It’s been a busy August so far, putting the finishing touches to a report on Enterprise Architecture (EA) which has just been published by JISC TechWatch.
EA is a strategic management technique which aims to align business strategies and goals with information systems. The process involves mapping out both the current situation within an organisation, what’s termed the ‘as is’ and then laying out a vision for the future, the ‘to be’.
It has been in use in the commercial world for a decade or so, although it is new to the education sector. The report synthesises the results of a year-long pilot project by a group of pioneers who looked into the day-to-day practicalities of introducing this technique into the higher education institutions. In particular they looked at the use of The Open Group’s TOGAF method for this kind of work.
The report comes to a number of conclusions, but I think the most interesting relates to the potential for the technique to be used to help the sector move to a low carbon future. As the report makes clear low carbon ICT is an area of activity that is strategically conducive to the EA approach as it needs long-term planning within, and possibly between, institutions. Work is already underway in the sector on the feasibility of shared data centres and the introduction of EA can only help these initiatives.
The report’s called Unleashing Enterprise Architecture and you can have a look at a PDF of the report on the JISC TechWatch website.
Tags: EA, Enterprise Architecture, JISC, Low Carbon, SOA
Tech Lunch 
August 10, 2009 at 5:26 pm |
I enjoyed reading your report. I had a few thoughts:
* I suppose that after all the good architecture work, when it comes to actual SOA implementations there is still a high risk of vendor lock-in. A lot of vendors already offer rapidly maturing SOAs based on software products which (by the time some HEIs come to implement) will look attractive and proven in the “real world”. Sometimes the very features that are most attractive involve acceptance of proprietary extensions or lock-in to vendor development tooling.
* Reading about the IB / e-Framework work on creating content or usable modeling I was struck by the national rather than international perspective. This contrasts with the global nature of the TOGAF work. I wondered if the UK HEIs are really that different from their continental or US cousins ? Did the Dutch universities produce anything useful in that area ? If HE is a new ‘vertical’ then I guess you’d expect that concrete work about models and exchangeable data would be best done at a global level, like it is with Medicine or in Finance.
* I think I’d put more emphasis on regulatory issues as a driver of organizational change. Nothing like external auditing of compliance to put the fear of God into people.
* I think the differences between HEIs and the private sector / commercial organizations can be overstated. There are plenty of commercial organizations with complex internal structures, discrete groups with high IT skill levels and inventiveness to spare, lots of devolved IT and no consensus between competing internal groups about IT priorities or futures.
* This is outside the scope of the report of course, but my feeling is that for most organizations there will always be a tangible and painful gap between the architected and CIO owned ‘official’ applications and the organizational need, particularly for these small talented teams which you alluded to, of which there are surely many in HE. If I was an HE CIO I would set aside some proportion of my budget for these Situational Applications or, failing that, provide an environment in which departmentally funded SAEs can be deployed as long as they reach some relatively low threshold in terms of (say) security compliance.
* A final thought: I think it’s probably worth saying that some basic boot-strapping level of IT provision and connectivity has to exist to allow the “excellent communications between groups of staff in different parts of the organization” which you mention. I’m thinking particularly here of IM and CMS / wikis so the gap between the private “at home” and the organizational “at work” experience is not too wrenching. I wonder how many HEIs would fail the test which says “Can your Finance person A, Admissions person B & IT staffer C collaborate using IM and a CMS on this new EA project today ?”
Cheers, Dave.
(Full disclosure: I work for IBM, but not in any areas to do with HE or these technologies & these thoughts are purely my own opinions.)
August 13, 2009 at 8:00 am |
Excellent feedback Dave. Many thanks. Some of the points you raise are very pertinent and I believe will be explored by HEIs in a new project which started on 1st July. In particular, I understand there are plans to explore the international dimension and work more closely with the Dutch.
See this link for more details: http://www.jisc.org.uk/whatwedo/programmes/flexibleservicedelivery.aspx
August 14, 2009 at 2:06 pm |
Very interesting report, Paul, and a very nice piece of work.
I congratulate JISC (and you) with this structured approach.
A few comments:
With regard to the Pilots, I observe a remarkable difference in acceptance of EA (or ease of introduction/adoption) between LJMU and Cardiff. Maybe this is caused by the fact that EA is a top down activity (in principal), therefore adoption of an EA practice is easier for an organisation with a more centralized steering concept.
I am beginning to believe that the degree in which the workers in an organization are willing to support a common goal is a good predictor for the degree of success of an EA introduction.
In most universities I know this willingness to support or even acknowledge a common goal is not very strong. in fact this seems to be the result of a historically grown scientific attitude, which comes very natural to the researcher and is in no way detrimental to the quality of the research.
My conclusion: yes, HE is a special case.
I take the liberty of supporting the recommendation of the consultant about the use of tools. It is true that views can be constructed using Visio (or OmniGraffle). Maybe it is even better to use these tools when presenting a view to the CEO or the Board. However the true value of the tools lies in the repository, where you can keep, manage and update the (ArchiMate) model of your organization. so my advice would be to use the tools from the start and when relevant translate to Visio etc.
Bert