Comments on: Unit Tests and Developers under Pressure http://www.elilopian.com/2008/12/04/unit-tests-and-developers-under-pressure/ Creating better software Mon, 02 Jan 2012 09:09:26 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Eli Lopian http://www.elilopian.com/2008/12/04/unit-tests-and-developers-under-pressure/comment-page-1/#comment-30305 Eli Lopian Mon, 15 Dec 2008 19:03:35 +0000 http://www.elilopian.com/2008/12/04/unit-tests-and-developers-under-pressure/#comment-30305 True Anon There is only one solution: The manager. It is the managers job to create professional software and to do so the manager must make sure that the cowboys become professional too. Even if it seems counter-productive to writing unit tests instead of another feature - this is what professionalism means. True Anon
There is only one solution: The manager.
It is the managers job to create professional software and to do so the manager must make sure that the cowboys become professional too.
Even if it seems counter-productive to writing unit tests instead of another feature – this is what professionalism means.

]]> By: Anon http://www.elilopian.com/2008/12/04/unit-tests-and-developers-under-pressure/comment-page-1/#comment-30115 Anon Tue, 09 Dec 2008 23:58:02 +0000 http://www.elilopian.com/2008/12/04/unit-tests-and-developers-under-pressure/#comment-30115 I think this is an excellent point/post. Often the cowboy coder feels he doesn't need tests anyway because he 'knows what to do' and 'knows what the current code does'; so he uses this 'pressure' from management(or whomever) as an excuse to 'get 'er done'. I can't count how many times I have sat and watched the unreformed cowboys I work with do build after build after build after build on a revolving door of bugs all related to some project that was 'high priority' and therefore the cowboy coder just went out and did it, totally abandoning the process. On many of these occasion once the dust settled the time spent on the revolving door of bugs was greater than the original coding time, and far greater than if the code would of been done right, following the process. But then, that shouldn't come to a surprise to anyone who truly understands the value in testing. But note: there are two problems here: the cowboy coder, and the catalyst - the manager who encouraged this kind of behavior. Both need to be dealt with, not just the coder. I think this is an excellent point/post. Often the cowboy coder feels he doesn’t need tests anyway because he ‘knows what to do’ and ‘knows what the current code does’; so he uses this ‘pressure’ from management(or whomever) as an excuse to ‘get ‘er done’.

I can’t count how many times I have sat and watched the unreformed cowboys I work with do build after build after build after build on a revolving door of bugs all related to some project that was ‘high priority’ and therefore the cowboy coder just went out and did it, totally abandoning the process. On many of these occasion once the dust settled the time spent on the revolving door of bugs was greater than the original coding time, and far greater than if the code would of been done right, following the process. But then, that shouldn’t come to a surprise to anyone who truly understands the value in testing.

But note: there are two problems here: the cowboy coder, and the catalyst – the manager who encouraged this kind of behavior. Both need to be dealt with, not just the coder.

]]>