[commandlinefan]

> Never been good at estimation

Me neither. When I was first starting out, that really stressed me out a lot until I realized that I didn't work with anybody else who was "good" at it - that is, I didn't work with or know anybody who could take a list of requirements written out in English and produce a timeline that had any relationship to how long the software would take to be ready to use.

Been doing this professionally since 1992. I still haven't met anybody who was "good" at estimation.

[handrous]

I've reached the point where I think that if you're not going to do the NASA Space Shuttle program thing of specifying the whole program to the smallest detail before you start writing the actual code, you may as well just start working, release often, and evaluate periodically whether the thing looks on-track to be worth the cost, cancelling if it's not. Just spend the estimation money on development instead.