Comments on: Why do TDD? http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/ Strong opinions, weakly held Mon, 17 Jun 2013 21:27:00 +0000 hourly 1 http://wordpress.org/?v=3.4.2 By: Anonymous http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4206 Anonymous Mon, 12 Dec 2011 13:34:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4206 Ha! Fair enough. It sounds like you may be describing a symptom, however. What is expensive about TDD? (In my experience, only the very smallest projects have any expense of TDD, unless there is a lack of technical leadership on the project) Ha! Fair enough. It sounds like you may be describing a symptom, however. What is expensive about TDD? (In my experience, only the very smallest projects have any expense of TDD, unless there is a lack of technical leadership on the project)

]]> By: Andrew Reiser http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4205 Andrew Reiser Mon, 12 Dec 2011 04:34:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4205 TDD is rubbish. Expensive beyond what customer's can afford. I'm currently watching a test driven project crash and burn. It aint pretty. TDD is rubbish. Expensive beyond what customer’s can afford. I’m currently watching a test driven project crash and burn. It aint pretty.

]]> By: Anonymous http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4065 Anonymous Wed, 02 Nov 2011 17:37:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4065 Ha, well, no it was my own code (which is often alien to me). But doing test-first, the first step is to write a failing test. Sometimes that test that is supposed to fail, actually passes. Ha, well, no it was my own code (which is often alien to me). But doing test-first, the first step is to write a failing test. Sometimes that test that is supposed to fail, actually passes.

]]> By: Thomas Weller http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4064 Thomas Weller Wed, 02 Nov 2011 16:54:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4064 What I imagined from your post was a developer staring at some alien code (e.g. some legacy codebase, or possibly adding some additional tests to a TDD-driven codebase), writing a test for it, and then being surprised because the test passed unexpectedly. This would be close to my experiences, so that's what I understood from your post. But I'm suspecting from your question that this not the case. Right? What I imagined from your post was a developer staring at some alien code (e.g. some legacy codebase, or possibly adding some additional tests to a TDD-driven codebase), writing a test for it, and then being surprised because the test passed unexpectedly.
This would be close to my experiences, so that’s what I understood from your post.
But I’m suspecting from your question that this not the case. Right?

]]> By: Anonymous http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4063 Anonymous Wed, 02 Nov 2011 16:45:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4063 Test-after? Curious, what gave you that impression? Test-after? Curious, what gave you that impression?

]]> By: Thomas Weller http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4062 Thomas Weller Wed, 02 Nov 2011 16:40:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4062 I totally agree, failing test are mostly more enlightening than succeeding tests - or at least trigger a deep review of your system's understanding. One thing though: Sounds more like you were talking about some sort of 'test-after' scenario,  not necessarily about TDD. Regards Thomas I totally agree, failing test are mostly more enlightening than succeeding tests – or at least trigger a deep review of your system’s understanding.

One thing though: Sounds more like you were talking about some sort of ‘test-after’ scenario,  not necessarily about TDD.

Regards
Thomas

]]> By: Fir3pho3nixx http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4061 Fir3pho3nixx Wed, 02 Nov 2011 13:45:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4061 One thing you could say is that TDD is perhaps as much as design activity as what it formalizes assumptions(all benefits). I find that in order to write better, easier to understand tests that prove behavior or interaction, you naturally tend to evolve your code in properly separated, single responsibility objects. Gone are the days of dealing with entangled messes for which there is no evidence to support that it works! :) One thing you could say is that TDD is perhaps as much as design activity as what it formalizes assumptions(all benefits). I find that in order to write better, easier to understand tests that prove behavior or interaction, you naturally tend to evolve your code in properly separated, single responsibility objects. Gone are the days of dealing with entangled messes for which there is no evidence to support that it works! :)

]]> By: Anonymous http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4059 Anonymous Tue, 01 Nov 2011 21:25:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4059 For me, TDD != unit tests. I'm very deliberate when I mention TDD that I don't mention unit tests. TDD just means test-driven development, not "unit-test driven development". For me, TDD != unit tests. I’m very deliberate when I mention TDD that I don’t mention unit tests. TDD just means test-driven development, not “unit-test driven development”.

]]> By: Scooletz http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4058 Scooletz Tue, 01 Nov 2011 20:42:00 +0000 http://lostechies.com/jimmybogard/2011/11/01/why-do-tdd/#comment-4058 I hope you don't mean only unit tests but also integration ones. Recently I reviewed the way Castle is tested and must admit that being strict to unit tests is not always the best way. I hope you don’t mean only unit tests but also integration ones. Recently I reviewed the way Castle is tested and must admit that being strict to unit tests is not always the best way.

]]>