PhotoCredit: Eleganza
Somewhere along the way I think we slightly messed up the commands we give our dog. "Sit" or "LayDown" can be used interchangeably - I am quite sure Winston knows which is which he just doesn't much care. "LayDown" is in itself interpretative and must be clarified with the command "DownDown" if we want him to stay in that position for more than a nano-second.
So I often find myself saying "Down" and quickly followed by "DownDown" as he goes through a his circus-like goodboy routine.
For some reason this all flowed through my head as I thought about Sturgeon's Law: there is Done and there is DoneDone. Sturgeon's Law provokes that 90% of everything is crap. Sturgeon was a Sci-Fi writer (!) and his offering is extremely useful in a world of overload. I find it very calming to know that don't need 90% of any given sample set. My immediate issue is how to set my discovery phasers for optimal crap-filtration in my life? The real issue is how to ensure the work I produce is not relegated to the CrapPile.
How do we optimize the work we do for the last 10%. DoneDone is a lot harder than Done. For insights on how to solve this, I immediately looked to Tom Cargill's thinking on software development. Cargill came out of Bell Labs (arguably the birthplace of our modern technological world) and observed that 90% of work takes 10% of the time and the last 10% of work takes 90% of the time. Some would even suggest the last 1% of development takes 99% of the time. As long as I have been involved in technology, I have always observed this to be true AND also can be applied to where the value lies in every project.
The immediate implications of this are that we need to leave more time (much more time) for finishing projects. I worry about the NeverEndingStoryVibes of this as a solution. It also has some very real money-pit implications.
HusbandMark has always said you should never give a painting back to an artist because they will mess with it. They will not be able to help themselves. A work of art is never done. In many respects software projects are never done either. Code needs inevitable updates and security fixes let alone evolving the product informed by customer use.
So yes, your project will take a lot longer to complete than you think it will be. Yes, it's also much more likely to be crap than you would care to think. And yes, it may never feel like a masterpiece. Finishing is hard. Bake this into your process from the start. Give yourself more time at the end than you think you need. Consider also convening an editorial/finishing committee to help you establish what non-crap might look like. Finally, know that we live in a world where post-release updates are the rule and not the exception. Yesterday's crap might be one clinical trial/peer review/focus group/update/brushstroke away from triumph today.