We are sometimes told that, “quality is in the eye of the beholder” and if this is the case, then many attributes of a product might please that eye. Statements like, “the product is safe” imply security testing and backup and recovery testing, while “the product is aesthetically pleasing and easy to use ” imply usability testing “. All of these quality aspects and more are validated by software testing.
Another definition of quality is that, “quality meets the actual user requirements or needs”. In this case, it is testing that validates that actual user needs have been met or verifies that each step in a delivery process follows through from, and can be logically related to supporting the user needs.
There are more software testing suppliers in SA now than 2 years ago, with many being small start-ups. Some are credentialed by their BEE scores rather than by their ability to deliver, others are able to deliver on niche market opportunities to give real value.
Others are large foreign suppliers who enter the local market, thus upping the competitive stakes, more so since developed economies have been suppressed by the recession.
There are a few local, focused software testing suppliers who have helped build and develop the software testing market in SA and thus have intimate knowledge and experience of what works and what is likely to fail. It is important that local suppliers also adapt to the changing market conditions and client needs, in order to remain relevant and competitive.
Many SMME’s don’t have deep software testing skills, as would be represented by persons holding advanced ISEB / ISTQB qualifications with over 5 years of testing experience and a successful track record applying these qualifications.
Although larger corporates have easier access to these individuals, testing professionals are not always favoured to oversee their testing projects against entrenched managers who don’t have deep testing experience. This can lead to sub-optimal quality or testing decisionswhich often have long term cause-and-effect consequences and are not always intuitive. Market forces are bound to correct this trend over time, particularly as more corporates start to measure results in terms of quality, quantity, schedule and cost, rather than in rands per hour per person (a crude and outdated measure). I strongly support any effort to measure actual testing productivity, including testing efficiency and effectiveness.
Companies number crunch and steer policies towards correct BEE ratios, but this has some negative aspects to growing local testing skills such as biasing the selection towards ethnicity and thus losing persons with deep software testing experience and skill to overseas markets.
Much testing work has been given to persons outside SA, thus building testing and quality skills in India, for example, at the expense of developing our own testing professionals. Large corporates have been able to get cheaper skills from India and have had little incentive to develop new SA. Ironically, the ‘lower’ foreign costs are often at rates which the local market could find attractive and are sometimes even higher than local market rates when total costs and benefits are accounted.
Weakening output from educational institutions combined with increased competition has led to a shortage of suitable IT graduates, with many companies opting to employ experienced contracted resources rather than develop new software testers.
Selecting software testing staff has become very important, as many software testing CVs in the industry are ‘somewhat creative’ when reflecting actual experience and ability. Even qualifications cannot be trusted without verification.
Some unscrupulous suppliers are exploiting the current market by fielding under-qualified or very inexperienced candidates at higher rates than warranted. It is good practice to ask the testing supplier how long they have been in the industry, as well as obtaining verifiable industry references.
The NQF has high ideals for training advancement but no way of enforcing training choices in software testing where the bar is mostly set by international qualifications such as the International Standards Examination Board (ISEB) and International Software Testing Qualifications Board (ISTQB).
Software success in terms of cost, quality and schedule has generally been poor world-wide in the past and is at a 5-year low, according to the Chaos Report (ref. Chaos Report – Standish Group, 2009).
This low success rate can be somewhat explained by confusion in the market, as Information Age management approaches collide and compete with Industrial Age management culture. This is to some extent mirrored by traditional software development life cycles being challenged by Agile software development life cycles and practices.
Only those software testing suppliers or individuals with deep roots in both approaches are likely to be able to chart the turbulent waters of these widely-divergent approaches as they compete in the current market conditions. Simplistic approaches claiming that either the ‘V-model’ or the ‘Agile approach’ is right, are doomed to poor quality as they oversimplify and miss the possibility of a balanced approach which selects the best elements from both.
What questions should purchasing clients ask in order to benefit optimally from testing suppliers?
It seems that perhaps there are more questions than answers and so let’s boil this down to a single question: “Are we applying our minds effectively to the broad national and narrow company interests of software testing?”.
This approach is relevant because we are preparing the local software testing and quality industry for future generations. What we achieve or not will impact SA’s future competitiveness and productivity.
So whether you manage testing and quality suppliers or work within them – don’t leave your brain in the parking lot when you arrive for work.
Wayne Mallinson
info@testfocus.co.za
The reproduction, adaptation or broadcast, without permission, of any articles
or photographs published in this
publication, is forbidden and copyright
is expressly reserved for Test and Data Services (Pty) Ltd.
Opinions expressed by contributors are their own and do not necessarily express the opinions of the Publisher, Editor or members of the Editorial Panel.