Brian BurkeChris Wilson recently alerted me to an opinion piece on the British Computer Society (BCS) website written by Peter Kemp and Dr John McManus that is critical of enterprise architecture in general and the National Program for IT (NPfIT) of the UK’s National Health Service (NHS) in particular. Some members of the BCS have been quite critical of the NPfIT in the past and the program has also been often criticized in the press as well. While NHS bashing a favoured pastime in the UK there are some very good aspects to the NPfIT that I pointed out in a case study research note last year.
Mr. Kemp and Dr. McManus state “But, we have to say, that were not sure we’ve yet seen an EA strategy that is anything other than impractical, unachievable and, even if it could be achieved, unsustainable.” At the core of the argument is that EA is too high level and future focused for practical application. The problems they cite with EA are common but not universal. I would argue that the majority of EA programs deliver valuable principles, standards and high level models of the future state providing valuable direction to projects. Mr. Kemp and Dr. McManus clearly have a project level world view and from that viewpoint EA may seem onerous. It’s easy to fall back on the belief that projects should be left to their own devices – and the optimization point should be the project. I believe the objective of EA is to shift the optimization point from the project to the enterprise. EA is about the big picture. It’s about the future and it does not (nor does it attempt to) solve all project level problems. Are there problems with some EA programs? Of course, but that doesn’t mean that EA should be abandoned and we should retreat to the failed practices of the past.
The NPfIT is a particular target of Mr. Kemp and Dr. McManus. While the program is costly and has had its share of problems it is also moving forward to solve real health care information sharing issues. In fact many other countries are starting similar programs. On February 24, 2009 President Obama addressed a joint session of Congress and stated, “Our recovery plan will invest in electronic health records and new technology that will reduce errors, bring down costs, ensure privacy, and save lives”. The NHS program is actually the envy of many governments – too bad it doesn’t enjoy the same support at home.
I do agree with their conclusion that enterprise architects should be able to justify their activities in terms of overall business benefits. When I first spoke to Paul Jones (NHS Chief Technology Officer) he told me that this was the first position where he’s had the opportunity to save lives. I’m jealous. I’ve never worked on a project that had that kind of a business benefit.
3 responses so far ↓
1 Chris Wilson March 20, 2009 at 5:32 am
Thanks Brian – good response! I think what I’ve seen here is a reaction to the misuse of the EA title… there are a lot of people using the EA title, when they are actually Technical Architects with an Enterprise scope. They then use the EA models to enforce their view of a technical solution, which alienates them from their customers, as it’s still primarily a bottom up approach. If, as a customer, this is all you have seen of EA, with the associated mandates, standards, and “police” enforcement, it is unsurprising that you feel that EA doesn’t not deliver a business benefit!
2 Dr John McManus April 4, 2009 at 7:18 am
Dear all,
I always welcome constructive feedback and whilst I may not agree with all that is said – I do believe that papers such as this help to stimulated debate amongst professional practitioners.
I will in due course be following-up this paper.
Dr John McManus
3 Peter Kemp and Dr John McManus April 7, 2009 at 8:43 am
Further to Dr McManus earlier comment
We would stand by the concrete examples from the NHS and other government projects and other examples. E.g. the “Enterprise Application Platform” is still a common concept. In a very recent discussion, an advocate of the EAP confided that the one problem was “making the applications run on it”. So, maybe you should chose the hardware based on the applications, not the other way round? That is if you want the best choice of useful business systems.
Another recurring theme is the strategy to have all UI through a common “portal”, oblivious to a) the fact that a COTS product vendor may have designed a very good UI suited to a particular business task, and b) the question of who is going to pay for the time and effort to rewrite the UI in the portal technology (even if the off-the-shelf product supports such a thing).
So, this is definitely still relevant.
We have no opinion on the methods or effectiveness of business strategy. If an EA is working on the business strategy side of EA, then we have no opinion on the matter. But, lots of EA’s are working at the level where the Enterprise Architecture determines what can be implemented and how and forces projects to chase absurd, abstract goals.
The test is:
Does the EA or IT strategy make it simpler or cheaper to deliver useful business functionality to the end user? And, not in theory, but in practice!
Because, at the end of the day, that is all IT systems do. If they do not do that, then they have no reason to exist.
Peter Kemp and Dr John McManus