Thread by @tottinge

I can estimate how long it will take me to drive to Des Moines, and be "close enough." There are lights, stops, traffic, weather to consider, but mostly it's me choosing route and rate.

Route has speed limits, traffic features, raw distance. Those are not up to me, really. I have limits, but the limits make the trip predictable.

The fewer items are under my control, the less likely the estimation.

Can I get to des moines in 3 hours? Probably, if you provide me a fast plane or fast helicopter and clear weather. Otherwise, no.

Now, how long will it take me to learn to dance ballet?
Learn music theory?
Or to come up with a diet that totally works for you to reach your ideal weight?
I've no idea. I don't know how readily I will learn what I need to know and develop what I need to develop along the path.

I don't know if I'm even capable of these things, and will have to wade in to find out. If I have to know before starting, I'll never start. I can't know.
Can I be a professional ballet dancer in a year? Probably not. Who knows?
Within a decade? Probably not, who knows?

Within a lifetime? Probably not.
Heck, I don't know how to play baseball or basketball. What do I know of acquiring physical grace?

"How long will it take you to do a thing you don't know how to do and have never done before, and for which you will have to acquire knowledge and skill along the way?" is a inherently non-estimable situation.

Somewhere inside that range is a software skill. I know a number of things, but not all the things. I'll have to learn about domain, tech stack, deployment, maintenance, etc. A lot of things are _not_ up to me, and some are. Is it semi-estimable?

Will I get it done faster by NOT getting the knowledge and skill that the work really requires, or by diving in and getting all the knowledge and skill that the work seems like it might need?

Is muddling through going to get a faster, safer, better result than developing expertise? Is it a sweeping difference or a marginal one?

How much does the choice to muddle, pre-skill, or JIT sklll affect the task? Can I know it?

Back to helicopter and fast planes: what would be the circumstance at which this could be done more quickly, the enablers and support needed?

Skilled partners who can act as sounding boards and mentors? An appropriate, well-selected framework? The right reading material and videos? Some kind of sandbox where I can try things and make mistakes freely?

Ah. If we work together and share skills, then the work becomes *more* estimable (or less inestimable).

If it becomes more predictable, and predictability is important, then working alone and muddling through (or preskilling for anything that may possibly arise) are both losing strategies.

Burma shave moment:

#pairProgramming #mobProgramming #teaming #ModernAgile