In the lead editorial of the February issue of EngineerIT, Dr. Mike Rodd states “it is painfully evident that an unacceptable number of large scale so-called IT projects are continuing to fail … “<\/p>\n
So what is the problem?’ Of course he is correct – it is amazing how we all just accept that IT projects fail, or do not deliver what was required or are cancelled – it just seems to be the cost of doing business! Will changing IT into a proper professional discipline – as is the purpose of the IFIP and BCS and other projects – change the situation? The answer is not as straightforward as it may seem!<\/p>\n
Dr. Mike Rodd compares the IT discipline to other professional disciplines, and asks “what would be the consequences if a third of new bridges fall down, a third of new power stations do not work or a third of new aircraft simply could not fly?”<\/p>\n
Why then do so many (a third?) of IT systems fail?<\/p>\n
At least one answer to this is that “Grade 12 learners do not build bridges, power stations or aircraft, but Grade 12 learners create software systems!” Having created such (small) software systems which basically work, they grow in confidence and create larger and larger software systems, without ever being exposed or taking note of best practices, efficiency, comprehensive testing etc. Many of these learners start companies, and accept contracts for larger and larger software systems, with, often, the results as mentioned above. Nobody would allow a Grade 12 learner who performs some basic operation on his\/her pet, to just carry on and later perform an operation on a human – we simply do not allow it!<\/p>\n
It seems that inherently in the problem of software systems failing, is the fact that developing software had been made so “easy”, that anyone can do it, and soon sees himself as an expert (professional?) and therefore capable of developing bigger and bigger software systems.<\/p>\n
The other professions we refer to, do not allow this, and are therefore “true” professions. They are supported by a vast legal infrastructure which will not allow an organisation to accept a tender from a company to build a bridge if professionally qualified, and registered, engineers are not involved. However, that same organisation will award a contract to a software company consisting of “self-made” programmers to develop a multi-million software system which may have a catastrophic effect on the economy of a country if it fails, or which may lose millions of rands of tax payers’ (or investors’) money if it is cancelled!<\/p>\n
So can the situation be changed? Of course it can, if the will and commitment exist.<\/p>\n
Firstly, the IT discipline needs a similar legal infrastructure as those regulating other professional disciplines, to regulate software development – any voluntary system has little chance of succeeding.<\/p>\n
Secondly, we need tax payer and investor involvement to demand accountability if money is lost because of software systems failing and costing the stakeholders money. This is part of good corporate governance, and promising developments in this area are becoming apparent.<\/p>\n
Thirdly, we must realise that developing software systems is one of the most complex actions ever to be created by man, and that we must ensure that is it done by qualified people.<\/p>\n
Of course, the reasoning above does not imply that the Grade 12 learner, or any amateur, should be prohibited from developing software – just as an amateur is not prohibited from building a wall in his garden. He is however (legally) prohibited from building a bridge used by the public.<\/p>\n
Also, it does not mean that an academic foundation is necessarily needed – many brilliant engineers build bridges based on experience over many years – in the same way, many brilliant software engineers build very good software systems based on their experience over many years.<\/p>\n
Can the IT discipline become a truly professional discipline? My answer is surely “yes”, but some legal and regulatory infrastructure is essential!<\/p>\n