Failure of the US Government's Enterprise Architecture Programposted by John Spacey, February 04, 2011
There are three kinds of Enterprise Architect:
1. those who learn from the failure of others
2. those who learn from their own failures
3. those who never learn
If your Enterprise Architect can't describe five cases where Enterprise Architecture has failed miserably, it is time to find a new Enterprise Architect.
Enterprise Architecture (EA) is a high risk activity that it is prone to failure. This is not to say that it has no value: EA can pay huge dividends to the few organizations that are able to get it right.
As a cautionary tale consider the failure of the Federal Enterprise Architecture program of the US government.
BackgroundIn 1999, the US Federal Government's CIO Council published the Federal Enterprise Architecture Framework (FEA).
In 2003, the US General Accounting Office (GAO) released a scathing report — pointing to the failure of FEA and a general lack of EA success across the the federal government.
Enterprise Architecture?GAO suggested that FEA completely missed the mark.
Should the FEA be described as an enterprise architecture? .... more akin to a classification scheme for government operations than a true enterprise architectureWhen you spend loads of taxpayer money on EA and what you develop is not EA — you have some explaining to do.
Security what?GAO pointed out that FEA does not cover security.
it plans to address security through what itAt this point FEA had been more than 5 years in the making — yet officials could not even comment on the status of a promised security framework.
terms a “security profile” to be added to the FEA. However, OMB
officials could not comment on the profile’s status or development
Two steps forward three steps backGAO also criticized the general lack of progress in enterprise architecture initiatives across the entire federal government.
while some agencies have made progress in
improving their enterprise architecture management maturity,
progress for the federal government as a whole has not occurred
22 agencies increased in maturity since 2001, 24 agencies decreased and 47 agencies remained the same
Make that 22 steps forward, 24 steps back and 47 sleeping federal agencies.
Why does any of this matter?If you are not a US taxpayer footing the bill for federal EA programs you might be tempted to ask — why should I care about any of this?
To answer this we defer to the philosopher George Santayana:
Those who cannot learn from history are doomed to repeat itMost enterprise architects are quick to explain the theoretical benefits of EA — but few have a realistic view of the risks involved and probability of failure. It is essential for the future of EA that failure risks be openly discussed.
Imagine your hardcore IT geek talking to a company executive. What would they talk about? |
Understand the threats to your organization.|
Explaining complex business and technical concepts in layman's terms.|
CxO level executives are demanding metrics for EA — can you deliver? |