[Previous] Learning From Emergencies | Home | [Next] Making Progress and Jordan Peterson Quotes

Integrating Goldratt's Philosophy with Fallible Ideas

i do lots of stuff and mention it and ppl usually don’t follow along and do my cool new projects. sometimes a few ppl read or learn 1/4 of it, a bit late, and then don’t take it very far.

and they aren’t doing a stream of their own awesome projects either.

this is my overall impression.


well they are overreaching. they don’t have spare capacity to jump on opportunities as stuff comes up. they are booked up too much. if they only booked 50% of their time, then they’d have time available for cool opportunities as they come up.

in the name of efficiency, ppl will schedule all of their time, and then there’s no flexibility to get anything else done, and they miss out on tons of great opportunities. and they are often in crisis mode b/c they can't even do everything on their schedule, but they made promises to do those things, and then they have to take inefficient steps to deal with the crisis.

this is a ton like in The Goal, by Eli Goldratt, where factory work stations need excess capacity. and ppl intentionally try to get rid of any idle time in the name of efficiency. (read the book if you wanna understand this post better, and also b/c it's a really great book. there's tons of relevant stuff in the book which i don't go over here. also read his book, The Choice.)

this attitude ends up chasing lots of local maxima at the expense of global ones. that’s what busy schedules do too.

in The Goal, for example, they run robots 24/7 at their manufacturing plant so the robots won't be idle. but they end up producing too much of the stuff the robots can make, which makes things worse. and they have ppl at the heat treat oven go do other tasks while it runs, but then they're late to come back and lose super valuable time starting the next oven batch, so what they're doing is actually much, much worse than having ppl just sit around and wait by the oven and never be late to get the next load started.

paradoxically, ppl have their lives so full they never get anything done. which is exactly one of the main problems in The Goal.

it’s also like poor ppl who stay poor partly b/c, lacking spare money, they can’t jump on sales or bulk discounts very well. ppl who lack spare/idle time, b/c they are so time poor, end up using their time really inefficiently exactly like poor ppl use money inefficiently.

ppl often will make some attempt to free up time for a specific purpose. they want to do X so they manage to shift some time on their schedule from something else to X. but doing that never solves the problem of being overbooked in general, let alone giving one the excess capacity to jump on good opportunities and do things with longterm benefit (in the same way having excess money in the bank lets you use your money better).

i'm not advising you sit around doing literally nothing for long periods of time. but for most ppl it's really not hard to find things to do! most ppl have plenty of stuff they want to do. if you aren't doing anything for a while, don't rush to start an activity. give yourself time to think sometimes! and if you aren't doin much thinking right now or you finish thinking, ok, then consider your priorities and pick something good to do. also place a higher value on activities with flexible scheduling that are easy to interrupt and take breaks from, and activities that offer fast returns on investment so it's fine if you quit.

that's how i organize lots of my time. i avoid scheduled obligations and longterm commitments. instead of literal idle time, i have a lot of flex time where i have some things i can do, but also it's no problem to not do them if something else comes up.

examples of flex activities including reading, writing, playing games, watching videos, listening to podcasts, and having discussions. at a moment's notice i can redirect all my time away from those things and have all day to do something else, no problem. (it doesn't have to be something else, either. if i get really into reading something, and i'm learning a ton and excited by it, then i can redirect time to that. when i want to read something, i'm usually pretty much only limited by getting tired, not by my ability to free up time for it immediately. b/c of how i run my schedule, raw time is much less of an issue for me than for most ppl, so i have gotten to put a lot of thought into some more advanced issues of managing energy/attention/focus, which are where people's bottleneck should be. there are more hours in the day than hours per day you have the energy to do good thinking, so if time is your bottleneck you're doing something wrong!)

people managing their time, attention and mental energy very badly is also why Paths Forward seems so implausible to them. so to try to save time they actually act irrationally which is very destructive, especially for supposedly truth-seeking intellectuals.

also, in the majority of cases, concretizing plans in advance is premature optimization. don't tie down what you're going to do in advance when you have less information about what's best to do. in some cases you get benefits from partially deciding early (e.g. you can buy supplies for a project in advance), but you should usually avoid it. don't concretize plans early without a very clear reason it's beneficial, and only concretize to the extent needed to get the benefit, and be wary of activities that require a lot of concrete, advance planning.

ppl often push you for concrete planning, implicitly. cuz they are overbooked on concrete plans, they never get to the less concrete plans. so you have to schedule concrete plans with them in advance or you can't do anything with them. so their lack of idle time is causing every additional thing added onto their schedule to be concretized in advance (which is worse). the busy schedule itself is preventing the best activities and only letting lesser activities get onto the schedule! the busy schedule is causing tons of premature optimization.

making the right changes is so much more important than spending tons of time making tons of changes. it's not "efficient" to spend all your time that way, that's a local maxima. you'll get more productively accomplished, in a fraction of the time, if you do the right things. this is a theme of Goldratt which fits with FI well.

if flex time loses 10% against the local maximum to retain flexibility, and you do a different much better activity once a week (use the flexibility a minority of the time), you can come out way ahead.

e.g. you might play video games that are easily interruptible. this reduces your selection of games, so you get to play a slightly less ideal game on average. some days, as a result of doing this, you come out behind – play a less ideal game with no upside on that day. however in the bigger picture you come out ahead – the days where you gain a large benefit from the flexibility more than make up for the minor downsides.

don't screw up the important stuff b/c you were too distracted optimizing minor details! this is a big deal, and is in The Goal (having ppl wait around at the bottleneck work stations, spending most of their time idle, improves production b/c those work stations are so important that it's better to wait around for hours than to be a little late from multitasking.)

you can look at this in terms of local vs. global evaluation of results, or seen and unseen (Bastiat's take on the broken window fallacy and selective attention).

what's natural for me personally is to see that, right now:

1) play game with 9 value per hour, and have a 10% chance to interrupt it to do something worth 50 value per hour

is more points, now, than:

2) play game with 10 value per hour.

cuz i look at expectation value (a well known poker concept meaning the avg expected value of what ur doing, regardless of what happens today), so i even get the local case right. the expectation value is 90% * 9/hr + 10% * 50/hr which is 13.1/hr (that math doesn't cover the mixed case where you do it for a while then interrupt partway through, so the real advantage is smaller but, given reasonable assumptions about the unstated details, it'll still be an advantage. note: you do need to be careful with risk when dealing with expectation values, especially when dealing with high stakes all at once. e.g. i'd rather have a million dollars than a 50% chance of 2.1 million dollars, even though the second option is a little more money on average. the only way i'd take the second option, which comes out to 50k more on avg, is if i made a deal with a bank or rich guy. e.g. the bank gives me a million dollars plus 10k, and i pick the 2.1m choice and give them the full amount from it. then i get 10k more than if i picked the 1m option, and the bank comes out 40k ahead on avg, and they have enough money to deal with the variance.)

Elliot Temple on February 7, 2018

Comments (8)


> what's natural for me personally is to see that, right now:
> 1) play game with 9 value per hour, and have a 10% chance to interrupt it to do something worth 50 value per hour
> is more points, now, than:
> 2) play game with 10 value per hour.

The logic of what you say is straightforward, but I have great difficulty estimating the values involved. That makes it very unnatural for me personally to see these kinds of relationships in real life situations.

If we fix game #2 as 10 value per hour, what's the actual value of game #1? You say it's 9 for your example, but if I'm thinking about real stuff I do in my life I would find it hard to develop an explanation that survives criticism for why game #1 is 1, or 5, or 9 value per hour. The relative value of interruptable and non-interruptable stuff I can do is really hard to rate, or I lack the skills and a good method for rating it. If we're talking straight-up dollars, the interruptible / unscheduled stuff I *know* how to do without adding risk to my life versus the scheduled / non-interruptible stuff I'm doing, my best guess is a lot lower than 9.

But suppose I solve that problem, and arrive at your estimates for the value of game #1 and game #2 by some method I'm satisfied with. Now I'm faced with determining, what's the chance I'll encounter an opportunity to do something with higher value if I choose game #1? You estimate 10% for your example. Here again, thinking of my actual life I find it hard to come up with an explanation for why that chance would be 1%, 5%, 10%, or 20%. So I need a method I'm happy with for that too.

By now you can probably guess where this is going. You say the value of the opportunity you get to do if you choose the interruptable game in your example is 50, but I don't have a method for arriving at that either IRL either. Could be 20, 50, 500, or 10,000.

I know my current life is time-poor, filled with local maxima and missing out on opporunties that might take me closer to global maxima if I had more free time.

However: I have no good way to estimate how much value I'd actually generate by doing something that was less scheduled / time intensive. AND I have no good way to estimate the chances I'll find more valuable activities to do in the freed-up time. AND I have no good way to estimate how much higher than the local maxima I'm currently at the opportunities would take me if I siezed them.

All of that uncertainty is why I treat freeing up my schedule as a luxury. It's something that I intend to do solely for pleasure, when and to the extent that I can afford it. Because I just can't make the kind of value case for it in my own life that you say comes naturally to you.

PAS at 9:54 AM on February 9, 2018 | #9492
the point was about thinking in terms of expectation value, not specific techniques for estimating values in situations. put another way: i can see the *option value* of things even when the option doesn't get used. (I don't think this is super amazing, but it's important and a lot of people seem to find it pretty unintuitive.)

in the big picture: i'd just free up time on principle, not b/c of a specific estimate. you can't plan out and calculate every detail of your life, but you can follow big picture guidelines that make sense.

i didn't say the word "naturally" and it's misleading. what i've done for expectation and option value stuff is get my fast thoughts reasonably in line with my slow thoughts (and make the slow thoughts reasonable). anyone can do that kind of thing.

curi at 11:28 AM on February 9, 2018 | #9493
> i can see the *option value* of things even when the option doesn't get used.

I agree the option to do something else has value even if it doesn't get used. What I don't know is whether that value is greater or less than its cost.

> in the big picture: i'd just free up time on principle, not b/c of a specific estimate.

I think such a principle in my own situation is *luxury*. Cuz I don't know enough to know on principle that I can expect to derive more value from having the free time than I'm currently deriving from my scheduled time.

> i didn't say the word "naturally" and it's misleading.

I assumed that naturally was substantially the same as "what's natural for me personally". Other than the fact you didn't say the specific word "naturally", I don't understand what's misleading about it.

> what i've done for expectation and option value stuff is get my fast thoughts reasonably in line with my slow thoughts (and make the slow thoughts reasonable). anyone can do that kind of thing.

I don't understand how this applies to what we're talking about. I don't see any disconnect between my fast thoughts and my slow thoughts regarding time and schedule stuff. In both cases I'd prefer to have more free time than I currently do. But I can't defend that preference against criticism on value grounds.

PAS at 12:58 PM on February 9, 2018 | #9494
"natural" = intuitive = fast thoughts/estimates **NOT** natural = inborn (the potential misleading thing).

you used the phrase "comes naturally" which makes it sound like the 2nd one (was learned naturally or was inborn, instead of just feeling natural now, at the current time, for some reason).

i didn't say fast/slow disconnect was your problem. i was explaining what i was talking about. from your comments, seems like you're unclear on the right slow analysis. ok.

curi at 1:11 PM on February 9, 2018 | #9495
Ah OK. I recognize some people do take "comes naturally" to imply inborn. I didn't intend it that way.

Ex: Driving comes naturally to me now, but it definitely wan't inborn. Driving was pretty hard for me at first compared to other people.

Do you have any ideas about where I am unclear on the right slow analysis of free time versus scheduled activities?

PAS at 1:21 PM on February 9, 2018 | #9496
big picture, i'd guess you of underestimate the value of flex and idle time. i know it's tricky to create a lot of those in your specific life b/c you get paid for scheduled time. (my paid programming work is primarily flex time – i can do most tasks whenever i want on a non-rushed timeframe.) but i guess you could find some more improvements (and give greater consideration to some big changes) if you saw more value in them.

reading The Goal and The Choice could help (while also keeping some of what i said in mind so you can connect it to the books – esp The Goal's stuff about idle time and efficiencies – and understand it more). they're easier reading than Rand let alone stuff like Popper. they are great and worthwhile anyway.

curi at 1:30 PM on February 9, 2018 | #9497

Underestimating the value of flex and idle time

> big picture, i'd guess you of underestimate the value of flex and idle time.

Maybe. I've thought some more about what's driving my estimate.

If we de-personalize it for a minute and talk about people generally, I think some things are pretty common:
- People don't know what to do to create value
- People aren't self-motivated to create value
- People are more productive working in with others than working alone

This is all due to personal knowledge and choices, not nature. A few people aren't like those traits in meaningful ways. Fewer still aren't like any of them at all. Regardless, it's super common to approximately fit all 3. I think you've said this something like that elsewhere. I'll speak of those traits together as not having much of a self.

People without much of a self in the way I described generally do fuck-all with their flex and idle time. They often destroy value, and if they manage to create instead of destroy it's usually a lot less than they create when they're scheduled and directed.

Where you mention industrial examples from Goldratt like waiting for ovens, I think we have to be careful characterizing this as idle time in the larger, personal-life sense. The worker is not idle in the sense of being free to do whatever he wants while he waits for the oven. He's been given a task and is following orders to sit there and just wait until the oven is done. If such workers were allowed to do what they want - take a nap or, say, play interruptable video games while the oven runs, my guess is that factory output would drop.

As a result of how common lack of self is, lots of infrastructure in society is designed around it. Common arrangements are set up to get more value with less variance from people without much of a self. They aren't particularly well structured for people with a lot of self.

Even people with more of a self than average are trained (by schools and social pressure etc.) to fit into the low-self model better than they otherwise would, because so much of society's infrastructure is oriented that way.

That infrastructure relies on stuff like schedules and promises and hierarchy of authority and consistency and forced focus etc.

So I think the value of what an individual will do with flex and idle time *in the context of how much self they have* has to be compared with the value of what that individual will do with scheduled time *in the context of the infrastructure that goes along with that scheduled time*.

Back to my case: The main reason it's tricky to create a lot of flex and idle time in my life isn't because I get paid for scheduled time. If it were just that, it'd be easy to taper down a bit and experiment.

The reason it's difficult is because I'm part of an infrastructure that was designed to either pay you for **enormous and continually recurring predicatable blocks** of scheduled time, or stop paying you at all. It's set up that way because, for most (low-self) participants in the transaction, that's actually value maximizing.

I may attempt to "hack" that infrastructure and reduce the size of my recurring blocks. But I do so knowing that it's not something the infrastructure was well designed to accommodate, and it's going to cause some problems.

Back to the estimate: It's possible I'm underestimating the value of flex and idle time because I'm underestimating how much self I have.

It's also possible I'm underestimating the value of flex and idle time because I'm overestimating the value I get from fitting myself into an infrastructure that was designed for people with less of a self than I have.

But it's also possible I'm correctly estimating those things or going the other way (overestimating how much self I have and underestimating the value of fitting in).

Which is why I come back to treating idle and flex time as luxury. If I get more experience with it maybe I can revise my estimates.

I have started reading The Choice; thanks for the recc.

PAS at 6:13 PM on February 10, 2018 | #9498
I agree with the general analysis. I was talking about the value of flex or idle time in a life being run rationally. I agree the value changes dramatically if one runs his life in other ways.

Regardless of how much self you have ... my guess is you don't have a full, clear picture on how flex time fits into a great life, how all that value and upside works. So there's more there you could understand, which would raise your opinion of it.

curi at 7:26 PM on February 10, 2018 | #9501

What do you think?

(This is a free speech zone!)