Reality has a surprising amount of detail (2017)
johnsalvatier.org
johnsalvatier.org
It seems that everybody knows everything. Whatever the issue, a little bit of Googling and suddenly you know all there is to know. Even more wondrous, no matter what position you take on anything, some more search engine Kung Fu and you can find a hundred people willing to support you with arguments, surveys, facts -- whatever you need.
Everybody knows everything. It's quite amazing. And then when you take a tech team into an unknown domain, suddenly they find it very difficult to open up, admit ignorance, and reason about things.
I am reminded of some startup book or blog I read years ago. It was talking about the relationship between intelligence and startup success. The author said that there was a correlation. It was an inverse correlation. The more you have been rewarded in life for being smart and knowing everything, the more you felt intelligent, the less chance you had of making a startup work. You just weren't able to admit all the things you didn't know.
Understanding is not a matter of a google. It's a decade of relevant experience.
You might beat the PhD to the factoid phrase, but when you come to employ the fact in the relevant problem you won't even know where to begin or even what it really meant.
People aren't exercising their ability to distinguish parroting from understanding, as-if intellectual work is just different forms of trivial pursuit.
If I google, find a primary source with reasonable data, and form an argument based on that, then I’ve done research. It was simply easier based on the Internet, but the research is not proof-of-work - a well-supported conclusion is just that no matter the effort.
On the other hand, I google, copy and paste the first link and parrot what it said - then I have indeed cheapened it. I’ve skipped the evaluation step of researching, wherein I convince myself of the “facts” of the past and I’ve also failed to extend any new thoughts as we’ve just parroted the argument.
My main point being, to a critical reader only “bad” research is cheap. It is also the duty of the audience, just as much the researcher, to evaluate what they read critically and respond with skepticism.
Would you mind sharing the context?
I work in a field where the PR machine of my target "research" jumps ahead of the methodological reports. Then those reports are often vague (length limits?) about the details of their methodology. That makes me not trust much that I find - but I've got to have something - so I end up doing some sort of meta-analysis. It's time-consuming, exausting and frustrating - and has confidence-bars that might as well be non-existent - but without doing the on-the-ground fundamental-study - it's what I've got. For example - I just spent 60 hours over the last 4 days trying to triangulate a "probably correct-ish" value and range for a variable from a heck-of-a-lot of studies that didn't use the same methodology. Would that count as research in your book, or not?
People (who are intellectually honest) say "I don't know" when a question is specifically addressed to them, and so they feel a responsibility to give an answer. They don't respond at all when they don't know the answer and they haven't been addressed... except if they are motivated by curiosity and wish to indicate that they, too, are interested in the answer.
Incidentally, the mistake of assuming that everything on the Net is addressed to you is a common failing among people who have little experience. It happens on Facebook all the time...
Online written communications will never follow the same patterns as direct communication, and that is a good thing.
Even the very word discussion comes from late Middle English (in the sense ‘dispel, disperse’, also ‘examine by argument’): from Latin discuss- ‘dashed to pieces’, later ‘investigated’, from the verb discutere, from dis- ‘apart’ + quatere ‘shake’.
We generally don't do a very good job of talking about things in way that brings those things, or us each other, closer together.
Very often, English-language experts can appear to defeat subject-matter experts in debates. When this happens it's usually because of footwork on the "language layer."
My self esteem has also been heavily based on believing that I'm really smart, but I never had a problem admitting I didn't know something. "Smart" to me means my ability to learn, problem solve, and make connections.
One of the most significant factors in my success to date has been my almost eager willingness to admit my lack of knowledge, and to ask others to share their knowledge.
How else are you going to learn new things?
I remember the good old days, before the internet, when people would always admit their ignorance in real life and the Dunning–Kruger effect didn't exist.
Their system isn’t too bright.
But I'll often bookend a comment like that with points about what I don't know. And here, there are points I think someone might know - frame those as questions. But also there are point I doubt either myself or the person I'm replying to knows, frame those as problems of knowledge, statistics and so-forth.
Obviously, I think I know a general framework, that our beliefs are areas of semi-certainty bordered by heuristics, myth, ignorance and lack of awareness. But this is a belief that will often about what I or we, don't know.
1. Being right doesn't mean being smart. I met many people in my life that were smart or dumb, but their success had nothing to do with this. You can do the right things by accident.
2. Being rich often removes the need of being smart to varying degrees. People want to help you getting richer if they get a piece of the pie.
Who could qualify this?
I mean, you found a company, talk to the right people and get rich.
If you just happen to find the wrong people, you won't get rich.
Sure, you could say the sales people were skilled and found these people, but how often is this really true and how often do you just luck out by finding the right number on some obscure website?
However, he would be served very well to work on eliminating the moderate pomposity that permeates much of his writing. He is also overly verbose sometimes. Fixing those two issues would reduce a lot of the beef he gets, I think.
(I think I see scientists forget this too often. There are things that are true that we don’t know. Being true and being sure are different things and you should respect people who are going to show you true things otherwise you will never learn.)
My instinct is to question the causality anyway, because I think you highlight a strong candidate for the actual, underlying limitation on intelligent leaders: humility. My hypothesis would be that the negative leadership performance is actually a proxy measure of humility, which on average is lower with higher intelligence, likely because of the praise effect you describe.
Because I don't think it makes sense that intelligent leaders are less able to question their frame of thought or significant details supporting it. Or rather less capable in carrying out the actual cognitive task of self-questioning. Quite the opposite, because general intelligence is a measure of the correlation between performance on all cognitive tasks, so it's practically tautological that higher IQ individuals would be stronger within that cognitive domain. Which makes me think it must be their willingness or propensity to engage in the process of self-questioning.
To bring it back around to the OP, details matter, but the process for selecting relevant and filtering out irrelevant details matters most. Humility may be one way to enhance that process. Though maybe there is also a limit. Too much questioning prevents the establishment of a stable, actionable consensus. An inability to shut out irrelevant stimuli is disabling to individuals and organizations. Like Funes the Memorious, you get lost in minutiae.
I don't know where the ideal balance lies, but I wish I was better at finding it in practice.
You make an excellent point about filtering out the irrelevant. The wood is brown, for example, is irrelevant to building a proper staircase. The angle of decline is highly relevant.
But, with regard to leadership, the Peter Principle (identified by Laurence J. Peter and published in 1969) is in inverse correlation to performance, I believe. Particularly if performance is tied to being able to communicate with the success of the high performer at that level, because they might not have met their own level of the Peter Principle.
And, after communicating, one must be able to act. If a person has reached their Peter Principle level, they will be incompetent at the task -- which is why so many managers are hated.
There's nothing like that online, so any given advice or answer in a comment could be coming from someone with 20 years of experience, or from someone who googled the exact thing you already did, and copied the first answer they saw on StackOverflow.
It would be cool in addition to "comments", there was a Q&A section. I find I mostly click on the comments to get a more nuanced understanding of an article only to find a 20-commeny deep flame war about something unrelated.
I think there's something to this, but I want to present an alternative, which is that the more I know, the more I know what I don't know. And with regards to startups and business, the effect of this that I can observe is one of paralysis. I am scared to death of committing to any idea because I can see right away many of the unknowns, and I fear the unknown unknowns. After some education, I have enough history of thinking I know something and then discovering that I basically know nothing, that I am scared to ever say, well THIS is the idea that is going to work out, I'll put all my money and effort for the next 10 years into THIS because I am sure. Sadly, the more I know, the less sure I am about anything, and it makes me unwilling to take risks. Not just business -- even deciding what to study next, what jobs to apply to, or what side project to begin -- every time I start something that seems simple enough, it takes about 5 minutes to realize I have no idea what I'm doing and have to evaluate whether it's worth pouring more time into. Result: things don't get done. Because there is no done. I do learn a lot however. But for what? Personal growth I guess. But I have no idea what to do with it. I really envy those people who are able to decide they are going to solve a specific problem, that it's worth it to do so, and are able to see it through to the end, instead of learning everything they can about it, getting bored, and moving on, which seems to be my pattern. In the end, for me, it is the learning that is motivating, not the goal, and that has terrible consequences for what you describe. Perhaps "successful at startups" is not the only metric, but at least for that metric, it is maybe not good to take the "I'm going to learn everything about this before actually doing anything" approach.
Now once the risk reduced to affordable level, my curiosity often takes over and I often gets excited to explore my unknowns.
It is sad that for some people that all they see is unaffordable risks and they have suppressed and forgotten their curiosity -- the source of life's excitement.
They are over “you can be anything you want to, Billy” and they have justifiable fears.
But if you ran your own deli for years and ended up building your own bagel slicing machine because nobody else had one or they mangled too many bagels, why not sleep in a little in the morning and sell bagel machines instead? You know your customer because you were your first customer. You can just ask the deli owner in your head what they would like.
I think about this from time to time -- how people build up this sense of personality related to a forum or site. I think what tends to happen for an individual is that when you're talking to millions of people you don't have the perspective on who says what anymore, it becomes "twitter thinks" or "reddit says" instead of "these 10 people."
I think the reason everybody knows everything is that because we don't look at the individuals when we talk in a forum like this. There are always going to be some 'famous' people whose names pop up over and over, but as general responses you don't really know what a person thinks about something. We internalize a consensus of opinions and then it becomes just "well, HN said this."
This is most apparent when the forum contradicts itself. If there isn't another way to denote the opinions (like a sub-forum) then you start seeing comments like "everyone was against it yesterday and now everyone is in favor?" when in reality it's unlikely to be the same people responding anyway.
In one to many conversations like a thread on Hacker News, why would I say "I don't know" unless I'm specifically addressed or have a side point to make? It's noise, and doesn't help anyone.
Coming from academic circles, we consider such people as the "mildly" smart. As in, being uninformed is OK (you know you are uninformed). Being only a little informed is dangerous (you don't know what you don't know). And the most informed are OK too (they know quite well the limits of their knowledge). The smart people you are talking about do not usually survive academia.
In my own personal experience, I find the false confidence to be quite orthogonal to intelligence. I see plenty of low intelligence folks have the same problem, and plenty of high intelligence folks have it, and do not see a clear relation where one is more prone to it than the other. More like: You are humble and always doubt yourself (which is the academic's concept of "smart" - continually trying to tear down your own theories), or you are arrogant and do not. Intelligence is unrelated.
So, when I take tech teams into uncharted waters, the first thing we do is observe. Otherwise we create "ivory tower solutions" better suited to ourselves than actual users.
And, yes. Your're right. If you're supposed to be the "expert founder" then "I don't know" is not, generally in your vocabulary. However, I do suspect this is a male problem, more so, than female. Having said that, in this great time of social upheaval, vis a vis gender, perhaps it's a great time to learn from our sisters!
I have observed the same but I have observed the exact opposite in academia and academically oriented corporate R&D in complex technical fields.
People with academic background can surprisingly humble intellectually. They may be arrogant when defending their point of view, but if they come up with a new thing they don't know about, they admit it even if it's common thing. By constantly asking simple questions every time they can't follow up something they are able to learn constantly.
Hopefully a relevant quote:
"Emotions can hinder or uplift. We might hope that those in leadership positions possess strength and resilience, but vanity and fragile egos have sabotaged many of the businesses that I’ve worked with. Defeat is always a possibility, and not everyone finds healthy ways to deal with the stress. Each person matters. Established firms will have a bureaucracy that can ensure some stability, even when an eccentric individual is in a leadership position, but when a company consists of just two or three people, and one of them reacts neurotically to challenges, the company is doomed."
http://www.smashcompany.com/business/how-to-destroy-a-tech-s...
She also has a book, which I would recommend skipping. It adds nothing that isn’t covered in the video.
https://www.huffingtonpost.com/entry/what-goes-through-every...
After a week the monitors started to fail, because the cable to the monitor was too tight, so when they moved the monitor, the contact on the back of the monitor broke.
In sincerity, I found that interesting. Lessons noted!
Anyway, I figured it fit the subject. ;) Ricky's a natural born hacker.
Automating this sounds like a business worth several million in annual revenue, easily.
I’d had to deal with Covad before and so had my friend who used to own a mom and pop ISP. I thought the fact that someone wrote a management app just for them was funny and sad, brilliant and justified.
A few years later they sold... to Covad. D’OH!
:)
It's the just that gets me. They are too action-oriented to think about the details or ask someone who's done it before, so they start down the path of (to use the author's metaphor) buying cheap lumber and calculating the angles for cutting the stair boards. They follow through with that and finish the rest of the stairs with the same attitude.
Now the stairs are finished and installed (shipped) - they look terrible, they're wobbly and too steep and everyone who walks on them in either direction trips. Then begins the never-ending process of in-place improvements: using a sander to even out the angles, grinding down the too-long screws that stick up, pounding on warped boards to straighten them out. We can't fix the angle or the poor spacing of the steps though, because everyone's gotten perversely used to it.
When the whole house gets built this way, you live in a perpetual fixer-upper, where each component is either under revision or you've settled for it being "good enough" until enough work has been done to make everything else a little less broken.
If your building a house the best volume to surface area is going to be a sphere. Why don't we use that? Well it's a pain to work with curves. Well why not a Cube? It _
That _ is going to be a listing of the problems constraints, which most people can't simply list by thinking abstractly.
Well then don't do a shitty job? I mean if the budget is there, then yeah, just build me some stairs.
The level of abstraction you are working on matters. A manager can't just build a great business either. But for example if he encounters legal trouble, he can just hire lawyers to fix it.
His interface to the world is via delegation to people who can solve his subproblems competently and consistently. There's still a "surprising amount of detail" but at a higher abstraction level and solved with different tools(a really strong "biological" AI that can resolve and prevent the details from bubbling up :P).
The flip side of this are the people who can’t accomplish anything novel because they’re unable to simply try. At some point you have to “just” do it.
You can do research and seek out advice, and indeed you should. But for most things, you will not develop a perfect understanding or solution without jumping in and finding the practical gaps in your knowledge. You generally will have gaps and you can spend infinite time trying to fill them. Or you can learn enough to start and understand that you will likely make mistakes and learn along the way.
I’ve never done anything substantial that didn’t involve mistakes. Every major new system I’ve built had mistakes. Every home project I’ve done has imperfections that annoy me. Not once have I completed something substantial and novel and then said I’d do it exactly the same if I were doing it again. That only happens for things I’ve done repeatedly so I learn the optimal way to get it done. Build one deck and you’ll screw up at least three things. Build a dozen and you’ll get it down to a consistent product.
This is of course why trades have apprenticeship programs. You can screw up stairs a dozen times and eventually figure out all the details you missed, or you can work with someone to learn to do it correctly more efficiently. The person you’re learning from can probably fix your errors, too, so the end product is better than you’d have produced.
No one complied, of course, but I never let the sentiment die.
(Despite the name, it's great post about complexity)
The author mentioned bicycles and inadvertently missed an opportunity to show another "unknown reality" that beautifully illustrates his point.
Consider typical adults (not physicists) with 20+ years of bicycle riding. If you ask them, "how does a bicycle turn?", they'd respond that "you just lean into the direction of the turn and it turns."
But that's not the complete picture of what actually happens.[1] If you were to capture the bicyclists with a high-speed camera, you'd notice that there's always a micro movement of countersteering to make the bike momentarily lose its balance and "fall into" the turn. Then there's an immediate correction of the steering to match the turn. This all happens in the span of milliseconds.
Even though the human body "discovered" the countersteering by way of kinesthetic feedback, the bicyclists' brains don't explicitly communicate this intermediate step to others. (Tacit knowledge.[2])
It's not always the case that we "simplify" reality for innocent purposes of pedagogy. (E.g. we tell "lies" to children about "numbers", "functions", etc and as they get older, we successively remove the layers of lies as they get into high school and college math.) Instead, we often don't even know the reality (e.g. bicycle countersteering) to consciously omit it.
That extra reality may not even be important. If one human is teaching another human, you can leave the "countersteering" detail out and it won't matter. The 200 years of people learning how to ride bicycles is evidence of that. However, if you're a robotics scientist and want to build a self-driven motorcycle, the reality of countersteering becomes a crucial detail.
(I also experience the "clicking" he refers to around 6:10 when I switch between QWERTY and Dvorak layouts. It takes about 15-30 seconds to switch, even after all the years I've been doing this, and I can "feel" it slide into place. This despite the fact that I use Swype (or whatever it is spelled as) on my phone all the time, which means I stay very familiar with the QWERTY layout every day... but somehow that qualifies as a different "thing", because I always get that instantly.)
It's probably different for everyone, in my case my brain has found the look of the keyboard to be the best predictor, for others it could be something else. But it's really interesting, sometimes it takes me a while before I notice that, indeed, this keyboard looks a bit like a Mac keyboard.
When I was a kid, I had a teacher who said that cutting paper wasn't a chemical change- no bonds were broken. That seemed odd to me, but after I had a PhD, I understood a few things better. That when you cut paper, mostly hydrogen bonds- bonds, but a particularly weak form of bonds- are broken, but threre is enough kinetic energy to break some covalent bonds too. This turns out to have huge implications (besides the fact that the teacher was wrong).
Most people would say that covalent bonds hold the protein together and they dominate the overall structure- for example, you wouldn't ever see a strained covalent bond, so you can easily eliminate all structures with strained bonds. This was long an assumption. However, work I did at Google using Exacycle demonstrated that, in fact, large collections of hydrogen bonds can in fact work in concert to stabilize proteins with strained covalent bonds (http://onlinelibrary.wiley.com/doi/10.1002/pro.2389/full), which overturned the assumption, and made a modest contribution to the prediction of protein folds, and protein design.
That covalent bonds are broken seems obvious if you ditch the paper and cut through a sheet of long-chain polyethylene. The scissors don't just stop if they're about to break a polymer chain.
Certainly, if an object millimeters across has a charge of 1 electron that isn't noticeable.
This is actually a well-known fact about motorcycle, and it's taught in driving schools. It's just that their is no bicycle driving school and people just figure it by themselves. I'm pretty sure it takes more time though.
[0]http://www.msf-usa.org/downloads/BRCHandbook.pdf (sec.4, pg.8)
Anyway, learning to ride a moped was when I first consciously discovered countersteering. It made things a lot easier after that point.
When the wheel is forward of the axis of steering, you don't need to countersteer for small turns. When you turn the bars, the contact point on the wheel moves forward and to the side, dropping the front of the bike down and tilting it into the direction of the turn for you. That mimics the effect of countersteer. It's also one of the factors that makes some bikes, like mopeds, easier or harder to control.
During harder turns, the amount of lean created by the rake angle is less than required and you do need to countersteer. It's very hard to notice even when you're actively looking for it though.
For a guitarist, it's almost impossible to explain the myriad tiny details that go into something as seemingly simple as picking a note. Troy's analysis of instructional videos by great players, where they obviously don't understand what they're doing, is kind of amazing. And a real eye-opener, for those of us who have played a long time.
1. Turn the steering wheel towards the lane you wish to enter,
2. Re-center the steering wheel,
3. Counter-steer the same amount as in step 1.
Step 3 is what people tend to forget about.
[1] https://99percentinvisible.org/episode/episode-37-the-steeri...
[0] http://www.lizengland.com/blog/2014/04/the-door-problem/
[1] "I made the sounds the door creates when it opens and closes."
[2] "We could charge the player $.99 to open the door now, or wait 24 hours for it to open automatically"
Brings me the question on wheter the "reality" is the same for everybody.
We all have our personal realities that operate on their own timeline. Only when you are playing an MMO next to someone do you get a chance to notice it. But it exists both in game and in real life.
Not that I could conclusively proof that statement.
The way he does it, with the angle brackets? That's the easiest way, and by far the weakest. The strongest way is to notch out 2x12's (ideally at least 3, with one in the middle). Oh, and no step can be more than 1/8" difference in rise vs any other stair. This much harder, though.
Combine this with the math to know how many steps you need, and what angle the stairs need to lean overall (this one he touches on), and it's a real pain. Fail, and people get hurt. Your body is pretty sensitive to stairs being perfect.
That requires some futzing with the comfort formulas, and still trying to make that work with code and average-height people.
Watch the height on the first and last steps. The middle ones are the "easy" ones.
In bigger cities there are companies that only frame stairs. It is amazing to watch them - they can do it in their sleep. I watched one crew finish up some complex stairs in less time than our framing crew would have spent making a game plan.
There is detail in everything and until you've done that job you really don't notice it.
I did a concrete landing for these steps, which wasn't too bad, but my next door had his front steps re-poured by actual professionals, and it took them like 2 weeks.
Compare that to mine, which while a pain in the ass really only took about 20 hours total, and I made 7 "stringers" instead of the usual 2 or 3 (I used composite decking on the stairs, which requires support every 9-11 inches) which made it a lot harder (and heavier).
And of course, concrete steps really only look right when they go right up to the house, they don't look right attached to a deck, and they certainly aren't something you want IN your house.
1. The narrator thinks it's important to have in your toolset hammers of multiple hardness, because your hammer should be softer than whatever material you're hammering on at the time.
At least, that's how I remember it; I might well be wrong about -- well, about the DETAILS of that example. Please correct me if I am.
2. A student gets and assignment to write an essay about a particular brick wall. He can't see how to fill the assigned length with anything interesting.
The instructor then says OK, write instead only about the left-topmost brick. The student has so much to say about it he has trouble making his essay SHORT enough to meet the assignment's requirements.
But even though I recall little about its specifics, that book was very influential on my approach to epistemology.
One of them, a girl with strong-lensed glasses, wanted to write a five-hundred-word essay about the United States. He was used to the sinking feeling that comes from statements like this, and suggested without disparagement that she narrow it down to just Bozeman.
...
It just stumped him. Now he couldn’t think of anything to say. A silence occurred, and then a peculiar answer: “Narrow it down to the main street of Bozeman.” It was a stroke of insight.
...
He was furious. “You’re not looking!” he said. A memory came back of his own dismissal from the University for having too much to say. For every fact there is an infinity of hypotheses. The more you look the more you see. She really wasn’t looking and yet somehow didn’t understand this.
He told her angrily, “Narrow it down to the front of one building on the main street of Bozeman. The Opera House. Start with the upper left-hand brick.”
Her eyes, behind the thick-lensed glasses, opened wide. She came in the next class with a puzzled look and handed him a five-thousand-word essay on the front of the Opera House on the main street of Bozeman, Montana. “I sat in the hamburger stand across the street,” she said, “and started writing about the first brick, and the second brick, and then by the third brick it all started to come and I couldn’t stop. They thought I was crazy, and they kept kidding me, but here it all is. I don’t understand it.”
Neither did he, but on long walks through the streets of town he thought about it and concluded she was evidently stopped with the same kind of blockage that had paralyzed him on his first day of teaching. She was blocked because she was trying to repeat, in her writing, things she had already heard, just as on the first day he had tried to repeat things he had already decided to say. She couldn’t think of anything to write about Bozeman because she couldn’t recall anything she had heard worth repeating. She was strangely unaware that she could look and see freshly for herself, as she wrote, without primary regard for what had been said before. The narrowing down to one brick destroyed the blockage because it was so obvious she had to do some original and direct seeing.
He experimented further. In one class he had everyone write all hour about the back of his thumb. Everyone gave him funny looks at the beginning of the hour, but everyone did it, and there wasn’t a single complaint about “nothing to say.”
In another class he changed the subject from the thumb to a coin, and got a full hour’s writing from every student. In other classes it was the same. Some asked, “Do you have to write about both sides?” Once they got into the idea of seeing directly for themselves they also saw there was no limit to the amount they could say. It was a confidence-building assignment too, because what they wrote, even though seemingly trivial, was nevertheless their own thing, not a mimicking of someone else’s. Classes where he used that coin exercise were always less balky and more interested.
As a result of his experiments he concluded that imitation was a real evil that had to be broken before real rhetoric teaching could begin. This imitation seemed to be an external compulsion. Little children didn’t have it. It seemed to come later on, possibly as a result of school itself.
-- Robert Pirsig
Just sayin'
;)
I learn the damndest things every day. Thank you for a slice of history of my own home town.
It is a nice little park (and still there).
0: https://www.amazon.com/Mezzanine-Nicholson-Baker/dp/08021449...
https://kottke.org/18/01/noticing-a-new-weekly-newsletter-fr...
Lego and Minecraft fit into this. I once got my old Legos out and suddenly had dozens of ideas of what to build. Although I could've done the same in woodwork at any other time. The boundary is the most important thing to fuel creativity imho.
It's difficult, but invaluable, to truly see things from other perspectives. Once we begin learning the small details of one perspective, our minds seem to develop heuristics for quickly judging all things from that perspective. Attempting to see things from a different perspective slows us down and makes us uncomfortable.
Citation needed, but I sincerely believe our intellectual difference are just a consequence of viewing things through many layers of varied perspectives. So for people who have considered global warming and believe it to be false, it really is false. That is reality. It's all subjective.
But of course it's perfectly valid for our highly complex human scale world too. And then why do we think the Sun is what we see in pictures? Of course it's that, but it's much more too, and since we don't capture a lot of that in our models (be them photos in various EM gamuts or neutrino counts or whatever numbers), but we only see the surface, we will never experience it up close.
And the same goes for our experience of others' experiences, be it scholarly undertaking in unraveling the mysteries of Earth or simple carpentry/masonry/woodworking/sports.
Even though hundreds of millions of people watch and tens of millions play soccer, no one really has the correct model about playing it on the professional (world cup, UK Premier League) level. Yet every one has a model of soccer, and of course that's their day to day reality. (And they of course do some imperfect subconscious belief update on their models as they go through life, but that's not much compared to an actual rational inquiry using the scientific method - but who has time for that for everything in the world?)
>The idea does not necessarily imply that there is no objective truth; rather that our access to it is mediated through our senses, experience, conditioning, prior beliefs, and other non-objective factors. The implied individual world each person occupies is said to be their reality tunnel.
I've never replaced stairs but I have learned enough of the "details" by repairing them (or being aware of them) to know it's a really hard to get right.
A single 1/4 inch difference between two steps can feel wrong; a 1/2 inch difference and people will likely stumble or tumble down them on their first approach. Our intuition and pattern recognition impacts how we solve problems. Some call it "muscle memory" but it's not just in our muscles, it's in our unconscious thought.
Like the author explains so well, once we move a detail into our unconscious it becomes part of us, transparent to us, "common sense"; and an unnoticed detail is completely hidden from us.
The details of reality can only be taught through experience, because once understood, they move almost instantly from hidden -> transparent.
I think this is a similar phenomenon where the people tripping up the stairs have an imbalance causing their top half to fall forward. As a result the unconsciously speed their bottom half up to catch the top half from falling.
But then it dawned on me. Things somehow play out well in the end. Why? Well, because there I was, to care for details, to find solutions for what it seemed like an impossible problem to solve. If they cared for details as much as I did — we wouldn't even begin anything, because the amount of detail was so overwhelming. The general plan wouldn't be born in the first place.
Now I sometimes wish I could care less for details and try to be... well, easier, I guess.
An other pleasant part of software development is that all you need is a computer and a compiler/interpreter to get started. No need to buy dedicated tools or supplies. And if you realize halfway into a project that you did something wrong (that your "staircase" is not cut right) then you can easily undo and redo anything without any mess or waste. No cleanup needed either. The only limits are your skills and imagination.
Seriously, though, couldn't agree more, I've said some variation on everything you mentioned for many years. Especially the mess part. I would add the risk of personal injury, too.
Over longer time spans, the tools sometimes become outdated in terms of one's productivity while using, assuming non-artistic motivation for use.
Old motorized tools are more prone to this than mechanical-type tools. Which maybe gives some insight as to the factors that contribute to obsolescence. There are exceptions, of course, but generally my grandfather's hand tools are still perfectly serviceable while his motorized tools often have more drawbacks than benefits to use. A 50-year circular saw is as heavy as a boat anchor and is a huge pain in the ass to adjust and use compared to newer options.
Find the cause of a bug. Oh, but the fix isn't clear, because it turns out the "bug" isn't a programming error as it was - two people who had two slightly different but incompatible interpretation over the meaning and purpose of some subsystem (whose creator didn't document anything because hey, gotta meet deadlines). So now we have to decide what is the most correct way to resolve this while breaking the least amount of stuff... hmm, we could refactor this, separate that, give this a clearer purpose... oh, but surprise, it's tightly coupled to this, which is tightly coupled to that, which is tightly coupled to... FML.
If you overlook most details you are basically just 'doing stuff to get objectives done', which is meager imho.
Getting acquainted with detail takes us away from the vast sameness that we experience when we are just running on automatic and helps us cherish and savor each moment as complexity unfurls.
I learned to love talking to strangers when I realized the richness and variety of human experience is so vast that no one person can feel or experience everything another person can in their lifetime. In retrospect, it's amazing I had to realize this at all, but we all start out self-absorbed.
- The effect of a surprise when facing unexpected complications/details is because our brains tend to simplify things, or rather simplify the model of the world that we maintain. When you first hear that water boils at 100º your brain really really wants to assume it's that simple, no more than that. Simplifications are crucial for optimising the brain's power consumption, which it tends to minimize all the time.
- The amount of detail in human built stuff, as well as in our understanding of the world is a result of many centuries of perfecting and improving those things. Go a few millenia back and look at the stairs we were building then, they were awful. Or the fact that the process of boiling is so complex is a result of relatively recent discoveries in physics, just a few centuries ago.
Funny thing is, the details of the reality around us will only get more complex over time. Suppose in a few centuries from now stairs will be so complex and so perfect that no individual will be able to build them on their own.
If you're a cook, you don't really need to know all that, because you either live up in the mountains (at lower pressure) or you don't. You either have hard water (with more dissolved ions) or you don't. And your pots are never scrubbed perfectly clean.
Suppose that in a few centuries, stairs will have to account for different numbers of legs, or wheels, or different foot types, or varying amounts of gravity, or track gauge, or crystal habit, or effect on convection and ventilation, or whether classical Earth-standard humans will ever be expected to use them.
I'm not talking just about programming. At my University I used to make robot arms and RC stuff with my classmates. Polystyrene, Arduino and a lot of duck tape were our friends, winning even a NASA competition. It was actually quite fun and fairly easy.
After finishing and while I was freelancing, I tried to take it to the next level. Once you get to rigid pieces and tight couplings, there is a huge amount of details and gears that I don't even know how to start searching for. What I learned at my University (Industrial Engineering) won't work as well, since I am not ready to order 10000s of pieces.
Everybody who actually creates or makes something can agree that that stuff is hard, really hard but oh so rewarding!
The moment you take a step back, realize that a real-world project has so many details, that your team can get collectively stuck, and that you might be just flat out wrong is really humbling.
Why not ask someone who has done it before, or who does it for a living. There are many domains of human endeavour, but few with no experts. Experience is not necessarily expertise, but the experienced know what details to look for even if they don't know how to fix them.
That to me is another problem of capitalism - people reserve knowledge in order to attract finance with its value; whilst the greatest good is to share that knowledge.
Actually nominal 2 inch lumber has thickness of 1.5" [1]. So your screws should be shorter than 1.5".
[1] https://www.archtoolbox.com/materials-systems/wood-plastic-c...
And, as I've recently learned, two-by-four is in fact 1½ × 3½ inches.
How on Earth does that make sense? Why do people call things as if they were X, while knowing perfectly well they're not X? Is this an American thing?
Some more surprising detail in reality :)
And it's also why I think he paperclip maximizer and gray goo scenarios are silly. Maybe it's theoretically possible to create something that would eat the world, but in order to do so, it would have to overcome every obstacle the world throws at it. Again, easy to imagine, extremely hard to realize. All those details get in the way.
An AI is an "unlocked" brain. It can see all the things that it does and then see how it does them. If it learns to ride a bike -and if it's a humanlike AI, it will be able to- it'll be able to see exactly what it does and build a model from that. We would be able to do the exact same except we can't see inside our own brains and we have to study slow motion video and notice every little thing. The AI just sees oh, I turn the other direction slightly before initiating a turn. Voila, countersteering is discovered.
https://www.wired.com/story/when-it-comes-to-gorillas-google...
Except this AI doesn't exist. You're imagining there is such a technology, and this technology will just be able to see all the detail in the world it needs to learn, and presto it does everything better than opaque humans.
If you think deep learning is such a technology, then ask yourself to what extent ANNs understand themselves and you'll see they don't at all. They're just good at optimizing for certain problems humans are able to set them up for.
So, how will us opaque humans create such a transparent technology?
Obviously I'm imagining it. Strong AI does not yet exist. It's also obvious that it could exist, because humans do it. I'm only making two logical inferences here:
1: Future superhuman AI will have at least the capabilities of the human brain, because we know those abilities are possible, because we do them.
2. A future superhuman AI will have the ability to examine itself in memory and identify things about itself in a way that far exceeds human introspection: we can barely examine out own emotions, much less the actual neuronal contents of our heads.
> And it's also why I think he paperclip maximizer and gray goo scenarios are silly. Maybe it's theoretically possible to create something that would eat the world, but in order to do so, it would have to overcome every obstacle the world throws at it. [...] If you think deep learning is such a technology, then ask yourself to what extent ANNs understand themselves and you'll see they don't at all.
Well first off, they're quite good at it[1], but more importantly that's weak AI rather than strong AI. Arguing that weak AI is unlikely to be superhuman is plausible, but strong AI is definitely self-improving.
What I think you're saying is that you're skeptical of us being able to create strong AI out of current techniques, which is also reasonable. NN are not gonna evolve into skynet any time soon. But believing them to be categorically impossible requires the human brain to be special in some way- either beyond human comprehension or comprising a supernatural component.
[1]: https://en.wikipedia.org/wiki/Generative_adversarial_network
In context of the OP, the issue is all the detail in the world that takes many people over ages to work out. Will the superhuman AI be able to recognize all the detail it needs to know to accomplish tasks better than us (all humans)? Notice this isn't the same issue as being transparently intelligent.
Yes, if for no other reason than this: http://slatestarcodex.com/2017/11/09/ars-longa-vita-brevis/
One answer would be: generate thousands of short descriptions of input signal for its different aspects, find the ones which correlate with the outputs more, refine hypotheses using scientific method.
I hope they don't think that. Deep learning is not suited for building a proper general AI, but that's a feature of deep learning - nothing we know about in physics / information theory says that the only possible intelligence is a neural-like black box.
> So, how will us opaque humans create such a transparent technology?
The same way we identify bugs in our own thinking - by careful application of mathematical methods. It will take time.
Standardising and having a limited set of (appropriate, well-understood) choices at every level is key to building complex projects. Which is probably one of the reasons why software quality isn't really improving...
But when I'm playing around in my off time I employee alternative ways of doing things with unusual or niche languages so I can explore details in more depth.
> Baader-Meinhof is the phenomenon where one stumbles upon some obscure piece of information—often an unfamiliar word or name—and soon afterwards encounters the same subject again, often repeatedly.
It was there all the time, you just didn't noticd it.
Say you want to measure a length. Easy right? Just take a measuring tape or ruler and put it next to the thing to be measured. But what if you need really high accuracy over a reasonably long distance. Say, to within a tenth of a millimeter (100 μ) over a meter long distance or more. Now temperature matters: your scale will expand differently than the object being measured if they aren't the same material. Humidity has similar effects. Traditional length standards were metal bars, but metal bars sag under their own weight, changing the distance between the ends. Modern standards use interferometers to create a length standard, but those require very good mirrors and monochromatic light sources for proper operation. That's without getting into electronic measurement systems for the interferometers, and the required sensitivity there...
And you still need a way to transfer the accuracy from the interferometer to a practical measuring scale. That brings back concerns of temperature. Human handlers transfer heat, leading to local expansion. Light is absorbed differently by objects of different color, leading to different expansion. Etc.
The further you try to push the accuracy and precision the more details start to matter, and the more disciplines start to get involved.
[1] http://mooretool.com/publications.html (it can be found on Gen Lib, but is worth the price)
[2] https://www.tek.com/document/handbook/low-level-measurements...
[3] https://cds.linear.com/docs/en/application-note/an47fa.pdf
Edit: I guess my first line will read as a disclaimer if I'm wrong (aka missing an important detail!) or a brag if I'm right; it's intended as the former!
Let us assume that you rest the board in such a manner that the angle between the edge of the board and the vertical wall is 45 degrees. The edge of the board is tangential to the corner of the wall and the upper floor, and you scribe the horizontal line, then cut it.
When you now flip the board around so that the cut you made will rest flush against the ground, the other end will still be a line that's tangential to the 90-degree join between wall and upper floor; it won't be flush against the wall.
For a 45 degree angle, you could go "oh, I'll just scribe a vertical line upwards where the board is tangential to the corner, and cut it." When you do this, you'll find the board may be flush to the ground, but it extends past the first floor, because you basically cut it too long. You could, by careful repeated cutting, shave off enough from the bottom to get it to fit flush to the floor, and flush to the wall... but that's a lot of cutting.
Now, you could apply some simple (?) maths here, and realise that for a 45 degree angle, the answer is that the board has to be some multiple of square_root(2) long, but that's maths, and that's not allowed. :)
Back to the physical approach. What if we did that horizontal cut, then rotated the board so that it became a vertical cut? Well, now it should be flush against the wall, but the end on the ground is resting on the corner of the board, not a face. You could cut it so that the "back" (furthest from the wall) corner was flush to the ground, but now the top of the board will be a few inches lower than the upper story.
This is non-trivial :) It gets worse if you're not using a 45 degree angle which makes things symmetrical.
I should go find a bit of paper and scissors and try.
What's coming to mind next is a horizontal cut as first described (placing at the angle you want). Flip, then vertical cut, and finally trim the end with the horizontal cut by making a parallel cut offset by the amount the board extends above the upper floor. I'm sure there's a better way though. Non-trivial indeed. :)
A more complicated example:
Wait, do programmers actually think that programming is fiddly? The fiddly part is always the concerns of the real-world domain; what's special to programming is the ability to use abstractions.
Consider these pieces of eternal wisdom from our industry:
- https://www.stilldrinking.org/programming-sucks
- https://medium.com/message/everything-is-broken-81e5f33a24e1
The author suggests that similar problems are a feature of pretty much everything humans do.
I know this article is not about AI, even remotely- but, oh my dog, it so is. It's like, you can do really well if you carefully select the borders of your problem domain- ImageNet, Go, Phenix, AZ; but when you try to use the same super powerful tools in an unconstrained situation (imagine a self-driving car in Mumbai; or playing Warcraft) then all those little noisy, unpredictable, unmodellable details in the real world kick your models' accuracy to the curb.
In fact, I think most AI folks have figured this out by now and that this realisation is a very big reason why AI has advanced with leaps and bounds in recent years. But we're still up against impossible odds here.
And this should be put to the attention of the Singularitarians- you don't know what you don't know yet. It might look like things are about to go exponential, but you never know what's behind the next bend. As Solon said to Croesus, "Count no man happy until the end is known".
This is a great point. It has another corollary that people who are good X approach aren't necessarily good judges of whether approach X is good. The standard answer to "that seems like an excessively complex way to do it" is generally "that because you don't have experience with doing it the right way" and that could be right or it could be wrong. Often, the experienced and the inexperienced have about equal chance to guess on the meta-level whether this is true.
When OOP was the next big thing, a whole array of people defended against objections with arguments around those not liking it "doing it wrong", which we now, mostly, know were ridiculous right? Of course we know...
I feel like most programmers have it easy. For the most part the first step of a program is "take the universe and project it to some fixed structure". The entire universe of complexity gone.
The box has to be the right shape, but even if it's not things can mostly work.
You forget to account for wind and suddenly your New York skyscraper is falling over. Forget a field in your database and you just have a slightly busted thing.
It really depends what the software is used for. Software is just peoples thoughts and intentions codified for quick access/evaluation.
If they are used for a game it might not be important, if they are used for something important it also becomes important and exactly because the complexity of real life is difficult to account for ahead of time.
There are massive failures, and now, more than ever, software interacts with other software in unpredictable ways.
I was just expressing the feeling that things are even worse in other domains. Though sometimes automating a huge unstructured mess is difficult, imagine having a huge unstructured mess and dealing with it by hand!
Programmers might have many problems, but at least a subset of the work happens in a purely mathematical space.
During that time, I've heard on more than one occasion a phrase that goes something like this "a month in the lab saves an hour in the library".
My response, was usually quote the Car Talk guys "Reality often astonishes theory"
The irony, and subject of endless entertaining remarks on the show was that Tom Magliozzi had a PhD.
https://en.wikipedia.org/wiki/Coastline_paradox
"The coastline paradox is the counterintuitive observation that the coastline of a landmass does not have a well-defined length. This results from the fractal-like properties of coastlines."
In rare moments, designs become simpler as you progress. Irrelevant or neurotic features seem to melt away. That is bliss!
But with details alone in mind it's easy to fall into the trap of "there must be 1 more detail I'm missing" ad infinitum.
If you do anything where you have to solve problems in a fixed period of time, it's important to decide when to pivot to a new, more productive problem (depending on the scale of your problem and freedom to change).
Take this advice or learn it the hard way.
I disagree. With physical objects there is elbow grease; a reality mistake on the level of a typo is minor compared to a programming typo which is as good as having no program written at all.
This. And how you could be pointing out something so self evident and it falls on deaf ears.
> Reality has a surprising amount of detail
You can't have it both ways. The amount of details, relative to your ability to perceive them all, is no-where near "mostly".
This is a good read for either of those kinds of people.
I personally took this to an extreme... I learned to program at quite a young age, and dual-majored in EE and CS specifically so that I could understand the systems that ran my software. I'm happy writing Python, but if I need to dig into the CPython run-time to figure out a bug I'm ok with that. And if I need to run tcpdump to figure out why a TCP stream is stalling, I'm good with that. The abstractions are beautiful and work 99.9% of the time, but if you don't know how the abstractions are built, you're going to have a hard time figuring out what's wrong when they break down.
One of the hardest things to do is make difficult processes and systems accessible. This article achieves that with a simple discussion of building a staircase (no mean feat, first time) to step up to the springboard for a much larger idea. It's almost as if observing the staircase details renders a scaleable model fractal that gets to measurement of a learning curve.
I wonder if, by observing the details of one's own most difficult task, focusing on the time to learn each detail and assemble one's own model, an overall speed for learning ANY difficult task, done by a particular person could be calculated. That could be a valuable tool for measuring human potential, particularly in apprentice-type learning.
The value of the time and the effort put into the apprenticeship -- to both the apprentice and the master craftsman -- could be assessed before hand. Achievement baselines could be set closer to reality; or it could be decided that the apprenticeship would not be worth the time and effort. But that last decision could be made on firmer ground, data-wise.
Da Vinci, I would suggest, was a great observer of detail; he was also pretty good at math, science, engineering, and architecture. But it all started with drawing -- observing the most minute of detail, in all it's imperfection, then transferring that to paper. As an artist, he also learned to sort of standardize many of the imperfections -- or at least the use of the tool to represent those imperfections on paper or canvas. The painting method he galvanized, sfumato, eschewed hard lines in nature, and worked hard to achieve the soft edge. I wonder if da Vinci saw the irrationality of pi as evidence that a circle has no true "edge," but that a circle was instead equidistant points stretching out in a small infinity, particular to a specif object
The downside of observation might be wrapped up in this last, though. Because the casual observer of a circle, presuming that there is a definite, finite edge to a circle could lose their faith in the dimensional space we've grown up perceiving (four dimensions.) This might be okay for a mathematician, a scientist, or an artist. To a suburban husband with a mortgage and a minivan, it might be terrifying. The observation might completely unmoor him, and send him drifting off.
Nature is like that. The infinite diversity of nature, at say, the class, order, genus, species, or individual level can be maddening to try to take in. Looking above, at "animal, mineral, vegetable" compounds that exponentially. How those pieces work together to build a staircase, or a giraffe, or a seam of granite, or the savanna of Africa, is the heart of scientific -- and artistic -- inquiry.
Thanks, John Salvatier, for a provocative and insightful article.
Interesting, Hofstadter's original motivation for the law was the observation (published in 1979) about the then delays in computers beating grandmasters at chess (first in 1997 and now routinely_. This gives us a data point about 'just how long' things can take (in addition to the numerous examples from software projects). Nuclear fusion is still out there...
To each their own, as they say.
By the way: great post.
VR can let you experience neat things that are impossible or difficult to experience in reality. Visiting Paris means an eight hour flight and thousands of dollars; visiting Mars is currently impossible; fighting aliens may never be possible.
"Why would you want to use VR for that" is a great criticism of "Using the Toilet Simulator 2019", sure, but it's a bit of a strawman.
Or financially incapable of having the experience, or time constrained - and everyone is mortal and therefore time constrained.
I don't anticipate using VR to experience a walk in the local park, or (personally) mountain biking on the local singletrack because those are things I can do in real life with much more fidelity and have a much better experience. However, I could understand taking a walk in a park on another continent (or six other continents in one afternoon), or advising a friend whom I've raved about mountain biking to try the trail in VR before buying expensive equipment.
I'd rather have a compromise of some vast array of experiences across the world in low-detail VR, and also enjoy the limited amount of experiences I can enjoy in real life in full resolution, than to constrain myself to the small slice of the world that I can afford to physically experience.