(These links give me a little bump if you decide to use them. Thank you!)
“The road to hell is paved with intractable recursions, bad equilibria and information cascades”
Welcome to 3 Things You Can Use, where Maddi and I decode self-improvement and entrepreneurship through books, three things at a time.
Before we get started I wanna shout out to Brandon at OnePercentBetter who has given us the time of day even though we have 30 subs and he has many more.
He reviews books as well so check out his stuff by clicking the card up there.
And on to the show…
This week’s book is Algorithms to Live By: The Computer Science of Human Decisions by Brian Christian and Tom Griffiths.
In a world where math is becoming more and more of the prefered method to finding solutions – statistics, algorithms and data collection the driving players in business, politics and civilization as a whole, it was about time for someone to give us regular folk some math to use to improve our lives.
And these blokes did it.
So, what are three things we can use from this book?
3 Things You Can Use
1. Home Use Algorithms
The three ways in which algorithms can change your home life.
Laundry, todo lists and filing systems.
Starting with, yes, the best way to do your laundry.
Well there are three factors in doing laundry:
Washing, drying and sorting.
The factor that this algorithm takes into account is the sorting.
More sorting translates to increasingly more time taken.
So the less sorting the better. To accomplish this, do laundry more often.
The authors say that “Doing laundry three times as frequently could reduce your sorting overhead by a factor of nine.”
So to put this in perspective, let’s say that Bob does all of his laundry once a week. If instead he did his laundry three times a week, he would spend down to 1/9th the time on sorting than he did before.
Moral of this one: do your laundry more often if you want to save time.
What does math tell us about the optimal way to get through a todo list?
Start by assigning two factors to each thing on your todo list:
The importance and the time it will take
(For importance, keep it simple, maybe on a 1-4 importance scale.)
Now for every task on your todo list, divide the importance by the time it will take to end up with your density number. Then, with each of your tasks with their own density number, do each task from highest to lowest density.
This is the same concept as an hourly wage. Value per time. If you were a freelancer and you had 3 jobs at three different rates, $15hr, $12hr and $9hr, you would want to do the one with the highest value per time first, the $15, and then move down from there.
So calculate your tasks’ densities and optimize your time.
And finally, the third home help, how to file?
Filing systems. The annoying drawer crammed with papers, folders and that apple core you dropped in there and never found. So many files unused for decades and so many files that you need all the time that you can never find. How can math help us here?
It’s called the Noguchi method or the LRU principle.
Every time you use a folder or paper from your filing system, return it to the far left of your cabinet.
Preparing what you used most recently is apparently the optimal system for approximating what you’ll need to use next.
They say that when using this method, “the total amount of time you spend searching will never be more than twice as long as if you’d known the future. That’s not a guarantee that any other algorithm can make.”
That means that if you file your papers to the far most left when you use them and look from the left to the right when searching for them and you’ll be half as good as if it popped out from as soon as you needed it.
2. Optimal Choosing
Whom to marry?
(Unless you’re already married, then we’ll say “Who to hire as your secretary”)
Is there a recipe to find the best mate?
There sure is.
It’s called optimal stopping theory and it’s the best way math can give us for whom to choose.
Say you will go through 100 potential mates, one at a time. You “interview” each one and then move on to the next. Some are good matches, some are great matches and some are poor matches. You’re sifting through the varying degrees of compatibility in order to find the best fit candidate.
There is one catch — you have to pick one and there is no going back.
This brings up some issues. What if the first one is the best? You blow right through them and are met with 99 more candidates that don’t measure up.
What if the last one is the best? Do you think you would hold out to the very end through all the uncertainty to end up with the best candidate? Prob not.
Or really anything in between? When do we have enough compatibility info to go off of and when do we stop and choose?
That’s right. The best mathematical instruction tells us to go through 37 of the candidates, discover the one that we like the best out of that bunch and then hold that candidate as a model for the rest of the potential partners that we go through.
As soon as we come across one that is better than our model, we pounce, and marry them.
This is the optimal stopping theory. See 37% then leap.
So if you had 5 to choose from, check out the first two, no commitment, and then for the next 3 begin choice mode.
While there are caveats to this one, regarding marriage, sometimes you can go back to choose, feelings can sometimes make objective judgement harder and sometimes they don’t want to marry you back, this is the best you can get from math regarding whom to choose.
And just to further help with this concept, and perhaps simplify a bit as well,
You can look at it from an age standpoint. Say you want to get married by the time you’re 30 and you’re 20 now. So you’ve got 10 years. For 3.7 years just date and date, no settling down allowed. Then after this time period, choose any person you find that is better than the best you dated during those 3.7 years.
3. Creating Strategies
Whether intentionally or not, we all have strategies for how we go about life. But there are a few factors to take note of, when designing and implementing these strategies.
You’ll see what I mean…
The first things we want to consider are the metrics we track. A metric you might track could be how much work you get done. This incentivises you to get more work done, so you can watch your metric go up.
But overfitting, as they call it, or targeting one metric too much without regard to others can end up being ineffective or shortsighted.
“At a job placement firm, staffers were evaluated on the number of interviews they conducted which motivated them to run through the meetings as quickly as possible without spending much time actually helping their clients find jobs. At a federal law enforcement agency, investigators given monthly performance quotas were found to pick easy cases at the end of the month rather than the most urgent ones. (…) A factory focusing on production metrics led supervisors to neglect maintenance and repairs, setting up future catastrophe.”
Focusing so much on one metric, how much work you can get done, can neglect the other metrics like how much sleep you get, how much enjoyment you allow yourself and how much you play with your kids, etc…
We have to be careful what we incentivise or we could end up with “the ruthless and clever optimization of the wrong thing.”
Moving on, the next helpful thing to our strategies is knowing when to relax.
A lot of this book was about figuring out and applying formulas to everyday life to make us more productive and more effective. But they also point out the dichotomy here of how if all of your time is spent calculating and figuring your plans, strategies and methods, you’ll never actually get anything done.
Sometimes instead of spinning your wheels it’s better to just jump in and figure it out from there rather than optimize your strategy beforehand.
They cement this concept with the quote, “Don’t let perfect be the enemy of good.”
And our final consideration given to us from computer science is interrupt coalescing.
This just means batch processing.
Take for example a comparison between email and snail mail.
Email, you are notified instantly, interrupted no matter what, smashing your train of thought off the tracks and abrasively sliding it through the mud and trees, off the cliff every time you get a dang email.
Though with snail mail, you get your mail at the same time every day, whether it’s 50 letters or two.
Snail mail is interrupt coalescing, keeping interruptions at a minimum, so you don’t waste too much willpower task-switching and stressing all day.
They suggest to set certain times of the day when you will catch up on your email or text messages, really anything that would usually interrupt you, turning off the notifications for such things until the times when you are doing your batch processing.
Fix your home-life by doing your laundry more, weighing your todos and filing to the left.
Find where to live, what college to go to or who to marry by first going through 37%.
Keep your strategies clean and effective by watching your metrics, knowing when to relax and turning off your notifications.
The explore, exploit theory, scheduling theory, Bayes’ Rule, Game Theory, Computational kindness, not to mention, big O notation, Gant charts, thrashing, Power laws, Occam’s razor, the bloom filter, packet switching and recursion, this book was jam packed with maths and proofs.
And while they did occasionally lose me with the jargon, that they tried to explain already, the end result is that you come away with lots of general purpose recipies to use in your life along with a better understanding and an appreciation for computer science.
I would say to stay away if you’re allergic to math, but otherwise it’s a great read and it’s 5 star reviews all over the place can reaffirm that.
That’s all for this one!
Thank you for watching, if you’re not subscribed, make sure to subscribe so you don’t miss next week’s video, and –
we’ll see you next week!
Want to Read it?
Algorithms to Live By (Physical Copy)
(These links give me a little bump if you decide to use them. Thank you!)
- Punching Sounds by: Mike Koenig
- Reading Photo at end of video by: Marketa