tag:blogger.com,1999:blog-25922407.post6828763080879194362..comments2024-03-25T22:12:49.064-07:00Comments on Gerald Weinberg's Secrets of Writing and Consulting: Why People Don't Instantly Buy Into Agile Methods: A Catch-22 Gerald M. Weinberghttp://www.blogger.com/profile/05902673055244863609noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-25922407.post-77744771112747837812012-12-06T20:42:12.076-08:002012-12-06T20:42:12.076-08:00Besides the argument about familiarity with bugs, ...Besides the argument about familiarity with bugs, there are also the arguments about the potential prohibitive cost or unacceptable risk of fixing the bugs (as you write about in other blog posts). I think this can offer a way around the catch-22.<br /><br />The way I see it, the argument to sell agile should not be based on wonderful new features, but rather the possibility of agile speeding up the feedback loop which makes it possible to do the same at a lower cost and less risk. Thus preserving familiarity and pleasing management at the same time.<br /><br />As a side note, I'll also offer an opinion on the question "If they payoff is so huge, why aren't we raising software quality to new levels?". I think the number "$600,000,000,000" should be put in context of (at least) the two following numbers:<br /> 1. The cost for fixing all bugs<br /> 2. The cost for working without the software (e.g. if it were not released yet due to bug fixing)<br />In that context I don't think the payoff is that huge.Georg Lysenhttps://www.blogger.com/profile/02028041823325365253noreply@blogger.comtag:blogger.com,1999:blog-25922407.post-9485719732066820852012-10-02T03:41:12.648-07:002012-10-02T03:41:12.648-07:00Hi Gerry,
Interesting. I wonder how different it...Hi Gerry,<br /><br />Interesting. I wonder how different it would be if we treated software problems more like aviation than ground transportation.<br /><br />When a plane crashes, there is always a thorough investigation looking deeply for root causes. As a result, travelling by plane is extremely safe. Meanwhile, every time an aircraft so much as hiccups it's front page news.<br /><br />I guess it's a function of the consequences of something going wrong. A 'fender-bender' at 35,000 feet will have an immediate and dire outcome, whereas one on the ground may result in some broken glass.<br /><br />The same applies to both software and the process used to create it. If the consequences of broken software are innocuous, then the functional and economic need to ensure that it's not broken is reduced. From a process perspective, if an organization is getting away with a serial, phase-gate process without any dire consequences, then there's little incentive to change.<br /><br />I'll conclude with the famous quote from Stalin, "The death of one man is a tragedy. The death of millions is a statistic." As messed up as that sounds, he hit the nail on the head with respect to human nature.<br /><br />Dave Rooneyhttps://www.blogger.com/profile/00934266695771206046noreply@blogger.comtag:blogger.com,1999:blog-25922407.post-84905932514686660192012-10-01T19:03:37.658-07:002012-10-01T19:03:37.658-07:00You've always had me as a fan, but this post k...You've always had me as a fan, but this post knocked it out of the park for me. Sure, I'm just an Agile guy, but I would be more than willing to have your baby.<br /><br />Ok... just kidding...<br /><br />On a serious note, we make the same points in our Agile QA courses and the biggest aha moment we have is when we state: quality is relative to the context. Putting your awesome number aside about the cost of poor quality software to Americans, I wonder how much we are spending as a nation on useless defect metrics that really are measuring quality of a desired product feature, but rather are measuring the ability to find bugs in software whether the market really wants a product feature or not. I probably don't have enough bits in my stack to calculate that number. Great post!!!Devin Hedgehttps://www.blogger.com/profile/08415796798988191870noreply@blogger.com