Comments on: Two ways to react

By: Edu

Edu — Thu, 16 May 2013 15:51:58 +0000

With some products customers want to test each release couple months before using them in production. Then it makes sense to ramp up testing. However, often the gut reaction is to add (non-autotested) specifications instead of testing.

By: Gus Power

Gus Power — Sun, 12 May 2013 19:29:15 +0000

Reminds me of discussions about MTBF (mean time between failure) vs. MTTR (mean time to recover) . Turns out that things eventually do go wrong and being able to recover quickly and gracefully (and to know that things have actually gone wrong in the first place) is pretty cool. Who'd have thought!

By: Roland

Roland — Fri, 10 May 2013 12:45:50 +0000

I understand you're probably talking about the reaction in general here, rather than specifics, but I think it is important to realise that fear is sometimes an appropriate response (I would prefer you were fearful of releasing something that would expose my credit card details for example).

I think the interesting thing is more that whilst one view of the world may be most appropriate for a particular style of project, figuring out how to recognize when the opposite approach is necessary on specific occasions, and how to deal with that in the team culture.

By: PilotBob

PilotBob — Thu, 09 May 2013 21:07:33 +0000

Why does it have to be one response or the other. Why not both at a measured level?

Remember the answer is always: It depends!

BOb

By: Dan Tao

Dan Tao — Thu, 09 May 2013 15:59:02 +0000

I agree with this dichotomy, though I think it depends quite a bit on the organization. It's easier to optimize for fixing mistakes when you're a small company and less is on the line (which is why it kills me to see small companies taking the former approach). The consequences of mistakes for larger companies with huge customer bases are more uncertain and hence scarier.

I also think some people reading this will naturally agree while others will naturally propose counterexamples. But that's just how people think, coming from our own experiences. Personally, if I make a really dumb mistake, chances are I have an intuitive sense of whether that was totally stupid and easily avoidable or if such mistakes are likely to happen no matter what safeguards are in place. I would rather prevent mistakes where preventing them is easy and obvious and imposes little or no friction on the organization in general. But I'd like to think that most of the mistakes I make aren't like that; hopefully, they're more often subtle mistakes that would have been difficult to anticipate. So I lean towards the latter approach.

I guess what I'm saying is that the way a person leans is likely to be shaped by what sort of mistakes he/she has generally experienced and observed. If most of the mistakes you observe seem to stem from stupidity, you might think the first approach actually makes sense.