.

8 June 2008

Psychology of why reuse fails - courtesy of GeekAndPoke



This amusing cartoon by Geek and Poke captures one of the important reasons why so much of our planned software reuse fails. Thus projects often fail to achieve the economy of scale benefits from the business case.


Within companies there is often little trust between departments and an unwillingness to rely upon things that other people have built.


The real question is how can we overcome this problem so as to achieve the benefits of shared resources and reuse?


The first step in resolving the problem for any organisation is recognition that this kind of problem exists. The next step is to bring the issue out into the open.

Often there are war stories within the business that explain how or why the lack of trust has evolved. It is important to uncover these reasons. Once things are out in the open they can be managed.

My rule of thumb for problems with technology projects is that about 70% of the problems are in our heads (us being all the people involved in the project), while the other 30% of problems are logistical (e.g. stuff takes longer to do than planned; or stuff gets delivered late).

Very rarely are the problems actually about the technology in and of itself. More often the problems that appear to be technical result from hidden issues with people that remain un-addressed, or architectural problems that lead to the selection of inappropriate hardware or software.

By Carruthers via Aide-mémoire

2 comments:

Saul Cunningham said...

Kate

I think you are quite correct in your assessment - technology is not the problem here - its human behavior. This cannot be solved except through rewarding reuse.

It is true that SOA solutions have the potential for reuse but ultimately if the organisation doesn't reward it, it won't happen because when push comes to shove it is put in the 'too hard basket'.

I know of one customer who basically axed projects that hadn't delivered after 90 days! The only way to meet that was to reuse existing things and to modularise projects into smaller pieces (think Agile methodology).

Carrot and stick approaches to rewarding reuse/punishing non-reuse need to be considered. No technology can solve the "you can lead the horse to water but you can't make it drink" problem!

Saul Cunningham
Oracle Corporation

Steve said...

Interesting post Kate. After working in the industry for over 15 years delivering technical 'solutions' I've yet to see, in use, anything similar to the holistic approaches put forward by (e.g.) Peter Checkland, in seeking to understand and bring out into the open all the agendas of the stakeholders - some can be a little too guarded to make a large project the success it could be.

The result is that we fall back on our methods and our experience but sometimes even that isn't enough.

Steve Teale
Oracle Corporation