Thanks for your reply! I agree with a lot of what you said.

First off, thanks for bringing up the point about underlying principles. I agree that there are often underlying principles in many domains and that I also really like to find unity in seeming messiness. I used to be of the more extreme view that principles were in some sense more important than the details, but I've become more skeptical over time for two reasons.

  1. From a pedagogy perspective, I've personally never had much luck learning principles without having a strong base of practice & knowledge. That said, when I have that base, learning principles helps me improve further and is satisfying.
  2. I've realized over time how much of action (where action can include thinking) is based upon a set of non-verbal strategies that one learns through practice and experimentation even in seemingly theoretical domains. These strategies seem to be the secret sauce that allow one to act fluently but seem meaningfully different from the types of principles people often discuss.

Another way to phrase my argument is that principles are important but very hard to transfer between minds. It's possible you agree and I'm just belaboring the point but I wanted to make it explicit.

One concrete example of the distinction I'm drawing is something called the "What Are Monads Fallacy" in the Haskell community where people try to explain monads by conveying their understanding of what mondas really are even though they learned about monads by just using them a bunch which lead to them later developing a higher level understanding of them. This reflects a more general problem where experts often struggle to teach to novices because they don't realize that their broad understanding is actually founded upon lower level understanding of a lot of details.

I think I would like it if LessWrong had more engineers/inventors as role models and that it's something of an oversight that we don't. Yet I also feel like John Carmack probably probably isn't remotely near the level of Pearl (I'm not that familiar Carmack's work): pushing forward video game development doesn't compare to neatly figuring what exactly causality itself is.

I tentatively agree, but it's pretty hard to draw comparisons. From an insight perspective, I agree that Pearl's work on Bayes Nets and Causality were probably more profound that anything Carmack came up with. From an economic perspective though, Carmack had a massive, albeit indirect, impact on the trajectory of the computing world. By coming up with new algorithms and techniques for 3D game rendering at a time when people had basically no idea how to render 3D games in realtime, Carmack drove the gaming industry forward, which certainly contributed to development of better GPUs and processors as well. Carmack was also the person at Id who insisted on making their games moddable and releasing their game engines, which eventually lead to the development of games like Half-Life.

That said, a better point of comparison to Pearl is probably Jeff Dean, who, in close collaboration with Sanjay Ghemawat, first wrote much of Google's search stack from scratch after it starting failing to scale and then subsequently invented BigTable, MapReduce, Spanner, and Tensorflow!

There might be something like all truly monumental engineering breakthroughs depended on something like a "scientific" breakthrough. Something like Faraday and Maxwell figuring out theories of electromagnetism is actually a bigger deal than Edison(/others) figuring out the lightbulb, the radio, etc. There are cases of lauded people who are a little more ambiguous on the science/engineer dichotomy. Turing? Shannon? Tesla?

Agree that science tends to be upstream of later technology developments, but I would emphasize that there are probably cases where without great engineers, the actual applications never get built. For example, there was a large gap between us understanding genes fairly well and being able to sequence and, more recently, synthesize them.

Shockley et al with the transistor seems kind of like an engineering breakthrough, and seems there could be love for that.

I agree with this and would add Lynn Conway, who invented VLSI, one of the key enablers of the modern processor industry and Moore's Law.

A little on my background: I did an EE degree which was very practical focus. My experience is that I was taught how to do apply a lot of ehttps://www.lesswrong.com/shortformquations and make things in the lab, but most courses skimped on providing the real understanding that left me overall worse as an engineer.

To be clear, I shared this frustration with the engineering curriculum. I started as a Computer Engineering major and switched to CS because I felt like engineering was just a bag of unmotivated tricks whereas in CS you could understand why things were the way they were. However, part of the reason I liked CS's theory was because it was presented in the context of understanding algorithms.

As a final point, I don't think I did a good job of my original post of emphasizing that I'm pro-understanding and pro-theory! I mostly endorse the saying, "nothing is so practical as a good theory." My perceived disagreement is more around how much I trust/enjoy theory for its own sake vs. with an eye towards practice.

NaiveTortoise's Short Form Feed

by NaiveTortoise 1 min read11th Aug 201885 comments

In light of reading Hazard's Shortform Feed -- which I really enjoy -- based on Raemon's Shortform feed, I'm making my own. There be thoughts here. Hopefully, this will also get me posting more.