By Stuart Boardman, Getronics
Right now work is starting on the next major release of TOGAF®, which for now is known as TOGAF® Next. That makes it a very good time to look at what else is going on in the world and what kind of contribution that might make.
A lot of the best ideas come from unexpected directions. Enterprise architects (fortunately) often have passions that don’t have much directly to do with that discipline. Let’s be honest, the best ones almost always do. Peter Bakker recently drew our attention to a current debate in the world of photography and photo journalism. People are using apps like Hipstamatic to make deliberately grungy images – to make the results less “realistic” and more “impressionistic” (same thing Claude Monet and his pals came up with in the late 19th century except they didn’t have apps back then). Apart from the intrinsic interest of the topic, Peter suggested this might be applicable in EA. That made me think. We’ve invested vast amounts of time and effort (and therefore money) in being able to specify things in enormous detail according to increasingly tightly defined models. In fact, people used to complain that those tight models were what TOGAF® lacked. Hmmm. Sometimes the result is not seeing the wood for the trees. Or assuming that detail equals fact. Or getting realism muddled up with reality. Or information with knowledge (never mind wisdom). The Impressionists wanted people to be able to get a feeling of what it was like to be there — not precisely what it looked like at a specific moment in time. So while I’m sure they weren’t thinking about quantum mechanics (that would have been quite an achievement!), they were certainly leaving things open for probabilistic interpretations. Could we do the same in EA – without just producing vagueness? Why not – at least down to a certain level? If you use the Business Model Canvas, for example, you can build up a very meaningful picture of an enterprise’s business model without vast amounts of detail. It provides a lot of knowledge and even some wisdom on the basis of an optimal amount of information. And that has the great benefit of allowing you to fill in the detail where it’s actually going to be useful to you. So why wouldn’t we do something similar in general in EA?
Ross Button is developing an idea he calls Scatter Architecture. You could visualize it as a lot of puzzle pieces that you scatter on a board and see what kind of a picture you can make out of them. They might turn out to fit together in more than one way. That’s actually a good thing, as it probably makes you more adaptable and less exposed to change. Some of the pieces will duplicate each other wholly or partly. Viewed from a TOGAF® perspective we can say that these duplicates occur both on the Enterprise Continuum and on the Solution Continuum. Duplicates are allowed in this architecture. I don’t suppose you’d find them in the Enterprise Strategy or in the Architecture Strategy but you might well find partial duplicates among your propositions, activities, resources and partners – particularly the latter. After all, you probably don’t really want to be dependent on one supplier but that doesn’t mean they’re all exactly alike. So your architecture strategy might even codify that, which means your architecture models will need to take account of it. On the solution side of things it’s just as likely. Ross has explicitly pointed to Cloud as an example of this. Just as in the “real” world, if you can avoid being locked into just one supplier (without the cost implications being too high), you have much more room to manoeuver. The Amazon crash a couple of months ago provided some good positive and negative examples. Moreover, just as in the “real” world, these partners might become part of your value creation process as opposed to just cost elements. So this introduces my second theme, multiplicity.
Louisa Leontiades has just launched a social media integrated business. It’s a great example of how enterprises are changing and why we need to understand them in non-traditional ways. What can we say about her business? Well, it’s an Internet company but it’s not selling technology. It sells real people skills but everything lives in the blogosphere. You can buy her stuff via the site but it’s not an eShop. It’s Louisa’s company but in some ways it’s a virtual enterprise. What does that mean? Well, there will be multiple contributors generating and selling content and the quality and commercial success of the content will shape how the company develops. Or to put it another way, the contributors are not merely suppliers but actually investors, who benefit from the success of the company. Oh and it has its own website but the marketing happens via separate blog sites, via Twitter, Facebook, Google+, LinkedIn – you name it. It’s easy to see then how capturing the architecture of such an enterprise is about capturing the essence and not getting distracted by detail that can change at any moment – exactly due to the multiplicity of contributors and propositions. It’s a daring concept – jumping into the unknown – and of course we won’t see this model in the large enterprise world for quite some time but in the non-profit world or perhaps even in education one could imagine a more rapid adoption. In fact you might reasonably expect to see it adopted in education. It was after all educational and research organizations that gave us the Web in the first place. And back then the web was all about collaboration and sharing – co-creation.
Tom Graves has been looking at extending the Business Model Canvas into Enterprise Architecture as a whole. One part of this is extending it upwards (or outwards – depends how you look at it) to reflect the extended enterprise context in which most organizations “live” today. This involves taking concepts which we already apply to the single enterprise and applying them to a world we don’t control, where multiplicity is the rule and in which our objective is to be an equal partner. This gives rise to relationships, which are both complex and shifting. I would argue that one consequence is that we need to put the emphasis on capturing the entirety of the situation, so we can understand its dynamics and reach (breadth), and we need to avoid the distraction of those details, which we know can and will change without our being consulted (anyone see a similarity to Cloud here?). Another part of what Tom is doing is a mapping with Archimate. I don’t know whether Tom sees it exactly the way I do, but I think one of the advantages is that it combines the impressionist approach with a standardized modeling technique and allows us to provide detail where it’s meaningful and useful. And what it also does is provide a semi-formalized way of using techniques coming from a different discipline within (or along with) familiar EA frameworks. Well, I say “does” but I should say “will do”. It’s work in progress, just like Scatter. Just like TOGAF® Next. You can contribute to these things, influence them or adapt them to your own purposes. You can read and leave them aside but at least you’ll have thought about it. And that in and of itself will enrich your practice.
Stuart Boardman is a Senior Business Consultant with Getronics Consulting where he co-leads the Enterprise Architecture practice as well as the Cloud Computing solutions group. He is co-lead of The Open Group Cloud Computing Work Group’s Security for the Cloud and SOA project and a founding member of both The Open Group Cloud Computing Work Group and The Open Group SOA Work Group. Stuart is the author of publications by the Information Security Platform (PvIB) in The Netherlands and of his previous employer, CGI. He is a frequent speaker at conferences on the topics of Cloud, SOA, and Identity.
Great post, Stuart.