unironically probably. all this overthinking of how best to prioritize things ignores the fact that the problem is almost never that a person doesn't distribute their time working well, it's that they don't spend enough time working, or they spend too much time switching tasks because they keep changing their mind about which task is most important (which itself is typically a method of avoiding work, rather than an actual issue of prioritization). If you have 3 things to do, we can assume without loss of generality the optimal order to be 1 2 3. But even completing the tasks in order 3 2 1 will be fine in 99% of cases, even if the work would have been completed 20% quicker if done in the optimal order. 20% more time seems like a lot of time until you realize that you might spend 50% or even 100% or more of the time it would take to complete the tasks on stressful procrastination/avoidance, including time spent 'prioritizing'. That being said, gives that the majority of this wasted time is psychological, I've found that the 'optimal' way of actually working is to pick the simplest, easiest tasks to start with, even if they are low priority, and knock a few of them out. This creates quick wins that create momentum and will also unburden the mind of the psychological stress of having a long todo list (which distinct from having a todo list that requires a long time to complete). With that momentum, I then jump into the most psychologically distressful talk, ie the one I am most inclined to avoid (which is also typically the one that most needs to be done) and complete as much of it as I can. At this point I generally am in work mode and can go for 8-10 hours straight with few breaks.
@@edism medium is shit because anything I try to read, it hits me up with a sign up. Never ever read anything from medium. I’d rather read from anywhere else. Prime making videos like this means me indirectly being able to read medium articles
You can tell they’re a student not just because they remember all of this math, but also because they naively assume that the outputs from all tasks have equivalent value.
I love how this man went and decided that the curve should look like that without any real world measurements and built a whole thesis around it instead of using that part of his brain to estimate the time it will take to learn Physics.
Right. I don't care to pay for medium and I didn't pay much attention to anything not mentioned in the clip, so I'm just going to assume that none of this guy's assertions were attained from or sourced to experts in relevant fields i.e. psychology/biology instead of math... peak solipsism IMO. We don't solve all of our problems with philosophy anymore. Nothing is taken as a given when it doesn't have to be.
Well, the Poisson distribution is a statistical distribution that models number of occurrences of some event in a specific time. In his model, I noticed what looks like an exponential distribution (the part with e) which is used to model the time it takes for an event to occur. These distributions are well established. If you are modelling anything similar, that's the best starting point. But I think the author could have explained more clearly. There seemed to be too many asspulls in the article
@@NphiniTYea that's why Prime went straight to drawing a different curve. There was no non-mathematical explanation for the values inserted. There are tons of amazing statistical models with strong accuracy at high sample sizes... but none of them work if your underlying facts are incorrect. This happens regularly, a good example being political polls. Asking the wrong questions or in the wrong way gives you fantastic data on something that isn't your intended question. The same thing will happen here. As the questions being asked by the formulas are not about the multitude of psychological and physiological states of the user, those mechanisms which determine the vast majority of your decisions will go unaccounted. Enjoyment and Effort are far too simple. I would be interested to find out if other disciplines were consulted and somehow accounted for, it just doesn't seem likely.
@@Varadiio If that is so then I'd appreciate if you linked the source that the person used. I simply wish to confirm that the time/effort graph isn't completely made up. I have not said or implied that philosopy is relevant, so I'll just assume you are projecting since pepople not involved with philosophy wouldn't use "solipsism" to describe a self centered person. I assumed there were no citations because I didn't see any, which is, in my opinion, completely fine. Besides I wasn't arguing wether or not the graph is accurate, I simply pointed out that they can clearly utilise assumptions and still refused to use them when they are probably the optimal way forward.
@@NphiniT Thank you for clarifying where that graph came from. What really bothered me is that there was no reasoning provided for using that graph. I can get behind it being accurate to their experience, or any number of reasons, but I can't get behind needing a reason to study for a certain amout of time when we apparently don't need any for half the things we based a thesis off of.
@@chrispy_baeconsure, routine is the code that lives in the main function. subroutine is all function calls except for main (because you can always recursively calls main) micro task is deferred code that executed later (usually end of routine or subroutine) like destructor in C++ or defer block in Zig
@@nrwchd routine is something you do repeatedly usually in the same time of the day voluntarily. subroutine is just a part of that whole routine and a micro task is something small you have to get done, but it is not usually the same thing or at the same time. Is how I would describe these terms
My productivity hack for reaching flow state quickly is to ensure that you have an important reason to meet a deadline, then start the task with the exact amount of time needed to complete it left before the deadline. If you don’t have time to not be at peak performance, you will be at peak performance. (You will also quickly learn that flow state can last 12 plus hours, it’s the best feeling, better than heroin)
I have 3 todo "apps": - On my phone, note app - On my computer, sticky notes for permanent and notepad for temporary notes - On real life, a blank sheet of paper inside a plastic case which lets me use a whiteboard marker on it. It's like a ultra thin portable whiteboard and it's amazing. I learnt this trick preparing for an online exam once My brain sticks them together
I am in a hate-love relationship with physical notes at this point. I prefer them over digital note taking, but once I start noting things down, my desk is quickly covered in post-its and full sized paper. Once I look back at the insane amount of notes, I stress myself out again. Never ending cycle. :D (Because usually I write things down as to-dos, potential to-dos, things I have to think about etc.)
- I use Syncthing (which is a P2P open source sync app) to sync a folder called "Life" btw my phone and my PC. - That folder is full of markdown files, which I open with Material Files and edit with Markor on my phone. - Whenever I make a change, its automatically synced to my PC's "Life" folder over the network thanks to Syncthing. This is basically the best way to do it because: - You dont need a paid subscription to a cloud service. - You dont need to have a home server always running. - You dont need any specific program to edit the notes (All OS have preinstalled text editors for plaintext files) - You can take advantage of search features, of a fully fledged file manager to sort/rename/organize your "notes" (text files) - You can easily use scripts for automation, as scripts are also plaintext files that take plaintext files as input and produce plaintext files as output.
Flow state exists, it has diminishing returns. Sometimes you need to just walk away, or go to sleep. How about you pick a task and work on it until you're not being productive any more? Was getting my son ready for his piano recital. We were crunching away at it. Wasn't perfect and he had stopped making progress so we stopped after a bit. Next morning he absolutely nailed it, got through it without any hesitations, wrong notes, etc. A good night sleep with unfinished work is great for me as well as a sr software engineer. Thinking about these problems and various ways to attack them at night helps me sleep too... it's a win win. Then in the morning I'm motivated to get to work and try stuff out.
the brain is like any other muscle, it gets tired and needs rest. Sleep is vastly underappreciated. I play guitar, and I find after an hour or even 45 minutes of practice I'm cooked, there's no more progress to be made on the learning front. Best thing to do at that point is wait a few hours or until the next day. I believe when you practice, you really don't see the skill gains until the next day at least. High quality practice trumps long hours
Learned this the hard way on an assignment as a freshman that I put off because I underestimated the difficulty (surprise, surprise) and I didn’t get it entirely working before the submission time, so I submitted it incomplete because the professor wouldn’t take anything late. Went to bed, woke up the next day and the solution hit me like a brick, and I implemented it and it worked, and I learned to start early on assignments just to give myself the chance to sleep on a problem. I thought that shit was a cliche my whole life until I lived it lol.
Taking a break is extremely effective. When I find myself bashing my head against a problem too long, I go do something physical (go for a walk; play with the cat; martial arts training) and so often the solution just pops into my head.
Your level of gratification is directly correlated to the level of difficulty. Therefore, I agree, being mature pushes pass the enjoyability phase increasing difficulty and therefore gratification. Otherwise, you have a recipe for difficulty suppression and low gratification. I also feel that the more time spent immersed in high levels of difficulty the longer you need to break from those efforts.
holy crap when you started explaining "the trough of sadness" and how it'll last a long time after breaking your flow... I came to the realization that I'm only here watching your videos because I'm currently in that pit 🤣🤣🤣
15:00 - Done this researching Marching Cubes vs Dual Contouring for hours passed when I was feeling burnt out. Retained nothing and had to go back over everything a few days later.
Basically you make up a scale yourself and the user reports their perceived effort/enjoyability for a particular task. He goes over it around 11:50 Its a bit fishy and not truely accurate, but for the purposes of this a scale of this it is sufficient for the analysis.
14:55 Prime, I believe you when you say you can study math for 16 hours. But that doesn't mean there isn't a point of diminishing return. Some of us grew up in households that didn't believe in diminishing returns, and that's why we know it existzs. When you're stuck for 5 hours in the kitchen with your father, you can't leave until you master X thing, and the more time passes the less you're able to focus, the more mistakes you make, even basic ones you didn't make at the start, and the less progress you're making, then you feel the diminishing returns.
I think the point is more that you're able to increase your time you can spend being productive by practicing and cultivating your own motivation. I highly doubt Prime was able to go for 16 hours on day one. You build up to that. With that said, putting a child in a position that you were in is not a way to accomplish that. If the only motivation you have to do something is the threat of unbalanced consequences... That's not going to get you in a flow state. It gets you in a state of fear. Of course you're not going to think clearly. Of course you're going to make mistakes. That's not an effective way to "help" someone learn. I'd argue that has more do deal with the diminishing returns than the length of time you spent doing the task. I'm sorry you and others had to go through that.
@@davidlamar4816 I definitely think the point of diminishing return varies. My argument was more than it exists. Thanks for the compassion, but honestly, I was lucky enough to be talented, so my "horror period" ended after primary school. I sometimes think about all the kids who didn't have this chance. The stereotypical asian household is not just a cliché, I knew some, and I feel for them. They are zre under immense pressure.
I am not a fan of formulas for human psychology due the complexity of human mind. Motivation and enjoyment for instance is deeply tied to personal meaning which can be pretty fluid and depend on many factors and even the awereness level about reality of the actor, so it can drastically change even if all the variables expressed in the formula remain the same. For instance changing how you feel towards your boss or a change in religion can alter one's motivation towards the same task. Using formulas for human minds implies that we are deterministic machines that it can control itself mechanically
They are just models that can help to better understand some of the mechanics that are involved. It doesn't have to be perfect to be able to learn something interesting from them. That said, the author pulled these models straight out of his ass so I'd take any conclusions with a big grain of salt.
ProTip from decades of IT that will put you in good stead - always leave a simple task aside for the next day morning to start with. That way you can get up to speed and into your flow for the harder tasks. Look at it as a mental coffee.
I always write all my code, but do no testing. The next day I start with the testing. Two reasons: I'm feeling fresh / awake. I've stepped away from the code so hopefully won't do as much "confirmation bias" testing.
i had this on a playlist and forgot what the title was by the time it came on while i was playing some games. i keep hearing stuff about phi and rho and derivatives and mathematical stuff, sounds pretty interesting. i look over and see its about a fucking todo app
The integral is fairly easy u can do integration by parts on the e bit. Basically y= te^t y'=e^t + te^t So all u really need is some linear combination of e^t and te^t
15:00 I think Prime might be the one with the "condition." Normal people don't look at math for more than a couple of hours before things stop making sense.
Another fundamental flaw here is what if you don't finish the task in the given time? Sometimes, tasks are low effort, but then something comes up and makes it higher effort than expected. Sometimes the total time is just not enough time to get every task done (classic overcommitting problem). And can you add in hours of sleep as a variable to this equation?
the person writing this increased their productivity by 0% because he still hasn't done that math homework despite obviously understanding the subject.
Context switching is useful. Another thing the algorithm didn't take into account that not all tasks have the same fundamental value. Doing a bit more of a "less productive" task could be more valuable than others are "more productive" when "effort" and "enjoyability" are what's used...
14:40 while sometimes you do have to suck it up and keep going, it's vastly better to learn something in small bursts of high quality studying. When you're learning something your brain is literally reorganizing itself, and I'm pretty sure that process takes at least a day to complete. So especially if you're learning something with compounding information, you'd be better off breaking it up into smaller study sessions. After a certain point you're being very inefficient with your time
I'd say that there is 1 layer that is missing here. We had a point at which we already had the optimal time for a given task, but then we gave that up in order to be able to do all tasks in a single day, which shortens or lengthens all optimal times accordingly. However if we have let's say 2 tasks, of which the first has the optimal time of 3 hours, and the second having an optimal time of 2.5 hours, and we have 3 hours to spend on tasks per day, then as long as the tasks don't have an "impossible" deadline of tomorrow, (where we can't finish it, and just try to get maximum amount of unfinished work,) it's way more optimal to do only one task a day (day1: task1, then day2: task2, then day3: task1...) instead of doing both tasks on all days. So in my opinion the missing layer is skipping tasks in a way that makes all tasks get closer to their optimal time (adjusted for any deadlines that would not be possible with optimal skipping). Although there is another issue with this. It is easy to show this being better in the math, but skipping tasks for 1-2 days breaks the ability to continue them properly next time. Although this is also true for the original concept, just in a different way. There we have more breaks, but less spacing, resulting in more, but milder startup issues, while skipping causes less, but worse startup issues. Also, this is different from the startup cost in the original model, because that only considered mental state, but if I've been doing something for the last 3 hours, then trying 1 more thing in the same topic may take 15 minutes, but if I break it off now, then it may take 30 minutes tomorrow. Not because of the flow state or whatnot, but because I have to remember where I was, I have to open everything again, or get stuff ready for it (really depends on the topic). A very dumb extreme could be that if your PC needs 10 minutes to get ready (let's assume you can't do other stuff in the meantime for the argument's sake), then allocating 10 minutes to a task that needs the PC makes no sense, and this distorts longer times as well
25:18 YES!! Multitasking of intelectual work is Sh!t!! when you need to change tasks, you have a lot of mental effort to "close the boxes" and leave it in a way you can remeber where to restart. and start "open" and "unboxing" all the points for the next task. you need to remeber, the correct version, connections, updates, the planned logical workflow, the sources, open Jira, report task start, and finally start working to SOLVE the task. and this is considering that you dont have to overcome all the struggle to start something you dont want to do
My approach was always just to sit in my chair and and not move until the task was complete it used to work sadly it now seems to just lead to being awake for days achieving nothing though so I might have to reassess at some point haha
The challenges and tasks I face on a daily basis are actually curses that I need to destroy using cursed energy (attention). However I have not learned the reverse curse technique and therefore I suck at exterminating curses.
At school they taught us this pattern: \_/\_/ Explanation: Initial motivation; decreasing productivity as you get bored/tired; increasing productivity as the end comes into sight This probably applies more to less complex and more compulsory, disliked tasks such as studying. Now thinking about it, the flowstate model makes a lot more sense than what we were taught.
The article writer made reasonable assumptions about the model that was generated. There were obviously certain trade-offs that were made (such as the possibility of second winds, variability of attention spans in different people, etc.) in order to actually come up with a model that was simple enough to calculate but general enough to encapsulate the idea that "as time goes on, productivity decreases from the point of maximum productivity". This isn't about using experiments to verify a hypothesis, its an exercise in applying mathematics to map a widely understood heuristic to an actual number.
Diminishing returns doesnt mean spending more time doesnt net you more value, it just means spending more time now nets you less value than spending more time before. You still get better, just not as fast as you had the previous 6 hours. I feel like people don't understand what a derivative actually is when they say you should only spend 3 hours studying g for math.
Did he explain how he tunes the model's per-person parameters? I feel like I was supposed to have seen that at some point. And how does he validate his model empirically? That quick "sanity check" at the end does not really work as validation since it is just another model assumption.
I actually thought the post was gonna end with “ And it doesnt matter how much calculus you do if you keep postponing the task trying to improve you performance like you just did with this post”
To be fair, the dumbest thing is that this considers all tasks to be infinity. "Ah yes, let me spend 1,5h in the math homework even if it was finished in 30min"
There's the cost of switching, but there's also the gain of switching. Sometimes I switch tasks and I am so much more productive than usual on the new task because of how much the previous one dragged out.
i think that's the point of ThePrime when he tells us to stick with a task even if we're not enjoying it as much as before, so when switching to a new task it will feel amazing
That's why the method of each 30 mins a 5 min break is very applicable and useful, i cannot jump like that from task to another i would include in the 6 hours plan breaks
I tried it. Bro gave me 4 minutes to clean my house (which I set to priority 10), and 1h 15m for songwriting (which I gave priority 5). I think there's a naming problem because "effort" seems to be used as if it means the duration of the task. I think of cleaning as very low effort and songwriting as high effort because I can clean mindlessly while listening to a book, and songwriting is a labour of love.
I think that effort is supposed to determine how much time you have to spend on a task to achieve the desired results. So you effectively told the system that you don't need much time to complete the "clean your house" task. Also, based on the demo video, it seems that priority only affects the order of the tasks, not the time. So in essence, I'd say that what you call "low effort" the system calls "high enjoyability" (you can spend a lot of time on this task without getting diminishing returns) and what you call "high priority", the system calls "high effort" (you'll need to spend a lot of time to complete this task). I mean, I still think that the system isn't perfect, but your particular problem seems to be coming from user error rather than a flaw in the model itself.
The biggest problem with this model is that it assumes that the user's perceived enjoyability and effort for any particular task is fixed for the duration of the task. In other words it considers its users as robots. This has the side effect of predicting a phi (time it takes to get to peak productivity) that may be drastically over/under estimated. In addition to not accounting for a host of other variables like sleep, exercise, nutrition, current emotional state and attention span (which is understandable because it would be much more difficult to construct a model that accounts for those things), it will probably produce an output that is not optimal, and im not convinced that it would be more effective than just "eye balling" it, or brute forcing the tasks, or even allocating equal amount of time to tasks that have perceived equal effort. If the model took more input from the user, and if it even accounted for the users history of task completion, then I might consider using it.
My calculations indicate that the time a task takes is measured from when it begins to when it ends. The productivity is a boolean on whether it's done or not.
so, I paused for a moment there at exactly 13:28... and after hurting my own brain, I happened to look at the comment stream in the top right corner. Cohors: "what did I just tune into?" roo00t: "My eyes hurt" seck_loogie: "wait. but how many deep is the pit and how much can the horse pull." lmfao.... >.< that must have been one entertaining stream.
This is the thing about Prime that keeps amazing me. When, at the end of a perfectly decent article, he goes "I have a counter take". … and I am like "hey that was a perfectly decent article, and it sounds like you are coming from a very limited sample size with your take". … and when I let him cook, I have nothing else to say but "hey that made 110% of the sense and you are correct, I didn't think of any of that".
15:40 Maybe I'm programming for 5 hours and don't get anything done. And then I get an idea and am 10 hours productive and don't even want to eat. And then I want to sleep, but I just want to finish it, I'm almost done anyway. If I tried to sleep now, I couldn't anyway and would just watch stupid stuff, and still be tired the next day. And the next day, all my motivation would be gone. Now I have the whole day to finish it, and it's almost finished anyway. I'll first have a long breakfast, then I watch a few episodes of my current anime, then I slowly look into my project and realize I have no idea where to start, so I eat lunch and then go outside. And at 9 in the evening I think, I should just finish it, but since it's not finished yet at midnight and I have work tomorrow, I go to bed and haven't finished it yet. If I just finished it yesterday, it would have taken another 5 hours, but at least it would have been finished. Something like this isn't considered in this algorithm, right?
The tasks are analysed as if they were all independent and static but if you do the first you have more energy then when you start the second or third after 3 or 4 hours. It will take you longer to get into it and you will not reach the same full concentration as with the first one in the morning. You get exhausted eventually over the day working at 100% full concentration. I think you have to do much more math here. Wink. I on the other hand might eat the frog first, i.e. do the thing I like least and feel better after that is off the table or just do what I feel most enjoyable first depending on how I feel on that day.
The entire concept is flawed - instead of estimating roughly how long you think a task will take, or basing it on past experience when you'll start to crash, you are basically just guessing at a bunch of random constants for each and every task that you have defined arbitrary relationships between. It also doesn't hold for tasks where you have an explicit workload to complete. You can't just decide "I'm not going to be optimally productive if I don't finish this maths homework, so I simply won't finish it" - the model only holds under the made up rule set the author has defined. This same critique also holds for literally all quantitative finance - it only works in made up fantasy land
Oh yeah, its dumb. Everyone has different stale-task tolerance, short-task preference, post-task confidence, pre-task procrastination/anxiety, task-switch friction (stubbornness). Some days you just want to crunch a hard problem, some days you need to clear the minor tasks so you can focus. None of this can be enumerated, let alone calculated. Imagine making your own flawless personal formulas, but then you get irritated the by the rigidity and authority your own creation has over you, or lack the discipline to actually use the dang ol thing. There's being overly analytical, and there's trying to optimize humanity out of human life. Just wing it, or you'll never fly.
to be fair to the author, they went on to extend a feature inside the app where anyone can define a task as time-limited and just override the algorithm.
Man this sounds like an article I would absolutely write to procrastinate doing homework lol. (physics student here.) And it's so nice that I can understand almost everything they say (almost), and follow along with the math. And no, I am not procrastinating doing homework rn. I'll have you know I am watching this after finishing my homework!
I think this approach to maximizing flow state is fine, but he should have considered the boost in energy you get from finishing a task, the difference between tasks with a clear end in sight (like homework) with no follow-through, versus tasks with no clear end (like studying) with follow-through (having to remember things for exam), and also the difference in nature of tasks (video editing is not as mentally involved and more free-form as opposed to studying/homework) For his example, I would have started with the homework, since it's mentally involved and not free form, so its uncomfortable to keep in the backlog. Its good to start with, since there's no follow-through. I would ensure the studying for physics exam comes last as recency helps with remembering things (part of the follow-through which is the exam) and no clear end (you're never really done studying until the exam's over tbh) next I'd go for the video editing task, since it's more free-form it'd act as a sort of break, while also being an atomic task (with no follow-through) Basically, a more qualitative analysis is required when scheduling stuff optimally.
That kind of mathematical modelling is why all software sucks. Not only is the model wrong, it also reduces people to a few shallow dimensions. It's the opposite of sonder and a bad way to think about people.
![Lp. Сердце Вселенной #60 РОЖДЕНИЕ ЛОЛОЛОШКИ [Финал] • Майнкрафт](http://i.ytimg.com/vi/YoR0pAV9FVQ/mqdefault.jpg)
This guy totally wrote a medium article to procrastinate on his homework.
A classic “congratulations or I’m sorry that happened to you” moment
a cum man
unironically probably. all this overthinking of how best to prioritize things ignores the fact that the problem is almost never that a person doesn't distribute their time working well, it's that they don't spend enough time working, or they spend too much time switching tasks because they keep changing their mind about which task is most important (which itself is typically a method of avoiding work, rather than an actual issue of prioritization). If you have 3 things to do, we can assume without loss of generality the optimal order to be 1 2 3. But even completing the tasks in order 3 2 1 will be fine in 99% of cases, even if the work would have been completed 20% quicker if done in the optimal order. 20% more time seems like a lot of time until you realize that you might spend 50% or even 100% or more of the time it would take to complete the tasks on stressful procrastination/avoidance, including time spent 'prioritizing'.
That being said, gives that the majority of this wasted time is psychological, I've found that the 'optimal' way of actually working is to pick the simplest, easiest tasks to start with, even if they are low priority, and knock a few of them out. This creates quick wins that create momentum and will also unburden the mind of the psychological stress of having a long todo list (which distinct from having a todo list that requires a long time to complete). With that momentum, I then jump into the most psychologically distressful talk, ie the one I am most inclined to avoid (which is also typically the one that most needs to be done) and complete as much of it as I can. At this point I generally am in work mode and can go for 8-10 hours straight with few breaks.
I am happy you pay for medium. Because that means I get free medium🙌🏽🤝🏽
It's only a single search away... x)
You might say you have a Prime subscription to medium.
I pay for UA-cam so that I don’t get Ads while watching a man read a premium article. #FiscalResponsibility
Choosing your own articles to read is a good experience. Just saying.
@@edism medium is shit because anything I try to read, it hits me up with a sign up. Never ever read anything from medium. I’d rather read from anywhere else. Prime making videos like this means me indirectly being able to read medium articles
You can tell they’re a student not just because they remember all of this math, but also because they naively assume that the outputs from all tasks have equivalent value.
14:40 "Sometimes you just have to be more mature than how you feel" is honestly really good advice
23:23 Can't be a coincidence that the "trough of sadness" has the same initials as TypeScript.
Twats on Sugar
I love how this man went and decided that the curve should look like that without any real world measurements and built a whole thesis around it instead of using that part of his brain to estimate the time it will take to learn Physics.
Right. I don't care to pay for medium and I didn't pay much attention to anything not mentioned in the clip, so I'm just going to assume that none of this guy's assertions were attained from or sourced to experts in relevant fields i.e. psychology/biology instead of math... peak solipsism IMO. We don't solve all of our problems with philosophy anymore. Nothing is taken as a given when it doesn't have to be.
Well, the Poisson distribution is a statistical distribution that models number of occurrences of some event in a specific time. In his model, I noticed what looks like an exponential distribution (the part with e) which is used to model the time it takes for an event to occur. These distributions are well established. If you are modelling anything similar, that's the best starting point.
But I think the author could have explained more clearly. There seemed to be too many asspulls in the article
@@NphiniTYea that's why Prime went straight to drawing a different curve. There was no non-mathematical explanation for the values inserted. There are tons of amazing statistical models with strong accuracy at high sample sizes... but none of them work if your underlying facts are incorrect. This happens regularly, a good example being political polls. Asking the wrong questions or in the wrong way gives you fantastic data on something that isn't your intended question. The same thing will happen here. As the questions being asked by the formulas are not about the multitude of psychological and physiological states of the user, those mechanisms which determine the vast majority of your decisions will go unaccounted. Enjoyment and Effort are far too simple. I would be interested to find out if other disciplines were consulted and somehow accounted for, it just doesn't seem likely.
@@Varadiio If that is so then I'd appreciate if you linked the source that the person used. I simply wish to confirm that the time/effort graph isn't completely made up.
I have not said or implied that philosopy is relevant, so I'll just assume you are projecting since pepople not involved with philosophy wouldn't use "solipsism" to describe a self centered person.
I assumed there were no citations because I didn't see any, which is, in my opinion, completely fine. Besides I wasn't arguing wether or not the graph is accurate, I simply pointed out that they can clearly utilise assumptions and still refused to use them when they are probably the optimal way forward.
@@NphiniT Thank you for clarifying where that graph came from. What really bothered me is that there was no reasoning provided for using that graph. I can get behind it being accurate to their experience, or any number of reasons, but I can't get behind needing a reason to study for a certain amout of time when we apparently don't need any for half the things we based a thesis off of.
"dont do morning routine"
Proceed to tell us his morning routine imediatly after :
"Wake up, coffee and go"
is common error for to be mistaken to differentiate micro task, routine and subroutine
@@nrwchd This is interesting man. Can you explain a bit more? Just curious how you differentiate
@@chrispy_baeconsure, routine is the code that lives in the main function.
subroutine is all function calls except for main (because you can always recursively calls main)
micro task is deferred code that executed later (usually end of routine or subroutine) like destructor in C++ or defer block in Zig
@@nrwchd routine is something you do repeatedly usually in the same time of the day voluntarily. subroutine is just a part of that whole routine and a micro task is something small you have to get done, but it is not usually the same thing or at the same time.
Is how I would describe these terms
@@MorgurEditswe're talking about programming right?
My productivity hack for reaching flow state quickly is to ensure that you have an important reason to meet a deadline, then start the task with the exact amount of time needed to complete it left before the deadline. If you don’t have time to not be at peak performance, you will be at peak performance. (You will also quickly learn that flow state can last 12 plus hours, it’s the best feeling, better than heroin)
underrated comment
So you've experienced heroin, I see.
Does it really work as a cough medicine for children?
Not a meme, have you ever considered you may have ADHD?
Source: have ADHD. Cannot focus on a task unless under absolute pressure.
@@jackdixon6681 I was diagnosed with ADD, but in my country there are no options for proper treatment. But I just like to consider myself lazy...
I have 3 todo "apps":
- On my phone, note app
- On my computer, sticky notes for permanent and notepad for temporary notes
- On real life, a blank sheet of paper inside a plastic case which lets me use a whiteboard marker on it. It's like a ultra thin portable whiteboard and it's amazing. I learnt this trick preparing for an online exam once
My brain sticks them together
I am in a hate-love relationship with physical notes at this point. I prefer them over digital note taking, but once I start noting things down, my desk is quickly covered in post-its and full sized paper. Once I look back at the insane amount of notes, I stress myself out again. Never ending cycle. :D (Because usually I write things down as to-dos, potential to-dos, things I have to think about etc.)
- I use Syncthing (which is a P2P open source sync app) to sync a folder called "Life" btw my phone and my PC.
- That folder is full of markdown files, which I open with Material Files and edit with Markor on my phone.
- Whenever I make a change, its automatically synced to my PC's "Life" folder over the network thanks to Syncthing.
This is basically the best way to do it because:
- You dont need a paid subscription to a cloud service.
- You dont need to have a home server always running.
- You dont need any specific program to edit the notes (All OS have preinstalled text editors for plaintext files)
- You can take advantage of search features, of a fully fledged file manager to sort/rename/organize your "notes" (text files)
- You can easily use scripts for automation, as scripts are also plaintext files that take plaintext files as input and produce plaintext files as output.
@@lukasz96 just write them down on index cards individually. then you can stack them up into a pile and it looks like there's just one.
@@y00t00b3r genius
my todo apps are a wall calendar and a pocket notebook
Flow state exists, it has diminishing returns. Sometimes you need to just walk away, or go to sleep. How about you pick a task and work on it until you're not being productive any more?
Was getting my son ready for his piano recital. We were crunching away at it. Wasn't perfect and he had stopped making progress so we stopped after a bit. Next morning he absolutely nailed it, got through it without any hesitations, wrong notes, etc.
A good night sleep with unfinished work is great for me as well as a sr software engineer. Thinking about these problems and various ways to attack them at night helps me sleep too... it's a win win. Then in the morning I'm motivated to get to work and try stuff out.
the brain is like any other muscle, it gets tired and needs rest. Sleep is vastly underappreciated. I play guitar, and I find after an hour or even 45 minutes of practice I'm cooked, there's no more progress to be made on the learning front. Best thing to do at that point is wait a few hours or until the next day. I believe when you practice, you really don't see the skill gains until the next day at least. High quality practice trumps long hours
Learned this the hard way on an assignment as a freshman that I put off because I underestimated the difficulty (surprise, surprise) and I didn’t get it entirely working before the submission time, so I submitted it incomplete because the professor wouldn’t take anything late. Went to bed, woke up the next day and the solution hit me like a brick, and I implemented it and it worked, and I learned to start early on assignments just to give myself the chance to sleep on a problem. I thought that shit was a cliche my whole life until I lived it lol.
@@smocloud Oh wow you just opened my eyes on this, completely agree
Sleep builds muscle memory
Taking a break is extremely effective. When I find myself bashing my head against a problem too long, I go do something physical (go for a walk; play with the cat; martial arts training) and so often the solution just pops into my head.
The problem with TODO apps is that they will end up containing all your shame.
This had more calculus than I expected.
The man literally explains why when I know I’ll have to stop doing something soon I don’t bother starting LMAO!
Your level of gratification is directly correlated to the level of difficulty. Therefore, I agree, being mature pushes pass the enjoyability phase increasing difficulty and therefore gratification. Otherwise, you have a recipe for difficulty suppression and low gratification. I also feel that the more time spent immersed in high levels of difficulty the longer you need to break from those efforts.
The scheduling advice (do the most important thing first, do one thing at a time) is worth more than the entire article
Unironically true
Why though? I thought it was true?
@@thomasdinh2k I think this indeed is what they mean.
3:55 SOMEBODY call this man’s wifeee!!! he’s on the pills againn!!
i thought of this as a joke and just realized how bad this may actually sound… its not letting me delete soo yeaaahhh
holy crap when you started explaining "the trough of sadness" and how it'll last a long time after breaking your flow... I came to the realization that I'm only here watching your videos because I'm currently in that pit 🤣🤣🤣
Ohh no
15:00 - Done this researching Marching Cubes vs Dual Contouring for hours passed when I was feeling burnt out. Retained nothing and had to go back over everything a few days later.
how the hell are you supposed to accurately approximate a tasks enjoyability and difficulty??
Basically you make up a scale yourself and the user reports their perceived effort/enjoyability for a particular task. He goes over it around 11:50
Its a bit fishy and not truely accurate, but for the purposes of this a scale of this it is sufficient for the analysis.
this feels like that one time when i made a whole rocket simulation program for my hs physics class instead of just showing my work, but in reverse...
14:55 Prime, I believe you when you say you can study math for 16 hours. But that doesn't mean there isn't a point of diminishing return. Some of us grew up in households that didn't believe in diminishing returns, and that's why we know it existzs. When you're stuck for 5 hours in the kitchen with your father, you can't leave until you master X thing, and the more time passes the less you're able to focus, the more mistakes you make, even basic ones you didn't make at the start, and the less progress you're making, then you feel the diminishing returns.
I think the point is more that you're able to increase your time you can spend being productive by practicing and cultivating your own motivation. I highly doubt Prime was able to go for 16 hours on day one. You build up to that.
With that said, putting a child in a position that you were in is not a way to accomplish that. If the only motivation you have to do something is the threat of unbalanced consequences... That's not going to get you in a flow state. It gets you in a state of fear. Of course you're not going to think clearly. Of course you're going to make mistakes. That's not an effective way to "help" someone learn. I'd argue that has more do deal with the diminishing returns than the length of time you spent doing the task. I'm sorry you and others had to go through that.
@@davidlamar4816 I definitely think the point of diminishing return varies. My argument was more than it exists.
Thanks for the compassion, but honestly, I was lucky enough to be talented, so my "horror period" ended after primary school. I sometimes think about all the kids who didn't have this chance. The stereotypical asian household is not just a cliché, I knew some, and I feel for them. They are zre under immense pressure.
I am not a fan of formulas for human psychology due the complexity of human mind. Motivation and enjoyment for instance is deeply tied to personal meaning which can be pretty fluid and depend on many factors and even the awereness level about reality of the actor, so it can drastically change even if all the variables expressed in the formula remain the same. For instance changing how you feel towards your boss or a change in religion can alter one's motivation towards the same task. Using formulas for human minds implies that we are deterministic machines that it can control itself mechanically
They are just models that can help to better understand some of the mechanics that are involved. It doesn't have to be perfect to be able to learn something interesting from them. That said, the author pulled these models straight out of his ass so I'd take any conclusions with a big grain of salt.
I want to disagree with you because i like ML engineering but, at the same time, i know you’re not wrong
I appreciate the random outbursts of energy, its what keeps me coming back for informative entertainment
I usually like to procrastinate well until past the dead line then it somehow instantly becomes more obvious what to do first.
ProTip from decades of IT that will put you in good stead - always leave a simple task aside for the next day morning to start with. That way you can get up to speed and into your flow for the harder tasks. Look at it as a mental coffee.
I always write all my code, but do no testing. The next day I start with the testing. Two reasons: I'm feeling fresh / awake. I've stepped away from the code so hopefully won't do as much "confirmation bias" testing.
i had this on a playlist and forgot what the title was by the time it came on while i was playing some games. i keep hearing stuff about phi and rho and derivatives and mathematical stuff, sounds pretty interesting. i look over and see its about a fucking todo app
😂🤣
The integral is fairly easy u can do integration by parts on the e bit.
Basically y= te^t
y'=e^t + te^t
So all u really need is some linear combination of e^t and te^t
15:00 I think Prime might be the one with the "condition." Normal people don't look at math for more than a couple of hours before things stop making sense.
He is peak condition
Adhd brain
6:55 bro says he used to study math 16h straight and forgot how to find the maximum of a function ☠
All of AI is finding the minimum/maximum of a function, arguably all of problem solving is finding the minimum/maximum of a function.
@@jonathan2847functional programming???
@@jonathan2847 True though. Maximum Likelihood estimation and Least Squares estimation are still haunting me to this day 😅
Another fundamental flaw here is what if you don't finish the task in the given time? Sometimes, tasks are low effort, but then something comes up and makes it higher effort than expected. Sometimes the total time is just not enough time to get every task done (classic overcommitting problem). And can you add in hours of sleep as a variable to this equation?
the person writing this increased their productivity by 0% because he still hasn't done that math homework despite obviously understanding the subject.
Context switching is useful.
Another thing the algorithm didn't take into account that not all tasks have the same fundamental value.
Doing a bit more of a "less productive" task could be more valuable than others are "more productive" when "effort" and "enjoyability" are what's used...
I am very happy 😂 keep going! You are doing amazing job! I loved the JDSL article. But seriously! I like your content, very insightful
The guy really paid for medium?? Insane
I did it for you
A noble sacrifice
14:40 while sometimes you do have to suck it up and keep going, it's vastly better to learn something in small bursts of high quality studying. When you're learning something your brain is literally reorganizing itself, and I'm pretty sure that process takes at least a day to complete. So especially if you're learning something with compounding information, you'd be better off breaking it up into smaller study sessions. After a certain point you're being very inefficient with your time
26:15, you could argue the trough + dopamine charge roughly cancel out in the long term, and the graph is on average reasonable
25:00 cool theory Prime, very nice. Can't wait for your Master Thesis on it!
You're absolutely right. Flow is a huge factor when doing good productive work. This has been studies very well and it's a huge cost to productivity.
The winrar comment on the chat passed unattended 🤣 "Do you also pay for Winrar?"
I'd say that there is 1 layer that is missing here.
We had a point at which we already had the optimal time for a given task, but then we gave that up in order to be able to do all tasks in a single day, which shortens or lengthens all optimal times accordingly.
However if we have let's say 2 tasks, of which the first has the optimal time of 3 hours, and the second having an optimal time of 2.5 hours, and we have 3 hours to spend on tasks per day, then as long as the tasks don't have an "impossible" deadline of tomorrow, (where we can't finish it, and just try to get maximum amount of unfinished work,) it's way more optimal to do only one task a day (day1: task1, then day2: task2, then day3: task1...) instead of doing both tasks on all days.
So in my opinion the missing layer is skipping tasks in a way that makes all tasks get closer to their optimal time (adjusted for any deadlines that would not be possible with optimal skipping).
Although there is another issue with this. It is easy to show this being better in the math, but skipping tasks for 1-2 days breaks the ability to continue them properly next time. Although this is also true for the original concept, just in a different way. There we have more breaks, but less spacing, resulting in more, but milder startup issues, while skipping causes less, but worse startup issues. Also, this is different from the startup cost in the original model, because that only considered mental state, but if I've been doing something for the last 3 hours, then trying 1 more thing in the same topic may take 15 minutes, but if I break it off now, then it may take 30 minutes tomorrow. Not because of the flow state or whatnot, but because I have to remember where I was, I have to open everything again, or get stuff ready for it (really depends on the topic).
A very dumb extreme could be that if your PC needs 10 minutes to get ready (let's assume you can't do other stuff in the meantime for the argument's sake), then allocating 10 minutes to a task that needs the PC makes no sense, and this distorts longer times as well
25:18 YES!! Multitasking of intelectual work is Sh!t!! when you need to change tasks, you have a lot of mental effort to "close the boxes" and leave it in a way you can remeber where to restart. and start "open" and "unboxing" all the points for the next task. you need to remeber, the correct version, connections, updates, the planned logical workflow, the sources, open Jira, report task start, and finally start working to SOLVE the task. and this is considering that you dont have to overcome all the struggle to start something you dont want to do
I like how most people are trying to figure out the best way to get to and maintain a flow state and this guy is like get there and then move on.
My approach was always just to sit in my chair and and not move until the task was complete it used to work sadly it now seems to just lead to being awake for days achieving nothing though so I might have to reassess at some point haha
17:00 you are not alone prime we all got scared from that integration 🤣🤣🤣
This guy clearly went above and beyond for his essay... either that or his professor demanded way too much of him.
The challenges and tasks I face on a daily basis are actually curses that I need to destroy using cursed energy (attention). However I have not learned the reverse curse technique and therefore I suck at exterminating curses.
At school they taught us this pattern: \_/\_/
Explanation: Initial motivation; decreasing productivity as you get bored/tired; increasing productivity as the end comes into sight
This probably applies more to less complex and more compulsory, disliked tasks such as studying.
Now thinking about it, the flowstate model makes a lot more sense than what we were taught.
the dangers of over thinking and not using experiments to verify your hypothesis
The article writer made reasonable assumptions about the model that was generated. There were obviously certain trade-offs that were made (such as the possibility of second winds, variability of attention spans in different people, etc.) in order to actually come up with a model that was simple enough to calculate but general enough to encapsulate the idea that "as time goes on, productivity decreases from the point of maximum productivity". This isn't about using experiments to verify a hypothesis, its an exercise in applying mathematics to map a widely understood heuristic to an actual number.
You're right, dude. Prime is a funny dude. Especially these "use vim" jokes are awesome!
my god prime giving actual feedback on this math hell, gg honestly
5:21 - Oiler's number - the current price of barrel of oil on the commodities market
All intelligence, no wisdom. Solving equations for the sake of solving equations. I am appalled.
Imagine spending your productivity time coming up with equations to make a productivity app to be more productive.
I close my eyes and hear Micheal from office reading a medium article
Diminishing returns doesnt mean spending more time doesnt net you more value, it just means spending more time now nets you less value than spending more time before. You still get better, just not as fast as you had the previous 6 hours. I feel like people don't understand what a derivative actually is when they say you should only spend 3 hours studying g for math.
Did he explain how he tunes the model's per-person parameters? I feel like I was supposed to have seen that at some point. And how does he validate his model empirically? That quick "sanity check" at the end does not really work as validation since it is just another model assumption.
I actually thought the post was gonna end with “ And it doesnt matter how much calculus you do if you keep postponing the task trying to improve you performance like you just did with this post”
Whenever I hot flow state my boss reassigns me to another task, leaving me in shambles.
I have to say what prime said is exactly how I feel, heck I wrote the same thing in my journals, he's speaking from experience.
To be fair, the dumbest thing is that this considers all tasks to be infinity. "Ah yes, let me spend 1,5h in the math homework even if it was finished in 30min"
The guy who wrote this articule is trying so hard to look nerd
There's the cost of switching, but there's also the gain of switching. Sometimes I switch tasks and I am so much more productive than usual on the new task because of how much the previous one dragged out.
i think that's the point of ThePrime when he tells us to stick with a task even if we're not enjoying it as much as before, so when switching to a new task it will feel amazing
@@alionicle I've never seen someone call him "ThePrime", that's crazy, I just unlocked a core memory
That's why the method of each 30 mins a 5 min break is very applicable and useful, i cannot jump like that from task to another i would include in the 6 hours plan breaks
I tried it. Bro gave me 4 minutes to clean my house (which I set to priority 10), and 1h 15m for songwriting (which I gave priority 5). I think there's a naming problem because "effort" seems to be used as if it means the duration of the task. I think of cleaning as very low effort and songwriting as high effort because I can clean mindlessly while listening to a book, and songwriting is a labour of love.
I think that effort is supposed to determine how much time you have to spend on a task to achieve the desired results. So you effectively told the system that you don't need much time to complete the "clean your house" task.
Also, based on the demo video, it seems that priority only affects the order of the tasks, not the time.
So in essence, I'd say that what you call "low effort" the system calls "high enjoyability" (you can spend a lot of time on this task without getting diminishing returns) and what you call "high priority", the system calls "high effort" (you'll need to spend a lot of time to complete this task).
I mean, I still think that the system isn't perfect, but your particular problem seems to be coming from user error rather than a flaw in the model itself.
Honestly I guess the Prime addition is actually included since even the push is hard and the down is deeper the sum is equal to the original line.
The biggest problem with this model is that it assumes that the user's perceived enjoyability and effort for any particular task is fixed for the duration of the task. In other words it considers its users as robots. This has the side effect of predicting a phi (time it takes to get to peak productivity) that may be drastically over/under estimated. In addition to not accounting for a host of other variables like sleep, exercise, nutrition, current emotional state and attention span (which is understandable because it would be much more difficult to construct a model that accounts for those things), it will probably produce an output that is not optimal, and im not convinced that it would be more effective than just "eye balling" it, or brute forcing the tasks, or even allocating equal amount of time to tasks that have perceived equal effort. If the model took more input from the user, and if it even accounted for the users history of task completion, then I might consider using it.
My calculations indicate that the time a task takes is measured from when it begins to when it ends. The productivity is a boolean on whether it's done or not.
2:40 If you want to find the name of a character, just paste it into a search engine and it will tell you.
(It's Rho btw)
so, I paused for a moment there at exactly 13:28... and after hurting my own brain, I happened to look at the comment stream in the top right corner.
Cohors: "what did I just tune into?"
roo00t: "My eyes hurt"
seck_loogie: "wait. but how many deep is the pit and how much can the horse pull."
lmfao.... >.< that must have been one entertaining stream.
This is the thing about Prime that keeps amazing me. When, at the end of a perfectly decent article, he goes "I have a counter take". … and I am like "hey that was a perfectly decent article, and it sounds like you are coming from a very limited sample size with your take". … and when I let him cook, I have nothing else to say but "hey that made 110% of the sense and you are correct, I didn't think of any of that".
My brain is too small for those math magic runes
15:40 Maybe I'm programming for 5 hours and don't get anything done. And then I get an idea and am 10 hours productive and don't even want to eat. And then I want to sleep, but I just want to finish it, I'm almost done anyway. If I tried to sleep now, I couldn't anyway and would just watch stupid stuff, and still be tired the next day.
And the next day, all my motivation would be gone. Now I have the whole day to finish it, and it's almost finished anyway. I'll first have a long breakfast, then I watch a few episodes of my current anime, then I slowly look into my project and realize I have no idea where to start, so I eat lunch and then go outside. And at 9 in the evening I think, I should just finish it, but since it's not finished yet at midnight and I have work tomorrow, I go to bed and haven't finished it yet. If I just finished it yesterday, it would have taken another 5 hours, but at least it would have been finished.
Something like this isn't considered in this algorithm, right?
At some point in this article productivity turned from an integrated amount to a rate of work completed.
I'm confused.... Why are the other unchecked boxes even present ;)
So what you're saying is that he forgot the cooldown on the focus ability?? Is that reduced if you level up? =)
The tasks are analysed as if they were all independent and static but if you do the first you have more energy then when you start the second or third after 3 or 4 hours. It will take you longer to get into it and you will not reach the same full concentration as with the first one in the morning. You get exhausted eventually over the day working at 100% full concentration. I think you have to do much more math here. Wink.
I on the other hand might eat the frog first, i.e. do the thing I like least and feel better after that is off the table or just do what I feel most enjoyable first depending on how I feel on that day.
The entire concept is flawed - instead of estimating roughly how long you think a task will take, or basing it on past experience when you'll start to crash, you are basically just guessing at a bunch of random constants for each and every task that you have defined arbitrary relationships between.
It also doesn't hold for tasks where you have an explicit workload to complete. You can't just decide "I'm not going to be optimally productive if I don't finish this maths homework, so I simply won't finish it" - the model only holds under the made up rule set the author has defined.
This same critique also holds for literally all quantitative finance - it only works in made up fantasy land
Oh yeah, its dumb. Everyone has different stale-task tolerance, short-task preference, post-task confidence, pre-task procrastination/anxiety, task-switch friction (stubbornness). Some days you just want to crunch a hard problem, some days you need to clear the minor tasks so you can focus. None of this can be enumerated, let alone calculated. Imagine making your own flawless personal formulas, but then you get irritated the by the rigidity and authority your own creation has over you, or lack the discipline to actually use the dang ol thing. There's being overly analytical, and there's trying to optimize humanity out of human life. Just wing it, or you'll never fly.
to be fair to the author, they went on to extend a feature inside the app where anyone can define a task as time-limited and just override the algorithm.
thanks, need to review for my upcoming midterms
13:40 Need some CUDA-fu to tackle this in software.
But is it written in Rust?
Euler? I barely know her
Man this sounds like an article I would absolutely write to procrastinate doing homework lol. (physics student here.) And it's so nice that I can understand almost everything they say (almost), and follow along with the math.
And no, I am not procrastinating doing homework rn. I'll have you know I am watching this after finishing my homework!
The equations are IMAGES KEKW :) - good to you!
I think this approach to maximizing flow state is fine, but he should have considered the boost in energy you get from finishing a task, the difference between tasks with a clear end in sight (like homework) with no follow-through, versus tasks with no clear end (like studying) with follow-through (having to remember things for exam), and also the difference in nature of tasks (video editing is not as mentally involved and more free-form as opposed to studying/homework)
For his example, I would have started with the homework, since it's mentally involved and not free form, so its uncomfortable to keep in the backlog. Its good to start with, since there's no follow-through.
I would ensure the studying for physics exam comes last as recency helps with remembering things (part of the follow-through which is the exam) and no clear end (you're never really done studying until the exam's over tbh)
next I'd go for the video editing task, since it's more free-form it'd act as a sort of break, while also being an atomic task (with no follow-through)
Basically, a more qualitative analysis is required when scheduling stuff optimally.
What's the program at the end he uses for picture/graph manipulation
Sounds like prime truly believes that everyone has the same potential
21:40 red light therapy shining on your balls xD
truly a wise man
Anyone heard of "taking a break"?
at 4:54... potate...the equation is a potate
So thats the MathOverflow error I keep hearing about.
Not gonna, learning advanced math be like this.
"Troth of sadness" totally stealing that term.
🙌 scheduling algorithms mentioned
F liking the video, I wanted to love it. Great content👌
...sounds like the end of every semester in Uni I attended.
Hard disagree on morning routine. The hour I spend every morning prioritizing and organizing is extremely valuable for my productivity.
How can tee = fee when you pay the fee before you tee up in golf?
Man you really are smart
I don’t know how you came on him so hard
I am sure that he hasn’t even thought about what you talked about
I don't need all this to know writing the article wasn't productive.
That kind of mathematical modelling is why all software sucks. Not only is the model wrong, it also reduces people to a few shallow dimensions. It's the opposite of sonder and a bad way to think about people.