IT-centrism is killing enterprise-architecture
All right, I admit it: I allowed frustration to get the better of me in the previous post, ‘How not to define business-architecture‘.
But the real point is this: IT-centrism is killing enterprise-architecture. Gartner made that clear some months back (apologies, can’t find the link…) in their ‘Hype Curve’ on EA: the IT-centric view of EA is increasing the ‘business-IT divide’, not reducing it. And as IT necessarily becomes more and more interwoven into everyday business, that IT-centrism is triggering a serious pushback from everyone else. Sure, if we’re in ‘the trade’, it demands our attention, but we’ve indulged it now for far too long: the blunt fact is that the obsessive IT-centrism of too many – especially amongst the ‘big players’ – must stop, and stop now, or else there will be nothing left. It really is as serious as that.
First, though, I ought to acknowledge some of the Twitter-stream that came up in response to that last post:
- dougnewdick: Good points w/o the invective RT @tetradian: [post] How not to define business-architecture… http://bit.ly/p0mITZ #entarch #bizarch #togaf
- tetradian: @dougnewdick point taken about ‘invective’ 🙁 – but how on earth else do we stop Open Group from trashing the industry yet again…?
- dougnewdick: @tetradian I think that you weaken your case and alienate some of your potential audience with the invective. // I think you stop them by publishing a compelling counter-arg that respects all. I’m an enterprise IT arch & I’m listening 2 u
- tetradian: @dougnewdick yeah, I do know… – but after 5 solid years of repeatedly ‘publishing a compelling counter-argument’ it gets a bit wearing 🙁
- dougnewdick: @tetradian fair call! 😉 – but I’m starting here
- tetradian: @dougnewdick other point is that that arrogant IT-centrism is really annoying to non-IT folks – it’s causing a lot of damage to #entarch
- dougnewdick: @tetradian That’s a good point too. My POV is that non-arrogant bus-centric EITA is a good place to start and try to get a seat at the table
- tetradian: @dougnewdick strong agree: ‘non-arrogant bus-[oriented] EITA’ is essential – and ‘non-arrogant’ creates space for seat at table
There’s also a useful comment from Stuart Boardman back on the previous post.
What do I mean by ‘IT-centrism’? It’s the assumption (usually implicit) that IT is not only the centre of our own world and work (which is fair enough if we work in IT), but also necessarily the centre of everyone else’s as well (which is not ‘fair enough’). It’s the attitude that for every possible business problem, there is always an IT-solution; and that that solution will always be the ‘best’ solution, simply and solely because it’s IT. It’s the assumption that there are no limitations to IT, that it alone offers the golden dream of perfect control – and therefore has the ‘right’ to control all others. It’s the assumption that the world can be meaningfully divided into ‘IT’ and ‘the business’ – and that IT, by definition, is the only one that’s right. It’s the attitude that only IT-people know how the world works, and therefore we have the ‘right’ to tell everyone else how they should work, so as to best fit in with the way that we work. And it comes out in a myriad of other subtle, unacknowledged, amazingly destructive forms.
It is, in short, myopic, narcissistic, and arrogant: and it really annoys the heck out everyone else.
If you want to see how and where and why the ‘business-IT divide’ is created, and why it stubbornly persists as a wicked-problem despite all the best intentions of so many people on every side, all you need to do is watch how IT-centrism keeps coming back, and back, and back, in just about everything that comes out of the IT-consultancy industry and elsewhere.
I ought to emphasise here that this kind of ‘self-centrism’ is not specific solely to IT. For example, ‘business-centrism’ is beginning to be another real problem in enterprise-architecture, where ‘the business of the business’ demands to be treated as the sole centre of the architecture. Finance people do it; HR people do it; shareholders do it a lot; Health & Safety folks do it far too often too. Just about every domain does, probably, at some time or other. But it does seem to be particularly endemic in the IT-consulting industry; and since they still dominate the enterprise-architecture discipline at present, that’s where most of our current ‘-centrism’ problems arise, and why IT-centrism is such a serious problem for us right now.
A domain-architecture necessarily centres around a specific discipline: that’s why it’s a domain-architecture. A solution-architecture also usually focusses its attention on a single domain. But every domain and domain-architecture has to learn to ‘play nice’ with all of the other domains and their architectures – otherwise the whole will always break down into feuds and infighting, or fragment into fiercely-defended ‘us-against-the-world’ silos.
For architecture – the relationship between structure and purpose – there needs to be something, some form of discipline, that helps to hold everything together. And the only way that works is by accepting whilst everyone does need to be the centre of attention at some point, no-one can claim to be ‘the centre’ around which everything always revolves. It does not work – it really is as simple as that.
In a true enterprise-architecture, everywhere and nowhere is ‘the centre’ all at the same time.
Hence we cannot allow IT-centrism to exist, or any other ‘centrism’ to exist, because it kills the architecture, every time.
Hence we must challenge it, every time we see it – in others, and especially in ourselves.
If we want our enterprises to succeed, there is no other choice: that IT-centrism has to go.
Which is why I’ve ranted so often on this point over the past five years or so. I apologise if it’s annoyed people a bit too much at times: but hiding from facts never helped anyone for long, and as a profession we really need to face this fact now.
That’s it, really. End of rant? 😐
I couldn’t agree more. As a business analyst in a government organisation in the Netherlands, I have also been asked to come up with the information architecture. Naturally I started to look for the business architecture. Surely there is none. Therefore I started I started to get to know more about EA. Gradually I became more and more dissatisfied by the fact that EA is generally viewed to be about business architecture, information architecture, application architecture and IT architecture. 3 out of 4 parts of EA are therefore primarily focussed on IT. But what about the financial architecture or the workforce architecture? Even (in our case very much relevant) do I miss the physical architecture (being the way we are housed). Since we are a regional collection of firebrigades and related public safety organisaties, the way fire trucks are spread around the region in wich fire station is very much a concern. Currently this part of the organisation is very much an issue because of budget cuts. Taking this on with a more (enterprise) architectural approach (and therfore in conjunction with other changes) wil give a more future-proof solution.
To put my thoughts to paper I have coupled the various (what you might call domain architectures) in relation to the well known economic factors of production. On the highest level you then get 4 basic architectures:
(1) Production architecture (focussing on processes, etc.) in relation to (natural) land
(2) financial architecture (focussing on financial structures, earning models, etc.) in relation to capital
(3) organisational architecture (focussing on people and the way they are organised) in relation to labor & entepreneurship
(4) informational architecture in relation to the fifth factor of production information
Bart den Dulk
This is further complicated by the job requirements. I’ve seen more EA job opportunities specifically mention software environment requirements and no business or organization understanding … maybe a major business “application” only!
nice post. I think it clearly sums up your key point really well: that IT-centrism is the real issue here. As an Enterprise IT Architect, I know that the dangers you warn of are very real. I have to constantly check my own behaviour for it, and I see it every day in the IT people I have to interact with. Also as you mention, any part of the business can fall prey to some kind of narcissism – thinking they are the centre of the universe and the most important part. In this respect I like to think of enterprise architects as maintaining balance and coherence across the competing demands of the enterprise.