MCCUSKER
From an EA practitioner’s viewpoint, the gaggle of organisation and cross-organisational systems have more code, more interfaces, more configurable options, more “hidden features”, hence exponentially increasing the more things that can go wrong and increasing the amount of time to resolve the issues.
Australian academics no longer seem interested in resolving questions in complexity theory and optimising architectural patterns that would be useful to the EA. Is this your observation as well?
O’NEILL
There ARE pockets of researchers working on this exact NP-hard problems (those being the most difficult to solve). Just like architecture is all about the “whole being greater than the sum of the parts”, several hard problems combined (ie more code, more interfaces, more options) is much harder than just the sum of these problems themselves.
Some radical re-thinking is needed. Our founder, Professor John Leaney, recognised this at UTS many years ago and we have, through 15 years of patented research and expert consulting, made some world-leading progress in the area. Accordingly, technologies such as ABACUS (standing for the Architecture-Based Analysis of Complex Systems) are being commercialised through UTS spin-off’s such as my company Avolution as we speak.
Other institutions, like the Software Engineering Institute (SEI) in the US and the Friedrich-Schiller University in Europe are also making good progress. But yes, on the whole, academics have missed the boat (or more precisely, let it sail on the crest of the latest wave) on this one. Did I hear someone say business intelligence?
O’NEILL
How do you justify the business case for architecture in general and EA?
MCCUSKER
Often the business case for EA is driven around cost savings, simply because in most financial models proving cost savings is more tangible than proving the net benefits of service improvement or potential revenue growth. The high-level case for EA is that a coherent plan, in orders of magnitude, is much better than having no plan. The reasons are:
· it’s within our nature to want to know what we are doing, it makes us feel safer.
· the power of an aligned group in order of magnitude is greater than an unaligned group.
· getting a group to go “somewhere” is more beneficial than spending previous time and resources “going nowhere”.
· it is justified through the combination of the improvement to service provision, as measured by improved customer and service-provider satisfaction and improved profitability.
· it’s also justified by addressing issues that no other part of the organisation is likely to take on, especially reducing process and system complexity (you can’t reduce system complexity if you don’t know how it works), reducing key risks and the costs associated with operational outages. Many organisations can accurately calculate what an operational outage has cost.
MCCUSKER
Are business systems no longer considered a trendy area for research? Most IT students will end up working on business systems in their careers. Are academics letting their students down in this area?
O’NEILL
There are attempts to closely align under graduate and post-graduate courses with industry needs, and programs like UTS’ ITMP, Engineering and IT co-op programs are generally accepted as the best in the country, if not the world. But we can definitely do more.
On the research front it’s becoming more and more difficult to “do” research with industry because:
a) National ICT Australia has all the government leveraging money
b) one-year funding cycles in business but three-year commitments needed by universities (for PhDs), and,
c) academic speed versus corporate speed.
Write to the Editor at Technology and Business
* All fields are mandatory.