Software Testers as IT Rock Stars? Party-on! Excellent!

Gavin Clarke at theRegister reports that:

Microsoft hopes to do for software testers what it's done for developers - turn them into "rock stars"

With Microsoft's recent launch of Visual Studio Team System 2005 (VSTS2005) which includes an array of integrated Testing Tools, it is clear that Microsoft's stance isn't just for the greater good of software development teams. Clearly, it is also intended to boost the market for Team System.

Whatever the motive, I see it as a good thing. Too many development teams underestimate the value that a talented and skilled tester can bring to a project. Having a large brand, like Microsoft, promoting and advocating better testing can only support improvements in the quality of software and perhaps contribute to fewer project failures in the industry.

It may spell doom for those testers who don't have the skills and aptitude to become 'Rock-Star' Testers. They will soon be replaced with people that do have what it takes but, perhaps in the past, had avoided Testing as a career due to a perception that it is undervalued. My experience is that the value placed on testers varies wildly across business sectors and, in some sectors, is in need of a boost.

Gavin goes on to link Microsoft's public test-advocacy to the development community's perception that Microsoft's latest Visual Studio offering has too many bugs. In relation to this, Gavin writes...

[Microsoft] may have to get its own house in order before preaching to the industry.

Although it is true to say, that lack of testing is likely to result in buggy software this doesn't mean that buggy software releases are typically caused by lack of testing. After over a decade in Software Testing I can say with confidence that the main factors in the release of 'buggy' software are the commercial and/or political drivers and not the quality or status of the testers.

Rarely, if ever is it the testers' decision whether or not software is released, and rightly so. It is common for software to be released despite a long list of outstanding bug-reports and many more areas that are either untested or with many more tests that could/should be executed.

Product Managers (or Project Sponsors in the case of bespoke business information systems) have the tough choice of balancing the commercial impact of a delay with the commercial risks of releasing buggy software. Ideally, the Product Manager will make that decision when the balance is right. Sometimes, that balance is in favour of the users getting high-quality software. Other times, commercially speaking, being first is more important than being the best. Sadly, there are also cases where internal politics and date-driven bonuses that don't place an adequately balanced weighting on quality are the main drivers to release on a given date.

For the users of the software, this state of affairs is an unfortunate reality! But is mostly a commercially-driven reality and doesn't preclude publicly advocating Software Testers.

The question is, does the fact that vocal .NET supporters have slated the quality of the first release of Visual Studio 2005 mean that Microsoft hasn't taken its own testers or testing practices seriously? No, not necessarily and in my experience unlikely. It is more likely that a commercial decision (right or wrong) was taken to release sooner rather than later. That decision was not likely to have been made by a tester but instead by someone who could evaluate the risks against the potential commercial rewards.

In this case, I suspect that Microsoft know that many potential VSTS2005 customers are already so heavily invested in .NET and Visual Studio, that it is only a matter of time before they move from VS2003 to VS2005. Moving to anything else would just be too costly for most. With that in mind, it is also only a matter of time before VSTS2005 comes up to par as future patches fix the outstanding (important) bugs.

In the mean time, there will always be early adopters that will find ways to work around the bugs. As a result of releasing now, Microsoft will have attracted some revenue rather than none and without that choice necessarily being at the expense of future sales.

Convincing companies to adopt Team System may prove difficult in the first release, but ultimately, the quality of the first-release isn't going to be the deciding factor for most Team System customers.

I hope that Microsoft and other big-brand software houses aren't deterred from continuing to raise the profile of software testers and software testing practices, despite their commercial choices being misinterpreted as policies that are contradictory to their practices.