Faithful readers of JavaWorld’s panoply of blogs (and really, I can’t think of any reason why you wouldn’t be among them) have no doubt seen Andrew Binstock’s lament on the waning popularity of unit testing. Binstock is a proponent of unit testing, but everywhere around him he sees it withering — few commercial or OSS offerings to support it, no books, etc. What’s going wrong, he asks? Mike Taylor responds with a bit more good cheer (not surprising as he’s the CEO of Instantiation, maker of the CodePro Analytix testing suite). Paul Keeble over at JavaLobby responded with what I found to be an even more interesting take. He’s all for unit testing as well (it’s interesting to me that nobody in this conversation is actually against it), but he feels that further evangelization of the technique is sort of pointless. The benefits of the process, while intellectually clear, are hard to quantify, whereas the costs are very easy to determine (in extra lines of code and programmer-hours to create those lines of code). To his mind, unit testing naturally recommends itself to the sort of thoughtful person who’s going to be a good programmer. “Right now it’s hit the perfect point of identifying the top 20% of developers and making the other 80% easy to spot.” Software Development