Viewpoint: Fear of feedback?
Why are suppliers of software development usually so reluctant to bring in others to provide a second opinion?
Writing software is a relatively new discipline and, as a result, it is something that we are still learning how to do well. The state of the art circa 2009 is not good; we only have to read reports of the latest collapsed, delayed or over-budget public sector project, or experience one of the many challenged commercial projects, to see how limited the industry is.
There are successful projects, of course, and these often use a particular kind of technique at the micro level: techniques based on independent review. For example, defect rates in code are known to be significantly reduced by code inspection.
Having someone other than the author read through code, however, is a seriously challenging - but hugely beneficial - practice. Teams can be reluctant to adopt these kinds of measures at the macro level, for a number of reasons.
Some developers just do not want to expose their work to review. Rather than seeing feedback, especially early notice of defects, as a welcome opportunity to increase quality, they view it instead as a threat.
Some developers, and some teams, resent the distraction from the supposed 'real work' of writing more code. Some development managers are reluctant to devote resources to strong quality policies, sacrificing long-term productivity for short-term throughput. Similar syndromes can be seen at the scale of entire projects and programmes.
It is rare for a purchaser to institute a review of an on-going project by another supplier other than to confirm their suspicions of, or diagnose an already recognised project failure. The expected mode of a second supplier is problem solving or trouble shooting, or, in the worst case, project rescue.
It is vanishingly rare for a supplier to voluntarily bring a potential competitor into a project. Given the attitudes of purchasers it is easy to imagine that to do so would be seen as an admission of failure. And yet is not the lesson from the longer-established disciplines that it is to the benefit of both purchaser and supplier to mitigate risk through independent review?
On a project where the supplier had voluntarily brought in another, not as an admission that they are incompetent but out of a desire to do the best job, might we not expect better internal quality and customer satisfaction?
We might notice sooner that the wrong question had been asked (too often the case in those public sector projects), or that the wrong answer was being pursued with vigour (a private sector failure mode), or even avoid starting to have those problems in the first place.
This seems as if it would be a good thing, but it requires a very deep change in attitudes in the IT industry.
Keith Braithwaite, Zuhlke Engineering.