The Good Old Ways ?
Submitted by Phil Kirkham on Thu, 26/01/2006 - 16:46.
general software testing
Here's the EXE now bash away at it and see if you can make it crash.
If you can't then ship it.
So much easier than having to think about testing in all it's flavours and methods.
But thinking about testing is starting to happen here. The developers are investigating VS Team Edition and as well as looking at the new cool coding features are also looking at how it runs unit tests. One of the developers has bought 2 books about NUnit. Quite the change, developers talking about testing and not in the usual "Why didn't QA test this ? " when a crash is found on a customer site.
Should I reign in their enthusiasm slightly and tell them that unit tests still mean it is possible to build a concrete life preserver ? Or pass on the article from Software Test and Perfomance that has as the first rule of thumb "if unit testing tools are your only strategy for quality, you have no strategy " ?
But I don't think we'll be using concrete in our life preservers - the lead developer has taken away my copy of "User Stories Applied" - just leaving books lying around on my desk is paying off in spreading the Test Gospel.
Now when should I ask them who (and when) will be doing the integration tests, the stress testing, usability testing, security testing...
Or should we just go back to hammering an EXE ?
Main lessons learnt this week ? BELIEF. You really do have to have a
belief that applying new methods will pay off. Cynics have to be overcome, the apathetic have to be enthused, the traditionalists convinced that the brave new world is worth the journey.
If you can't then ship it.
So much easier than having to think about testing in all it's flavours and methods.
But thinking about testing is starting to happen here. The developers are investigating VS Team Edition and as well as looking at the new cool coding features are also looking at how it runs unit tests. One of the developers has bought 2 books about NUnit. Quite the change, developers talking about testing and not in the usual "Why didn't QA test this ? " when a crash is found on a customer site.
Should I reign in their enthusiasm slightly and tell them that unit tests still mean it is possible to build a concrete life preserver ? Or pass on the article from Software Test and Perfomance that has as the first rule of thumb "if unit testing tools are your only strategy for quality, you have no strategy " ?
But I don't think we'll be using concrete in our life preservers - the lead developer has taken away my copy of "User Stories Applied" - just leaving books lying around on my desk is paying off in spreading the Test Gospel.
Now when should I ask them who (and when) will be doing the integration tests, the stress testing, usability testing, security testing...
Or should we just go back to hammering an EXE ?
Main lessons learnt this week ? BELIEF. You really do have to have a
belief that applying new methods will pay off. Cynics have to be overcome, the apathetic have to be enthused, the traditionalists convinced that the brave new world is worth the journey.