for creativity and get innovative.
Because yes, innovation is one of the KEY for your enterprise to get through and progress. Let’s us stop here for today. Next time I will tell you more about “my way”…