I've been thinking a lot lately about this little old lady who lived in my neighborhood in Chicago. It was on this street lined with interesting old brownstones with cool little apartments. She lived in a garden level apartment (for those who don't know, this essentially means street level apartment) From what I understand she was a florist so she didn't make a ton of money and sometimes struggled to make ends meet but had been scraping by, renting the place for like 20 years. I'm not really much of a "flower" guy, but the flowers and plants she had on her little like 6 foot by 6 foot patio/stoop were fuckin awesome. Seems like she planted them so there was almost always something in bloom. Most of the buildings in the neighborhood had a flower pot or two out front, but hers were pretty impressive. I was fresh out of college and had been renting in the neighborhood for about 3 years before she passed away. After she was gone it was impossible not to notice that the new people who moved in had no plants. Over the next few years all of the buildings stopped putting flowers out on their stoops, the burnt out porch lights were no longer being changed, etc... I don't want to imply the neighborhood fell apart, it was a nice neighborhood, it was just that the neighborhood was noticeably less pretty, the details were missing. I'm totally aware that I'm making a leap here, but after talking to some neighbors, I truly believe her love for flowers and her beautiful little garden had a ripple effect, 2nd and 3rd order effects that none of us really understood until she was gone. That her little garden inspired other renters to add their own little piece to the neighborhood. I don't know how we measure and reward someone like her, how we measure someone who's little daily actions cause positive ripple effects, but fuck I really wish we knew how.
I think we'll always measure with numbers, be governed by numbers, but more and more I find myself hoping we get better at it--that we figure out better metrics by which to measure success and quality of life--to notice the ways things ripple out. Lately every time I hear someone degrading someone else because that person isn't rich enough or educated enough, her little garden pops into my head.
Just to be clear, I recognize that markets often do a good job of measuring certain things, but lately it feels like our current iteration of measurements, if we really wanted to, we could maybe make them better.
Some things shouldn't be measured, or be about any external validation. I'm sure that old lady wasn't doing it for recognition or reward, but just because she felt it the right thing to do. A sense of personal or civic pride if you like, or wanting to make her little bit of the world that bit nicer, for everyone.
It's something we mostly forgot. Parks and gardens are maintained at lowest, most efficient cost; businesses build the cheapest box building planning laws let them; people couldn't care less about much of anything beyond them and theirs. Everywhere you look things are that much more dull and uninteresting, or messy and rubbish strewn than they once were. I suspect it was an intended consequence of the Thatcher/Reagan years. It's certainly removed a lot of the humanity.
I remember two stories on HN - one about a guy somewhere near NY, who is quietly digitizing tons of old newspapers. It is all available on his website for free. Another about a couple who are collecting seeds, I think.
Something like 2 years ago, a friend of my flat mate came to visit from her home country. She is a very energic person and transmits happiness, we would be just having a drink and talking and suddenly she would get up and start dancing, no reasons needed, "I just feel like it".
We were talking avout drawing and I suggested her to draw something nice for us, told her that if she did I would proudly stick it on the outside part of our flat door so other neighbors could see it.
And boy did they love it. Not because the drawing was beautiful per se, but because of the good idea of bringing some colorful touches to a boring and standard common corridor. Also, most of my neighbors have kids, so they started making their own drawings and sticking them on their doors.
Fast forward to present day. Drawings change every so often, the children draw new pieces wishing Merry Christmas, or the arrival of the Spring. We at home also encourage our visitors to draw something, most are shy about it because "they don't know how to draw" without realizing the expectations are those of a 3-years old kid :-)
Such an inocuous idea has had a very positive impact on a very little aspect of our lives in this particular building. Also the first impressions from new people coming to the place are always fun to see.
We need less robotic-like number optimizations in our lifes, and more of the good-old "do as you feel like" (even if it's not optimum or maximizes some metric)
I suspect we are a fairly long way from measuring that. And even when we do manage to measure that positive ripple effect, do you want to?
The reality is, numbers are used for accounting. If we start keeping an accounting of small actions such as that, what freedom will we have?
People being mean, nice, depressed, etc. they are what make is human. By accounting for the little things we’d incentivize uniform behavior and lose all diversity of any kind.
I too think on those little things a lot. I recently had a family member move away, and since then every family get together is just flat in comparison. I think in a small way, those flowers did the same to the neighborhood.
Sometimes I think it’s enough to say, we lost something when they left. Or gained something when they were here. Maybe that’s enough of a quantified “number”.
> And even when we do manage to measure that positive ripple effect, do you want to?
> The reality is, numbers are used for accounting. If we start keeping an accounting of small actions such as that...
This is a really good point. I think your point fits well with a sibling comment to yours, Gibbon1 pointed out something called The McNamara Fallacy which says we often trick ourselves into believing that the most important things are those which we can measure quantitatively, that they must be more important if only because they are measurable.
I do still wish we were better - on a level above The Individual - at recognizing the ways we add value to one another's lives. I do feel like, on a societal and systems level we're missing out on a lot of the little ways we make each other's lives more colorful, the ripples or 2nd and 3rd order effects. Taken as a whole, all of these little value-adds, everything from the artist, a neighbor's little garden, to the guy on my team who writes hilariously snarky code comments, I think these things probably add-up and affect us more than we realize.
Maybe we need to be better at teaching from a young age to notice those little ripples? Anyway, thanks for your comment, it helped me see it differently.
You seem to be hedging and I'm not sure why. Your core point is one that every good engineer should understand: not all important things can be practically measured. We've all watched metrics or standards defined with the best of intentions driven into the ground by a narrow-minded focus on the wrong numbers. Why should we be surprised that this is true in economics or politics or philosophy?
Increasingly I think that there are many things that measurement simply doesn't apply to. That our industrial and post industrial society has conditioned us to believe that the applicability of statistics to complex but still mass produced items can be extended to humans and social situations in a simple way.
But actually, it can't. Humans are all very different, calculus's of human affairs must deal with penalty functions full of infinities of loss.
> I don't know how we measure and reward someone like her, how we measure someone who's little daily actions cause positive ripple effects, but fuck I really wish we knew how.
We do know how. It's called civic engagement. We try to teach it to our children in organizations like the Boy / Girl Scouts, which gamify personal development and social contribution in the form of badges. As adults, civic engagement is driven by communal recognition and appreciation, in the form of honorary plaques presented at fund-raising banquets in front of friends and neighbors. But this has been in decline, for two underlying reasons.
The first and biggest reason is the breakdown of community and communal institutions. In rural areas, communal institutions have been squeezed by population decline, as people have left for cities in search of economic opportunity. People working multiple jobs to make ends meet don't have the time or energy to volunteer in their communities. In cities, the failure of municipal planners to put enough housing on the market for people to buy affordable apartments has driven people into the rental market, which prevents people from putting down roots and committing to their communities. If you look at the Bay Area, most 20-somethings look around and know in the back of their minds that they are going to need to move away to start a family. Why would you put your heart and soul into an urban garden if you're not going to be around to enjoy the fruits of your labors?
The second reason is a fearful overreaction. Civic engagement is a spectrum, and on the extreme end you find nativism and xenophobia. The social stability required to promote civic engagement means that such civically-minded organizations tend to be led by socially conservative people, and if left unchecked, these organizations can be unwelcoming. But there's no reason why civic volunteering needs to be socially conservative, and indeed, civic volunteering is made better when it's more welcoming and inclusive. Unfortunately, many people have knee-jerk reactions and throw out the baby with the bathwater, and reject these kinds of organizations on principle, but many such people would be put at ease by strong community leadership that fostered an inclusive environment for them.
Ultimately, it takes strong community leaders to give civic expression room to take root, grow, and stay inclusive over time. So the real question isn't how can we promote civic engagement - it's why don't we have strong local leadership?
Speaking of knee-jerk, you should note social conservatives tend to be the bedrock of organisations for a reason. While I love a good ol bashing of yesterdays politics, placing blame on people who think differently to you is the exact thing you are complaining about.
You don’t need to capture such things with numbers. What’s needed is neighborly love... if you see her struggling (financially, physically, emotionally, etc), then help her and encourage the other neighbors to do so as well. Let her know she’s appreciated. Encourage each other.
I think this one aspect of what the Chinese are trying to build with their social credit system. Not to say this is a good idea, I think there is no way to implement something like this fairly, but I know they give social score for neighborhood improvements under some narrowly defined criteria.
Given extensive enough monitoring and ML wouldn't it be possible to monitor the emotions of the citizens in a neighborhood and reward people who positively impact their neighbors?
No, because you get what you measure. If there are concrete rewards attached, people will immediately try to game the system. You can ameliorate this somewhat by being careful what you measure, but you'll still wind up with a metrics-driven tick-the-boxes response rather than an organic result driven by individual human motivations.
To the extent this is a real phenomenon, it's because of competition and not measurement. Competitions that are judged qualitatively have the same effects, and measurements without competition (e.g. measuring the height and weight of babies) don't result in power accruing to the folks doing the measuring.
That said, people systematically overvalue anything with numbers attached due not understanding the epistemology of measurement and statistics. That's why managers at software companies are obsessed with pointing tickets, why scientists are obsessed with p-values, why the school system is obsessed with grades, etc. So if you want people to compete more fiercely over something, coming up with some numbers is a good way to make that happen. (At least for a while.)
It'd be nice to believe that this is just a temporary quirk of our current cultural values. But we may also have just topped out, at least for the foreseeable future, in terms of just how well people can really understand the world given a relatively fixed level of intelligence. Even though individuals are ostensibly becoming more transmodernist, it's difficult (as a layperson) to see how that or something like it could ever really could become the dominant philosophical paradigm behind our institutions.
Yes but don't too readily dismiss the notion that he who is tabulating will eventually tabulate into his own favor. Of course, this notion holds no water when looked at from a bureaucratic point of person but if we look at how BANKS are doing we might start wondering if this "overdraft fee" is really a beneficial thing for all or some. To use one concrete example for this camp.
Fair enough, and to the extent that the zeitgeist is sort of retroactively determined by whatever technological and political things are going on at the time, I wonder what new beliefs people will have about the state of the world in 50 or 100 years as the result of DLT.
On the surface the ability of banks to steal from folks seems like largely a different type of power than what the economist is talking about, which is the ability of metrics (and the people who define them) to shape human behavior. But maybe it just looks that way because it's difficult or impossible to really perceive the full extent of how living in a non-DLT world is affecting our behavior.
Everyone should know three things:
1. You can measure anything
2. Measurements always have error
3. Measures aren't always aligned to what they are trying to measure, so even validated measures need critical thinking based on qualitative experience.
We're doing a very dangerous thing here. We're crossing the line between computers computing, and computers advising. When a computer computes, you can be assured the answer is correct. When a computer advises, the average person takes that information at the same value. There's a long laundry list of reasons why this is horrible. I went over a few in my essay about platforms being the enemy. http://tiny-giant-books.com/Entry2.html?EntryId=recEUbufzhAv...
There's a deeper moral/ethical issue at stake here. If you program a computer, you are responsible for that computer never presenting a misleading view of reality to the user. People don't distinguish between their tax program, GPS, and voting on reddit. When you use the interface to guide or subtly mislead people? You're hurting millions of people just a tiny bit at a time.
So many net-level effects involve these tiny changes that are impossible to evaluate. That delays or completely obfuscates the feedback loop. Very bad stuff here.
I have always, falsely, taken it for granted this is common knowledge.
The fact that everything is quantified doesn't necessarily remove bias from the forthcoming decisions. Instead it provides additional fuel to hone and precision tune that bias for maximum manipulation. This occurs even unintentionally. The safety check to reduce bias is to ensure the quantified data is available to a wide number of stakeholders and open to scrutiny.
is there not some formal "rule of thumb" describing that attempts to improve a system through "point based" incentives ultimately lead to undesirable outcomes as the incentives naively skew focus away from the starting goal of improving the system to simply gaming the point structure?
what I wrote feels hard to put cleanly into words, but does that makes any sense to anyone else?
As well as Goodhart, there's also "People with targets and jobs dependent upon meeting them will probably meet the targets - even if they have to destroy the enterprise to do it." W. Edwards Deming
As seen everywhere including in hospital care outcomes, waiting lists and school standards.
Good question! Goodhart’s Law fits this quite well. :)
From Wikipedia: “Goodhart's law is an adage named after economist Charles Goodhart, which has been phrased by Marilyn Strathern as "When a measure becomes a target, it ceases to be a good measure." [0]
Yeah, but read his actual papers. Goodhart doesn't actually advocate against using measures. He just shows how measures can be corrupted and presents suggestions on avoiding this
Reading the thread there is a lot of pessimism in regards to whether measuring could ever be a positive and "life-giving" (in the sense of that flower story).
I'd like to argue that yes it can be just that, there's two reasons why I think so. First is Christopher Alexander and his work on wholeness.
Second is that I think the fundamental problem with our economic measurements is the "single source of truth" problem. When there is a just a single entity / stakeholder that prints currency and thereby defines baseline value then all the economic actors have less power to affect the definition of value. The development I'd like to see is democratization of stocks, making them the currencies of day to day life, and the phasing out of national currencies.
I haven't figure out yet exactly how to implement the intricacies of such a system but the general idea is that we need economic systems to acknowledge relativity in value better and stop hoping for a concrete truth to base our reality on.
... That economists get to pretend matter, while ignoring all the numbers that they can't measure and will kill us.
Economics on its own can't measure the impact of unknowns like global warming impact. It just gives the tools to the rich and elite and corporations to resist paying for it and dealing with it for as long as possible.
Everything is NOT QUANTIFIED. The theory of computation, physics, math, the uncertainty principle, and chaos GUARANTEE we do not have accurate quantification.
What does exist is bullshit quantification in favor of the powers that be that fund the studies and think tanks to justify their existence and political desires.
Yet another example of gross overreach of the pseudoscience of Economics.
This could also be phrased '[...] are increasingly reduced to numbers', in order to be rationalized by algorithms. How much wealth is counted multiple times by multiple models all recognizing the same qualities under different enumeration schemes?
I expect to see more and more references to Goodhart's Law as this increasing bias in favor of numerically digestible value amplifies the (often unacknowledged) premise that 'that which cannot be counted, doesn't count'
Readit News
I think we'll always measure with numbers, be governed by numbers, but more and more I find myself hoping we get better at it--that we figure out better metrics by which to measure success and quality of life--to notice the ways things ripple out. Lately every time I hear someone degrading someone else because that person isn't rich enough or educated enough, her little garden pops into my head.
Just to be clear, I recognize that markets often do a good job of measuring certain things, but lately it feels like our current iteration of measurements, if we really wanted to, we could maybe make them better.
It's something we mostly forgot. Parks and gardens are maintained at lowest, most efficient cost; businesses build the cheapest box building planning laws let them; people couldn't care less about much of anything beyond them and theirs. Everywhere you look things are that much more dull and uninteresting, or messy and rubbish strewn than they once were. I suspect it was an intended consequence of the Thatcher/Reagan years. It's certainly removed a lot of the humanity.
I can give no better example of striving to make the world a tiny bit better for no recognition or reward than this story from yesterday: https://www.theguardian.com/uk-news/2019/feb/22/man-who-tend...
If you click through to the twitter threads that initiated this, you may end up with grit in your eyes. :)
Then there is this man who built a road by himself - https://www.thebetterindia.com/18326/the-man-who-moved-a-mou...
The reason we measure things is so people can't make assertions like this without some kind of evidence.
We were talking avout drawing and I suggested her to draw something nice for us, told her that if she did I would proudly stick it on the outside part of our flat door so other neighbors could see it.
And boy did they love it. Not because the drawing was beautiful per se, but because of the good idea of bringing some colorful touches to a boring and standard common corridor. Also, most of my neighbors have kids, so they started making their own drawings and sticking them on their doors.
Fast forward to present day. Drawings change every so often, the children draw new pieces wishing Merry Christmas, or the arrival of the Spring. We at home also encourage our visitors to draw something, most are shy about it because "they don't know how to draw" without realizing the expectations are those of a 3-years old kid :-)
Such an inocuous idea has had a very positive impact on a very little aspect of our lives in this particular building. Also the first impressions from new people coming to the place are always fun to see.
We need less robotic-like number optimizations in our lifes, and more of the good-old "do as you feel like" (even if it's not optimum or maximizes some metric)
The reality is, numbers are used for accounting. If we start keeping an accounting of small actions such as that, what freedom will we have?
People being mean, nice, depressed, etc. they are what make is human. By accounting for the little things we’d incentivize uniform behavior and lose all diversity of any kind.
I too think on those little things a lot. I recently had a family member move away, and since then every family get together is just flat in comparison. I think in a small way, those flowers did the same to the neighborhood.
Sometimes I think it’s enough to say, we lost something when they left. Or gained something when they were here. Maybe that’s enough of a quantified “number”.
> The reality is, numbers are used for accounting. If we start keeping an accounting of small actions such as that...
This is a really good point. I think your point fits well with a sibling comment to yours, Gibbon1 pointed out something called The McNamara Fallacy which says we often trick ourselves into believing that the most important things are those which we can measure quantitatively, that they must be more important if only because they are measurable.
I do still wish we were better - on a level above The Individual - at recognizing the ways we add value to one another's lives. I do feel like, on a societal and systems level we're missing out on a lot of the little ways we make each other's lives more colorful, the ripples or 2nd and 3rd order effects. Taken as a whole, all of these little value-adds, everything from the artist, a neighbor's little garden, to the guy on my team who writes hilariously snarky code comments, I think these things probably add-up and affect us more than we realize.
Maybe we need to be better at teaching from a young age to notice those little ripples? Anyway, thanks for your comment, it helped me see it differently.
One of my points is there is a lot of necessary and beneficial things people do that don't fit into being exploited via paid work.
https://news.ycombinator.com/reply?id=19206174
Increasingly I think that there are many things that measurement simply doesn't apply to. That our industrial and post industrial society has conditioned us to believe that the applicability of statistics to complex but still mass produced items can be extended to humans and social situations in a simple way.
But actually, it can't. Humans are all very different, calculus's of human affairs must deal with penalty functions full of infinities of loss.
Just recognising this does not mean you are "anti-market".
Thank you for your beautiful story that highlights this!
We do know how. It's called civic engagement. We try to teach it to our children in organizations like the Boy / Girl Scouts, which gamify personal development and social contribution in the form of badges. As adults, civic engagement is driven by communal recognition and appreciation, in the form of honorary plaques presented at fund-raising banquets in front of friends and neighbors. But this has been in decline, for two underlying reasons.
The first and biggest reason is the breakdown of community and communal institutions. In rural areas, communal institutions have been squeezed by population decline, as people have left for cities in search of economic opportunity. People working multiple jobs to make ends meet don't have the time or energy to volunteer in their communities. In cities, the failure of municipal planners to put enough housing on the market for people to buy affordable apartments has driven people into the rental market, which prevents people from putting down roots and committing to their communities. If you look at the Bay Area, most 20-somethings look around and know in the back of their minds that they are going to need to move away to start a family. Why would you put your heart and soul into an urban garden if you're not going to be around to enjoy the fruits of your labors?
The second reason is a fearful overreaction. Civic engagement is a spectrum, and on the extreme end you find nativism and xenophobia. The social stability required to promote civic engagement means that such civically-minded organizations tend to be led by socially conservative people, and if left unchecked, these organizations can be unwelcoming. But there's no reason why civic volunteering needs to be socially conservative, and indeed, civic volunteering is made better when it's more welcoming and inclusive. Unfortunately, many people have knee-jerk reactions and throw out the baby with the bathwater, and reject these kinds of organizations on principle, but many such people would be put at ease by strong community leadership that fostered an inclusive environment for them.
Ultimately, it takes strong community leaders to give civic expression room to take root, grow, and stay inclusive over time. So the real question isn't how can we promote civic engagement - it's why don't we have strong local leadership?
Given extensive enough monitoring and ML wouldn't it be possible to monitor the emotions of the citizens in a neighborhood and reward people who positively impact their neighbors?
Dead Comment
That said, people systematically overvalue anything with numbers attached due not understanding the epistemology of measurement and statistics. That's why managers at software companies are obsessed with pointing tickets, why scientists are obsessed with p-values, why the school system is obsessed with grades, etc. So if you want people to compete more fiercely over something, coming up with some numbers is a good way to make that happen. (At least for a while.)
It'd be nice to believe that this is just a temporary quirk of our current cultural values. But we may also have just topped out, at least for the foreseeable future, in terms of just how well people can really understand the world given a relatively fixed level of intelligence. Even though individuals are ostensibly becoming more transmodernist, it's difficult (as a layperson) to see how that or something like it could ever really could become the dominant philosophical paradigm behind our institutions.
On the surface the ability of banks to steal from folks seems like largely a different type of power than what the economist is talking about, which is the ability of metrics (and the people who define them) to shape human behavior. But maybe it just looks that way because it's difficult or impossible to really perceive the full extent of how living in a non-DLT world is affecting our behavior.
Deleted Comment
https://www.amazon.com/Measurement-Psychology-Critical-Histo...
There's a deeper moral/ethical issue at stake here. If you program a computer, you are responsible for that computer never presenting a misleading view of reality to the user. People don't distinguish between their tax program, GPS, and voting on reddit. When you use the interface to guide or subtly mislead people? You're hurting millions of people just a tiny bit at a time.
So many net-level effects involve these tiny changes that are impossible to evaluate. That delays or completely obfuscates the feedback loop. Very bad stuff here.
The fact that everything is quantified doesn't necessarily remove bias from the forthcoming decisions. Instead it provides additional fuel to hone and precision tune that bias for maximum manipulation. This occurs even unintentionally. The safety check to reduce bias is to ensure the quantified data is available to a wide number of stakeholders and open to scrutiny.
what I wrote feels hard to put cleanly into words, but does that makes any sense to anyone else?
As seen everywhere including in hospital care outcomes, waiting lists and school standards.
https://en.m.wikipedia.org/wiki/Goodhart's_law
From Wikipedia: “Goodhart's law is an adage named after economist Charles Goodhart, which has been phrased by Marilyn Strathern as "When a measure becomes a target, it ceases to be a good measure." [0]
[0] https://en.m.wikipedia.org/wiki/Goodhart's_law
I'd like to argue that yes it can be just that, there's two reasons why I think so. First is Christopher Alexander and his work on wholeness.
Second is that I think the fundamental problem with our economic measurements is the "single source of truth" problem. When there is a just a single entity / stakeholder that prints currency and thereby defines baseline value then all the economic actors have less power to affect the definition of value. The development I'd like to see is democratization of stocks, making them the currencies of day to day life, and the phasing out of national currencies.
I haven't figure out yet exactly how to implement the intricacies of such a system but the general idea is that we need economic systems to acknowledge relativity in value better and stop hoping for a concrete truth to base our reality on.
Economics on its own can't measure the impact of unknowns like global warming impact. It just gives the tools to the rich and elite and corporations to resist paying for it and dealing with it for as long as possible.
Everything is NOT QUANTIFIED. The theory of computation, physics, math, the uncertainty principle, and chaos GUARANTEE we do not have accurate quantification.
What does exist is bullshit quantification in favor of the powers that be that fund the studies and think tanks to justify their existence and political desires.
Yet another example of gross overreach of the pseudoscience of Economics.
I expect to see more and more references to Goodhart's Law as this increasing bias in favor of numerically digestible value amplifies the (often unacknowledged) premise that 'that which cannot be counted, doesn't count'