OK, I freely admit it isn’t the most original of titles being a paraphrase of the title of John Seddon’s book. Now I have to admit that I have this book on my bookshelf, and I occasionally pick it up and skim, but there is something about it I can’t yet appreciate and so I remain ambivalent to it.
No, the reason for the title of this post is because at the moment I’m very much involved with looking at the future in two different, but interconnected, areas.
Firstly I am doing some work on the skills we will need in the future for delivering IT in partnership with a gentleman called Rajeev Bhardwaj, by day an Enterprise Architect but also a cartoonist who keeps his artwork hidden away on his blog.
Secondly I am looking at scenarios for business models in ICT as we increasingly use cloud services, and new ways of deriving information.
What is becoming very clear is that where responsibility lies is becoming an emergent challenge of these wider changes, but behind that simple statement lies a lot of complexity that can’t be ignored
It’s [Going To Be] Complicated
I posted a diatribe about fads in IT almost a year ago, but I have to hold my hands up here and say that Cloud Computing is finally moving from fad to being realistic. Through software vendors offering hosted solutions; our organisations are now being presented with credible business solutions delivered as demanded by customers, and embodying new financing models.
It’s early days, but these solutions are maturing and reversing the trend of “everything-in-one” solutions in favour of the customer having a menu to “Pick and mix” from with the promise of hard won money being deployed only to what is actually serving a purpose for that particular customer.
My problem here is that coming from a technology background I foresee challenges (that’s a euphemism for “problems”) including:
- As a consumer of business services selecting from this cornucopia of offerings, how do I know that a particular combination will actually work to serve my business in the best possible way?
- From a systems engineering perspective, how do I know that the combination will work efficiently together and give me effective information with which to manage my business (and herein lies a problem that will only get worse)
- As a vendor, how can I maximise my own revenue streams when I have customers using my offerings with other components that I have no control over, yet demanding wider business benefits that transcend my own offerings.
The wise man may well say something to the effect that this presents an opportunity for the business of aggregating, offering selections of services that can safely be used together with known results. This is going to be the new economy: service aggregation, and I look to Amazon as the emerging model here.
But more is not necessarily merrier, and complexity theory jumps in here suggesting that the more combinations you try to make available means that the management burden increases by the square law.
That’s great if the economic models supports it, but the cloud model is about paying only for what you use. How many business owners will accept that they need to “use” skills to integrate offerings in order to get information about how their business is doing, yet they are going to come under the same cost pressures. Those with the best understanding of how their business is running will do best.
|Let’s summarise here:||Our businesses are going to come under ever increasing cost & margin pressure, and those with end-to-end information about their business performance will survive and hopefully even win out.
Yet … getting to that position to control a business may well mean incurring more cost especially when you fragment the business systems you rely on to transact with customers
Who takes responsibility?
After that bit of background we should get back to command and control.
As we head towards these new complexities, the law of requisite variety suggests that we won’t be able to cope. Not because we can’t, but because we won’t be able to afford to.
We need a new way to manage this, as we cannot build a command and control system of sufficient variety to manage that in the underlying system i.e. all that complexity caused by trying to offer a smorgasbord of money making services that can work together.
One way is to create standards, but does that really work? The cloud market is characterised by small, innovative entrants working quickly to deliver new ideas. Standards slow that down, and tend to become the domain of the largest players.
I’d suggest here that the stage is set for moving responsibility from the top to the bottom, and that applies to customer/vendor chains, and to our organisations in general. We can’t rule for the total extent of the complexity, and deliberately leaving gaps in our rules creates work for lawyers. Instead, I suggest, we need to move to a world where the consumer of these business services has to work on an assumptive basis, and makes those assumptions clear so that the higher orders can work with those making the assumptions to correct any (note I say “work with”, not enforce). And this means communicating so that intentions are clear, and making sure that the emerging body of knowledge is promulgated for future use.
Many years ago I was drilled on what assumptions do, but I’d argue now that if we can’t control the whole we can only derive the controls as we need them. In the same way we can’t afford the time to formally create controls, and so we need a new way to make sure we understand what is going on, and our expected outcomes.
The reality is that it isn’t just “assume”, but what we need will be based on “assume …. and communicate that assumption”. Intuitively this feels like something that will meet a lot of cultural resistance, but my forecast here is that this approach will evolve naturally.
After all, is there an alternative?