Glen B. Alleman, MSSM, Vietnam 69-70 · @galleman
21st Jul 2015 from TwitLonger
Interview with Zach and Ryan #Noestimates Part 2
By the way all the conversation about not being to decide without estimates do not remove the huge issue of bad management, bad estimates, and bad results from bad management and bad estimates.
All this "estimating problem" has been around forever. From simple 6 weeks and two developers to Joint Strike Fighter, to ERP, to major infrastructure. Estimates are a problem. This is well documented, studied to death, and solutions galor. We have solutions in our domain of "software intensive systems." They work sometimes, many times not.
The problem is hard. It's not "wicked" because we know what the problem is. So if anything good has come from #NE it's recognition that estimating is hard. It's needed and many times mandated, but it's hard.
So what? Lot's of things in our domain are hard. Does that mean "let's stop doing them?" Not likely useful to those paying us.
One conclusion is the visceral response to "estimating as evil" likely comes from developers experiencing bad management. And Not estimating isn't going to fix that. It's treating the symptom not the root cause.
But it is clear, there are times when estimates add little value and times when they are critical to the success of our work. How to determine "when" is a problem that needs to be addressed. And the answer has to come from those providing the money, not just those spending the money. Until that is understood, #Noestimates will be an "argument without end."