Wednesday, October 01, 2008

Pink Floyd meets Frank Sinatra - anti-pattern

This anti-pattern is all about training and learning. Its the "we don't need no education" anti-pattern. It occurs when organisations want to "do everything themselves" and develop in isolation from other efforts and experiences. There can be some reading done but the predominant theme is that it will be done "My Way".

The effect of this is that the organisation starts creating its own definitions of terms. These are created in a "folksonomy" manner from local opinions rather than taking reference to external sources. Common elements are definitions of "services" that map to things that already exist (see Defensive SOA) rather than looking at a new approach. The other side is that when using technology the company will "learn themselves" how to do the implementation with very little involvement from either the vendor or experienced 3rd parties, the reason for this will be that the company wishes to "learn themselves" how to do SOA, the reality is that they are concerned that external inputs will demonstrate that their current approaches don't work.

The effect of this is that SOA becomes simply another label attached to the companies standard way of working and any changes are purely superficial. SOA rapidly becomes a "best practice we've done for years" and new technology projects fail to exploit the new technology successfully so little, or negative, improvement is seen in productivity.


The causes of this is an inward facing IT department which sees external companies, including vendors, as competitors rather than collaborators. Sometimes this is combined with a huge arrogance that the company knows best, but mostly its a mistaken belief that in order to remain in control you must do everything yourself. A large part of the issue comes from a fear of new ideas being demonstrated as measurably better than what went before and when tied up with a blame culture this results in a need to protect reputations over open collaboration and improvement.

The first part of the resolution is to assess whether you really are that good. If you turn around and find that you are the favoured love child of Google and Amazon and Bill Joy, James Gosling and Donald Knuth are fit only to be junior project managers then carry on, you are right and you are the best. If however you only think that Bill Joy is rather smart, or even employ Bill Joy then remember his mantra at Sun "Most smart people work elsewhere". The second stage is to get rid of the Blame element. You've done an okay job and you want to do a better job. That needs to be seen as a good thing not a bad thing, it needs to be seen as the right behaviour. The next stage is a budgetary one, how will you measure the benefit of external spend and how will you justify it. You need to have clear cases on what you expect external spend to bring and how you will measure its impact. This way you keep in control and use external people to help you improve where you are either weak or where you don't want to waste your time.

Technorati Tags: ,


Daniel Spangler said...

It seems that this mentality could also be something that grows out of an underfunded IT department where doing it yourself is rewarded because of the perceived cost reduction in the short term.

Steve Jones said...

Good point Daniel and one worth noting.