The Relativity of Wrong
Collection of seventeen essays on science "The Moon and We" (April 1986) "The Minor Objects" (May 1986) "The Second Lightest" (June 1986) "Labels on the Molecules" (July 1986) "The Consequences of Pie" (August 1986) "The Enemy Within" (September 1986) "The Relativity of Wrong" (October 1986) "The Unmentionable Planet" (November 1986) "The Dead-End Middle" (December 1986) "Opposite!" (January 1987) "Sail On! Sail On!" (February 1987) "The Incredible Shrinking Planet" (March 1987) "The Light-Bringer" (April 1987) "Beginning With Bone" (May 1987) "New Stars" (June 1987) "Brightening Stars" (July 1987) "Super-Exploding Stars" (August 1987)
Stats for nerds
34recommend
80mentions
54people
2010–2026years
57% recommend41% neutral1% critical
Mentions per year
What Hacker News says
- See the relativity of wrong by Asimov for a simple development of this theme.
Can it help? Of course! But I think the question is too vague here and you're (presumably) unintentionally creating a false dichotomy. I'll clarify with the next responses> Why not? Why can't faster typing help us understand the problem faster?
The main problem is that solution spaces are very large. There are two general ways to narrow the solution space: directly and indirectly. Directly by things like thinking hard, digging down, and "zooming in". Indirectly by things such as figuring out what not to do (ruling things out).> Why can't we figure out the right thing faster by building the wrong thing faster?You can build a lot of wrong things that don't also help you narrow that solution space. The most effective way to "build the wrong thing" in an informative way is to first think hard and understand your solution space. You want to build the right wrong thing. The thing that helps you rule out lots of stuff. But if you are doing it randomly then you aren't doing this effectively and probably wasting a lot of time. You probably are already doing this but not thinking too much about this explicitly, but if you think explicitly you'll improve on this.
You always build the "wrong" thing. But it is about how wrong you are. Despite being about physics, I think Asimov's Relativity of Wrong[0] (short essay) is pretty relevant here and says everything I want to say but better. It is worth the read and I come back to it frequently.> Presumably we were gonna build the wrong thing either way
Yes! But this is not quite the same thing. I do this too! I never know the full details of the thing I want before I start building. I'm not sure that's even possible tbh. You're always going to discover more things as you get into the details and nuance. But that doesn't mean foresight is useless either.> I often build something to figure out what I want
Let's say I'm somewhere in the middle of America and I want to get to NYC. Analogous to your framing you are saying "why can't moving faster help us get there faster?" Obviously it can! BUT speed is meaningless without direction. You don't want speed, you want velocity. If you start driving as fast as possible in a random direction you're equally likely to head in a direction that increases your distance than one that decreases. And you are very unlikely to head in a good direction. Driving fast in the wrong direction significantly increases harm than were you to drive slowly in the wrong direction.Analogy -------So what's the optimal strategy? Find a general direction (e.g. use the sun or stars/moon) to figure out where "east"(ish) is, start driving relatively slowly, refine your direction as you get more information about the landscape, speed up as you gain more information. If you can't find a general direction you should slowly meander, carefully taking in how the landscape/environment is changing. If it is very unchanging, then yeah, speedup, but only until you find a region that becomes more informative, then repeat the process.
If we already had perfect information about how to get to NYC then just drive as fast as fucking possible. But if we don't have that information we need a completely different strategy. Thus, t̶y̶p̶i̶n̶g̶ driving speed isn't the bottleneck.
Speed doesn't matter, velocity does
[0] https://hermiene.net/essays-trans/relativity_of_wrong.html
- Asimov and Feynman also spoke about similar things (along with many others)
In 1980, Asimov famously wrote The Cult of Ignorance[0], criticizing the rise of anti-intellectualism. Where there was a strong political push of "don't trust the experts". He criticizes claims that sound familiar today "America has a right to know" on the basis of this being meaningless without literacy. He clarifies that literacy is far more than being able to actually read words on a page, but to interpret and process them. Asimov isn't being pretentious, his definition is consistent with how we determine reading levels[2] and his critique would be that most people do not have that of a Freshman in High School. Hell, it is even in his fiction! It is even in The Foundation and is literally the premise of Profession[3].
Feynman is a bit more scattered, but I think his discussion about the education system in Brazil (in the 50's) says a lot[4]. He talks a lot about how the students could recite the equations, ace all the tests, and achieve everything that looks to be, at least on paper, perfectly academic; but how the students did not really have the deeper understanding of the equations. It is a discussion about literacy. Were he around today I'm sure he'd use the phrase "metric hacking". Anyone that knows Feynman may also be thinking about his Cargo Cult Science[5](a commencement speech at Cal Tech (1974)). This is where his famous quote
comes from. But there is a lot of important context surrounding this and it is worth knowing about.The first principle is that you must not fool yourself--and you are the easiest person to fool. So you have to be very careful about that.[0] Note: 1980 was an election year, and one with a sweeping victory...[1] https://people.bath.ac.uk/mnsbr/papers/Asimov-Newsweek-Janua...
[1] https://en.wikipedia.org/wiki/1980_United_States_presidentia...
[2] https://nces.ed.gov/nationsreportcard/reading/achieve.aspx
[3] Profession has been in discussion lately, directly relating to this topic. If you haven't read it I'll say it is one of my favorite's of his. Not as good as Foundation but up there with Nightfall https://en.wikipedia.org/wiki/Profession_(novella)
[4] https://enlightenedidiot.net/random/feynman-on-brazilian-edu...
[5] https://sites.cs.ucsb.edu/~ravenben/cargocult.html
[Edit]
I wanted to add Asimov's The Relativity of Wrong. Sometimes I feel it should be required reading before arguing on the internet. I find myself coming back to read it at least once a year
- I think The Relativity of Wrong (Asimov, https://hermiene.net/essays-trans/relativity_of_wrong.html) is a nice counterpoint to this. Sure, we may say everyone is an idiot in some area. But there are relative levels of idiocy, and there are basic tasks you can sort of master. At the very least you can minimize your own idiocy if never eliminate it. I think mastering most essential areas in life can make you unworthy of the title of 'idiot', at least not overwhelmingly so.
There are infinitely many things to know, but not all of them are important. Knowing finitely many things (which is all we can do) can still keep us alive and well, at least for a while. And we can know some of those finitely many things increasingly well, if never perfectly.
Just as an example, if you manage say your personal finances pretty well, your health pretty well, perform any civic duties you might have, maybe do some social good or social work or charity etc., if your relationships are reasonably agreeable, respectful and pleasant, etc. and if you have a good amount of joy or peace or satisfaction, etc. in your life, then I wouldn't call you an idiot. This is not an impossible ask to know infinitely many things or infinitely precisely.
And we can learn it over 30 or 40 years, or more, prioritizing the most essential first.
Moreover, I'd say whether you can be called an idiot is context-dependent. If you get a typical (non-idiot) person, and put him in a highly specific job (which he isn't qualified for), say manager of inspectors of nuclear power plants, then he might behave like an idiot; in this case the best ability is probably the meta-ability to recognize one's own limitations and refuse work you're not qualified enough for.
Like, any person (literally any person) can theoretically be put in a situation that he might do significant harm or something stupid, this just means we have to work in contexts and understand and do well within said context; we could only legitimately be called idiots while failing badly or unethically within a canonical chosen context.
I really just don't think it's generally a good idea to go around calling ourselves (or anyone else) idiots. Too broad, derogatory, and tries to put an irremovable label on a person, which as I've explained, almost never deserves such an absolute classification.
- I understand the spirit of what you're saying, but I think this phrasing can be abused by burn-it-all-down skeptics, and I would prefer to say the notion of forward iteration depends on there being some such real thing as truth. The "relativity of wrong" essay by Isaac Asimov, recently upvoted on HN, captures the idea pretty well imo.
- As a ML researcher and a degree holding physicist, I'm really hesitant to use the words "understanding" and "describing" (much less hesitant) around these models. I don't find the language helpful and think it's mostly hateful tbh.
The reason we use math in physics is because of its specificity. The same reason coding is so hard [0,1]. I think people aren't giving themselves enough credit here for how much they (you) understand about things. It is the nuances that really matter. There's so much detail here and we often forget how important they are because it is just normal to us. It's like forgetting about the ground you walk upon.
I think something everyone should read about is Asimov's "Relativity of Wrong"[2]. This is what we want to see in these systems if we want to start claiming they understand things. We want to see them to deduction and abduction. To be able to refine concepts and ideas. To be able to discover things that are more than just a combination of things they've ingested. What's really difficult here is that we train these things on all human knowledge and just reciting back that knowledge doesn't demonstrate intelligence. It's very unlikely that they losslessly compress that knowledge into these model sizes, but without very deep investigation into that data and probing at this knowledge it is very hard to understand what it knows and what it memorizes. Really, this is a very poor way to go about trying to make intelligence[3], or at least making intelligence and ending up knowing it is intelligent.
To really "understand" things we need to be able to propose counterfactuals[4]. Every physics statement is a counterfactual statement. Take F=ma as a trivial example. We can modify the mass or the acceleration to our heart's content and still determine the force. We can observe a specific mass moving at a specific acceleration and then ask the counterfactual "what if it was twice as heavy?" (twice the mass). *We can answer that!* In fact, your mental model of the world does this too! Yo may not be describing it with math (maybe you are ;) but you are able to propose counterfactuals and do a pretty good job a lot of the time. Doesn't mean you always need to be right though. But the way our heads work is through these types of systems. You daydream these things, you imagine them while you play, and all sorts of things. This, I can say, with high confidence, is not something modern ML (AI) systems do.
A good example of lack of understanding is the image OP uses. Not only does the right have the wrong number of fingers but look at the keys on the keyboard. It does not take much understanding to recognize that you shouldn't have repeated keys... the configuration is all wonky too, like one of those dreams you can immediately tell is a dream[5]. I'd also be willing to bet that the number of keys doesn't align to the number of markers and definitely the sizing looks off. The more you look at it the worse it gets, and that's really common among these systems. Nice at a quick glance but DEEP in the uncanny valley at more than a glance and deeper the more you look.== Edit ==[0] https://youtube.com/watch?v=cDA3_5982h8
[1] Code is math. There's an isomorphism between Turing complete languages and computable mathematics. You can look more into my namesake, church, and Turing if you want to get more formal or wait for the comment that corrects a nuanced mistake here (yes, it exists). Also, note that physics and math are not the same thing, but mathematics is unreasonably effective (yes, this is a reference).
[2] https://hermiene.net/essays-trans/relativity_of_wrong.html
[3] This is a very different statement than "making something useful." Without a doubt these systems are useful. Do not conflate these
[4] https://en.wikipedia.org/wiki/Counterfactual_thinking
[5] Yes, you can read in dreams. I do it frequently. Though on occasion I have lucid dreamed because I read something and noticed that it changed when I looked away and looked back.
