"It's tough to make predictions, especially about the future." —Yogi Berra
I've recently watched the “Software Estimation Best Practices“ Web cast. It has opened my eyes to what has to be one of the easier mistakes to make in software estimation: “The perfect is the enemy of the good.“ The idea being maybe a practice isn't perfect but overall and in general it is good. For example, I'd much rather have estimates based off of lines of code, screen shot number or area, then gut feelings or hope.
Anyway, I've noticed that mobile software estimates can be a very tough nut to crack. I haven't found any resources on mobile software estimates, since estimation practices tend to be language agnostic, for the most part. I figured I'd share what I have in hopes of discussion of what works and what doesn't work.
MSDN Webcast: Software Estimation Best Practices – Level 200
https://msevents.microsoft.com/cui/EventDetail.aspx?culture=en-US&EventID=1032249520
Reducing Bias in Software Project Estimates
http://www.stsc.hill.af.mil/crosstalk/2000/04/peeters.html
Painless Software Schedules
http://www.joelonsoftware.com/articles/fog0000000245.html