Conversation

Suppose you have a creative project that is 10% obvious tasks, 80% non-mysterious vague ones, where details need thinking through, 10% “inspiration-needed” creative-load-bearing strategic core bits. Should you plan a bit to change the 10-80-10 to say 20-75-5, or just dive in?
14
76
For eg: for a new website, the 10% = pick a domain name, CMS. 80% = basic information architecture, get content flowing. 10% = get the site’s voice to cohere and pop, usually around a couple of pieces of anchor content around which the whole project will start self-organizing.
2
9
The waterfall vs agile debate is sort of an approximation of this question. As is the debate around “plan to throw one away” vs not (Brooks vs Spolsky). If you actually try various approaches, it’s pretty subtle. I’ve never found a one-size-fits-all approach to how much to plan.
1
11
When you first think of a creative project, it is sort of naturally born with 10-80-10 proportions of clear/vague/mysterious task buckets on the to-do list. That’s what makes it a creative project rather than a cookbook project or a bureaucratic chore.
1
10
Planning is reshaping the 10-80-10 to trade-off do ability and motivation. There are 3 things you can do to reshape: 1. Make mysterious things vague & non-mysterious 2. Make vague things obvious and clear 3. Mystify obvious things 1+2 = ordinary planning 3 = creative planning
1
15
Most planning focuses on 1+2, which tends to increase doability while lowering motivation. That turns a more creative project into more of a cookbook or bureaucratic one. You basically try to squeeze out ambiguity (mysterious —> vague) and uncertainty (vague —> obvious).
1
15
The result is a project with “lower risk” that’s more ready for “execution,” that you’re less motivated to actually do. Partly you’ve just spoiled the surprise, but mainly because the upfront open-loop “visionary” derisking (mysterious —> vague —> obvious) tend to be uninspired.
3
12
Mysteries solved and vagueness clarified in improv mode *after* you’ve started tend to be way more inspired. They often simultaneously increase doability and inspiration, expanding Pareto and making you feel more powerful. But you’ll be more inefficient and wasteful getting there
1
12
But there’s a kind of upfront planning you can do where doability may or may not increase, and may even slightly decrease, but motivation can radically increase. This is moving things “upstream” from obvious to mysterious. Take a clear part of the project and make it mysterious.
Replying to
In our trivial example of a website, finding a .com domain name is an obvious task. But choosing instead to get a weird tld like say .club makes it a marketing mystery since you’re not quite sure how it will behave now. You have to discover it’s marketing properties.
1
7
These upstream moves tend to refactor the whole initial state of the project. The 10-80-10 proportions of obvious, vague, mysterious may not change, but the contents will move around in unpredictable ways based on the logic of the mystification, and often 10x inspirations.
1
2
The result is unmistakeable. You can take a project you’re feeling resolutely meh about and suddenly it is super inspired and you’re itching to do it so badly you can’t help but just dive in and start doing the first obvious thing. It can destroy “ordinary planning” impulses.
2
9
This can even be dangerous. The ideal amount of ordinary derisking planning is not zero. If you kill that impulse entirely you can go off half-cocked and really screw things up. Especially when the first few steps involve spending a lot of money, as in high upfront-capex projects
1
11
The trick is to do enough mystification through creative planning to trigger the itch/action bias, but not so much that it destroys all rational planning impulses, creating a whole new kind of impatience risk. A good way to think of it is trial-and-error budget risk.
1
16
If your budget is “3 attempts” and you over-inspire yourself so much that you’re now rushing to blow it all on 0.5 attempt where you’ll run out of money before you even have a first trial outcome, time to slow down and do some ordinary planning too.
1
3
Industrial modes of work are extremely high-capex biased, so naturally have safeguards against a 1-to-0 outcome. A one shot budget being blown to 0 by narcissistic overvisioning and implosions.
happy season 4 GIF
GIF
1
1
This is ironically a Soviet-style “planning” failure mode. Promise of a false dawn undermining actual rational planning and derisking impulses. The capitalist failure mode is to artificially separate out mystification impulse into a phase labeled “brainstorming” or “exploratory”
1
11
Often, when done in a highly privatized, corporate institutional setting, thus has the effect of creating innovation theater on the sidelines of highly conservative shareholder-value-focused bureaucratic chore/cookbook/formula main flow of work.
1
10
I’ve become highly suspicious of “vision” thinking outside an active tinkering/muddling-through experimental activity. The active trial-and-error, even in highly inefficient and unsinspired (= low hit rate) early muddling-through phase, acts as a control rod to prevent meltdown.
1
22
This is not about theory vs practice. Even in purely theoretical work, like physics or writing, there’s a doer context (doing the math, doing the draft writing) vs purely isolated theorizing context created by exercises like say mind-mapping.
1
7
The doer context cannot be a safe-failure pure test environment. It has to be production tinkering/muddling. It has to be at least open-play as they call it in war games. Or field trials over lab trials.
1
9
Half-assed things in a full-assed way > full-assed things in a half-assed way. Prioritize testing half-assed rocket (whole system) rather than a full-assed engine (subsystem) where possible. Of course eventually you have to do both. But sequence for maximal open-world discovery.
Quote Tweet
The key is to do half-assed things in a full-assed way rather than full-assed things in a half-assed way.
3
26