I’ve seen this quote before and always find it funny because when I read Greg Egan, I constantly find myself thinking there’s no way I could’ve come up with the ideas he has even if you gave me months or years of thinking time.
NaiveTortoise
Karma: 843
Sorry I was unclear. I was actually imagining two possible scenarios.
The first would be deeper investigation reveals that recent progress mostly resulted from serendipity and lucky but more contingent than we expected historical factors. For example, maybe it turns out that the creation of industrial labs all hinged on some random quirk of the Delaware C-Corp code (I’m just making this up to be clear). Even though these factors were a fluke in the past and seem sort of arbitrary, we could still be systematic about bringing them about going forward.
The second scenario is even more pessimistic. Suppose we fail to find any factors that influenced recent progress—it’s just all noise. It’s hard to give an example of what this would look like because it would look like an absence of examples. Every rigorous investigation of a potential cause of historical progress would find a null result. Even in this pessimistic world, we still could say, “ok, well nothing in the past seemed to make a difference but we’re going to experiment to figure out things that do.”
That said, writing out this maximally pessimistic case made me realize how unlikely I think it is. It seems like we already know of certain factors which at least marginally increased the rate of progress, so I want to emphasize that I’m providing a line of retreat not arguing that this is how the world actually is.
Isn’t it both possible that it’s a fluke and also that going forward we can figure out mechanisms to promote it systematically?
To be clear, I think it’s more likely that not that a nontrivial fraction of recent progress has non fluke causes. I’m just also noting that the goal of enhancing progress seems at least partly disjoint from whether recent progress was a fluke.
Yep, clicking “View this email in browser” allowed me to read it but obviously would be better to have it fixed here as well.
Thanks for your reply! I largely agree with drossbucket’s reply.
I also wonder how much this is an incentives problem. As you mentioned and in my experience, the fields you mentioned strongly incentivize an almost fanatical level of thoroughness that I suspect is very hard for individuals to maintain without outside incentives pushing them that way. At least personally, I definitely struggle and, frankly, mostly fail to live up to the sorts of standards you mention when writing blog posts in part because the incentive gradient feels like it pushes towards hitting the publish button.
Given this, I wonder if there’s a way to shift the incentives on the margin. One minor thing I’ve been thinking of trying for my personal writing is having a Knuth or Nintil style “pay for mistakes” policy. Do you have thoughts on other incentive structures to for rewarding rigor or punishing the lack thereof?
I’d be curious what, if any, communities you think set good examples in this regard. In particular, are there specific academic subfields or non-academic scenes that exemplify the virtues you’d like to see more of?
Makes sense—would some of the early posts about Focusing and other “lower level” concepts you reference here qualify? If you create a tag, people (including maybe me) could probably help curate!
I can make a longer comment there if you’d like but personally I wasn’t that bothered by the dreams example because I agreed with you that confabulation in the immediate moments after I woke up didn’t seem like a huge issue. As a result, I was definitely interested in seeing more posts from the meditative/introspective angle even if they just expanded upon some of these moment-to-moment habits with more examples and detail. Unfortunately, that would at least partly require writing more posts rather than pure curation.
Great post (or talk I guess)!
Two “yes, and...” add-ons I’d suggest:
Faster tool development as the result of goal-driven search through the space of possibilities. Think something like Ed Boyden’s Tiling Tree Method semi-automated and combined with powerful search. As an intuition pump, imagine doing search in the latent space of GPT-N, maybe fine tuned on all papers in an area’s, embeddings.
Contrary to some of the comments from the talk, I weakly suspect NP-hardness will be less of a constraint for narrow AI scientists than it is for humans. My intuition here comes from what we’ve seen with protein folding and learned algorithms where my understanding is that hardness results limit how quickly we can do things in general but not necessarily on the distributions we encounter in practice. I think this is especially likely if we assume that AI scientists will be better at searching for complex but fast approximations than humans are. (I’m very uncertain about this one since I’m by no means an expert in these areas.)
Did you ultimately decide not to continue this series?
Yes I can relate to this!
Yep.
Thanks, this framing is helpful for me for understanding how these things can be seen to fit together.
This is basically my perspective but seems contrary to the perspective in which most problems are caused by internal blockages, right?
Thanks for replying and sharing your post. I’d actually read it a while ago but forgotten how relevant it is to the above. To be clear, I totally buy that if you have crippling depression or even something more mild, fixing that is a top priority. I also have enjoyed recent posts on and think I understand the alignment-based models of getting all your “parts” on board.
Where I get confused and where I think there’s less evidence is that the unblocking can make it such that doing hard stuff is no longer “hard”. Part of what’s difficult here is that I’m struggling to find the right words but I think it’s specifically claims of effortlessness or fun that seem less supported to me.
I keep seeing rationalist-adjacent discussions on Twitter that seem to bottom out with the arguments of the general (very caricatured, sorry) form: “stop forcing yourself and get unblocked and then X effortlessly” where X equals learn, socialize, etc. In particular, a lot of focus seems to be on how children and adults can just pursue what’s fun or enjoyable if they get rid of their underlying trauma and they’ll naturally learn fast and gravitate towards interesting (but also useful in the long term) topics, with some inspiration from David Deutsch.
On one hand, this sounds great, but it’s so foreign to my experience of learning things and seems to lack the kind of evidence I’d expect before changing my cognitive strategies so dramatically. In fairness, I probably am too far in the direction of doing things because I “should”, but I still don’t think going to the other extreme is the right correction.
In particular, having read Mason Currey’s Daily Rituals, I have a strong prior that even the most successful artists and scientists are at risk of developing akrasia and need to systematize their schedules heavily to ensure that they get their butts in the chair and work. Given this, what might convince me would be if someone catalogue of thinkers who did interesting work, quotes or stories that provide evidence that they did what was fun, and counter-examples to show that it’s not cherry-picking.
The above is also is related to the somewhat challenged but still I think somewhat valid idea that getting better at things requires deliberate practice, which is not “fun”. This also leads me to a subtle point which is that I think “fun” may be being used in a non-standard way by people who claim that learning can always be “fun”. I.e. I can see how even not necessarily enjoyable in the moment practice can be something one values on reflection, but this seems like a misuse of the term “fun” to me.
For what it’s worth, this is very true for me as well.
I’m also reminded of a story of Robin Hanson from Cryonics magazine:
Robin’s attraction to the more abstract ideas supporting various fields of interest was similarly shown in his approach – or rather, lack thereof – to homework. “In the last two years of college, I simply stopped doing my homework, and started playing with the concepts. I could ace all the exams, but I got a zero on the homework… Someone got scatter plots up there to convince people that you could do better on exams if you did homework.” But there was an outlier on that plot, courtesy of Robin, that said otherwise.
Meta: this project is wrapping up for now. This is the first of probably several posts dumping my thought-state as of this week.
Moving on to other things?
A thing I really structured to capture was that “i did actual research and had actual models for why masks would help against covid, but it’s still not type-3”, which is why “know why” doesn’t feel right to me.
I share this feeling based on my understanding of the boundaries you’re trying to draw.
I tentatively think that some of what you’re calling engineering knowledge would fit into what I call scientific (which is a strike agains the names), and/or that I didn’t do a good enough job explaining why engineering knowledge is useful.
Yeah, like I said, I don’t want to bike shed over terms but I do think the distinction between science and engineering is an interesting one. Regarding the “and/or” isn’t it more like there’s often an interplay between the levels? First we get “folk knowledge” and then we “do science” to understand it and then we use that science to “engineer” it?
Also, I found an Overcoming Bias post where Robin quotes Drexler on the distinction:
The essence of science is inquiry; the essence of engineering is design. Scientific inquiry expands the scope of human perception and understanding; engineering design expands the scope of human plans and results. …
Scientists seek unique, correct theories, and if several theories seem plausible, all but one must be wrong, while engineers seek options for working designs, and if several options will work, success is assured. Scientists seek theories that apply across the widest possible range (the Standard Model applies to everything), while engineers seek concepts well-suited to particular domains (liquid-cooled nozzles for engines in liquid-fueled rockets). Scientists seek theories that make precise, hence brittle predictions (like Newton’s), while engineers seek designs that provide a robust margin of safety. In science a single failed prediction can disprove a theory, no matter how many previous tests it has passed, while in engineering one successful design can validate a concept, no matter how many previous versions have failed. ..
Simple systems can behave in ways beyond the reach of predictive calculation. This is true even in classical physics. …. Engineers, however, can constrain and master this sort of unpredictability. A pipe carrying turbulent water is unpredictable inside (despite being like a shielded box), yet can deliver water reliably through a faucet downstream. The details of this turbulent flow are beyond prediction, yet everything about the flow is bounded in magnitude, and in a robust engineering design the unpredictable details won’t matter. …
The reason that aircraft seldom fall from the sky with a broken wing isn’t that anyone has perfect knowledge of dislocation dynamics and high-cycle fatigue in dispersion-hardened aluminum, nor because of perfect design calculations, nor because of perfection of any other kind. Instead, the reason that wings remain intact is that engineers apply conservative design, specifying structures that will survive even unlikely events, taking account of expected flaws in high-quality components, crack growth in aluminum under high-cycle fatigue, and known inaccuracies in the design calculations themselves. This design discipline provides safety margins, and safety margins explain why disasters are rare. …
The key to designing and managing complexity is to work with design components of a particular kind— components that are complex, yet can be understood and described in a simple way from the outside. … Exotic effects that are hard to discover or measure will almost certainly be easy to avoid or ignore. … Exotic effects that can be discovered and measured can sometimes be exploited for practical purposes. …
When faced with imprecise knowledge, a scientist will be inclined to improve it, yet an engineer will routinely accept it. Might predictions be wrong by as much as 10 percent, and for poorly understood reasons? The reasons may pose a difficult scientific puzzle, yet an engineer might see no problem at all. Add a 50 percent margin of safety, and move on.
Yeah good point—given generous enough interpretation of the notebook my rejection doesn’t hold. It’s still hard for me to imagine that response feeling meaningful in the context but maybe I’m just failing to model others well here.