CompuServe Messages

Steve I cannot agree more with your comments about beta-testing … I've been doing it almost since the day I got my first PET (serial number 13 🙂 There's an ego-trap that developers can so easily fall into … they'll only take criticism when the comments agree with their own. Of the several developers I've `worked for', very few have been totally responsive to suggestions … all sorts of arguments are put forward for their rejection of ideas, and it's often crystal clear that they are just making excuses because all they're looking for is bolstering of their own egos. In addition, such developers will often dismiss carefully researched bugs because they themselves cannot replicate the problem … rather than spending a little time with the code, looking for the obvious, they'll steadfastly refuse to accept it's there. And beta-testers shouldn't just be looking for bugs … they should be looking for consistency (which is so often easy to lose through a program's development), logicality, intuitiveness and, above all, the ease with which a user can build a cognitive map of the software (a process the designer has _never_ had to go through, but _every_ user will have to). Of course, you're also right that a beta-tester must know his stuff … although, sometimes this can blind the tester to the real problems and he needs to be able to look at the software with different sets of eyes. Some time ago, I wrote an article (which never got published, sadly) all about the art of software testing (in alpha, beta and gamma stages) … it started off with something like `there are four things that any software author needs; four basic tools without which he cannot function. The first of these is his computer … the second is his own imagination … the third is his compiler environment … the fourth is his team of testers.' [ MORE ]