3 ms·

Well, we're mostly software developers - I'm sure we all understand being an hour away from completing a project for the past several days of development. Scal

• ZenoArrow 9y ago
  Sure, but there's a difference between setting short term targets and long term targets. Setting short term targets is only realistic when you're doing something that either has been done before or where it's a variation on something that has been done before. Perhaps you underestimate the time it takes to do something by being unaware of the underlying complexity of a task (or by telling people what they want to hear, if they can't handle pessimistic estimates), but they still have a reasonable chance of being completed on time. Long term targets work differently. If you want the time taken to be as short as possible, and you still give an estimate within a window of time which you have little to no hope to predict, you're basically saying "I don't know". To put it another way, I could have a reasonable guess at what I'd be doing one week from now. On the other hand, I don't really know what I'll be doing 20 years from now. Any estimates I put within a range that is outside my ability to predict would basically boil down to me implying I haven't got a clue. With that said, I'd still put a timescale on an baseless estimate if it served a purpose. It's the purpose behind pulling the number out of thin air which is the point I was trying to highlight.