I’m currently researching forecasting and epistemics as part of the Quantified Uncertainty Research Institute.
ozziegooen(Ozzie Gooen)
Karma: 2,059
Thanks! I think precommitement is too narrow (I don’t see dying as a precommitement). Optionality seems like a solid choice for adding. “Options” are a financial term, so something a bit more generic seems appropriate.
Are there any good words for “A modification of one’s future space of possible actions”, in particular, changes that would either remove/create possible actions, or make these more costly or beneficial? I’m using the word “confinements” for negative modifications, not sure about positive modifications (“liberties”?). Some examples of “confinements” would include:
Taking on a commitment
Dying
Adding an addiction
Golden handcuffs
Starting to rely on something in a way that would be hard to stop
Good luck!
To be clear, my Mac is connected via ethernet (but there’s no way to connect the headset with a wire). I’m really not sure how the iPhone would work, or if they support it.
I believe things are much nicer for Windows computers.
I’m hoping Apple releases their own soon, though rumors have it that the upcoming unit will be very expensive $1k to $3k).
The latency is too bad for my particular setup. It depends on how good the wifi connection is in your room.
Yea, I’ve looked into the SIP workaround, but am reluctant to implement that now. I’m hoping they just making drivers for Big Sur, but it’s taking more time than I’d like.
Hi!
I made this mistake of upgrading my computer to Big Sur, which has problems with wifidirect. I also changed rooms in the house, so now my wifi signal isn’t quite as good. This makes a very noticeable difference.
I still use this for writing here and there, but mostly I’m waiting for wifi direct support and/or better setups to come out. I’m keeping a close eye on developments.
Thanks so much, that makes a lot of sense.
Reviewing works can be tricky, because I’d focus on very different aspects when targeting different people. When describing books to potential readers, I’d focus on very different aspects than when trying to comment on how good of a job the author did to advance the topic.
In this case the main issue is that I wasn’t sure what kind of book to expect, so wanted to make that clear to other potential readers. It’s like when a movie has really scary trailers but winds up being being a nice romantic drama.
Some natural comparison books in this category are Superforecasting and Thinking Fast and Slow, where the authors basically took information from decades of their own original research. Of course, this is an insanely high bar and really demands an entire career. I’m curious how you would categorize The Scout Mindset. (“Journalistic?” Sorry if the examples I pointed to seemed negative)
I think you specifically did a really good job given the time you wanted to allocate to it (you probably didn’t want to wait another 30 years to publish), but that specific question isn’t particularly relevant to potential readers, so it’s tricky to talk about all things at once.I’d also note that I think there’s also a lot of non-experimental work that could be done in the area, similar to The Elephant in the Brain, or many Philosophical works (I imagine habryka thinks similarly). This sort of work would probably sell much worse, but is another avenue I’m interested in for future research.
(About The Village, I just bring this up because it was particularly noted for people having different expectations from what the movie really was. I think many critics really like it at this point.)
(I originally posted this to Goodreads)
TDLR: A good book with mass appeal to help people care more about being accurate. Fairly easy to read, which makes it easy to recommend to many people.
I’ve met Julia a few times and am friendly with her. I’d be happy if this book does well, and expect that to lead to a (slightly) more reasonable world.
That said, in the interest of having a Scout Mindset, I want to be honest about my impression.
The Scout Mindset is the sort of book I’m both happy with and frustrated by. I’m frustrated because this is a relatively casual overview of what I wish were a thorough Academic specialty. I felt similarly with The Life You Can Save when that was released.
Another way of putting this is that I was sort of hoping for an academic work, but instead, think of this more as a journalistic work. It reminds me of Vice Documentaries (which I like a lot) and Malcolm Gladwell (in a nice way), instead of Superforecasting or The Elephant in the Brain. That said, journalistic works have their unique contributions in the literature, it’s just a very different sort of work.
I just read through the book on Audible and don’t have notes. To write a really solid review would take more time than I have now, so instead, I’ll leave scattered thoughts.
1. The main theme of the book is the dichotomy of “The Scout Mindset” vs. “The Soldier Mindset”, and more specifically, why the Scout Mindset is (almost always?) better than the Solider Mindset. Put differently, we have a bunch of books about “how to think accurately”, but surprisingly few on “you should even try thinking accurately.” Sadly, this latter part has to be stated, but that’s how things are.
2. I was expecting a lot of references to scientific studies, but there seemed to be a lot more text on stories and a few specific anecdotes. The main studies I recall were a very few seemingly small psychological studies, which at this point I’m fairly suspect of. One small note: I found it odd that Elon Musk was described multiple times as something like an exemplar of honesty. I agree with the particular examples pointed to, but I believe Elon Musk is notorious for making explicit overconfident statements.
3. Motivated reasoning is a substantial and profound topic. I believe it already has many books detailing not only that it exists, but why it’s beneficial and harmful in different settings. The Scout Mindset didn’t seem to engage with much of this literature. It argued that “The Scout Mindset is better than the Soldier Mindset”, but that seems like an intense simplification of the landscape. Lies are a much more integral part of society than I think they are given credit for here, and removing them would be a very radical action. If you could go back in time and strongly convince particular people to be atheistic, that could be fatal.
4. The most novel part to me was the last few chapters, on “Rethinking Identity”. This section seems particularly inspired by the blog post Keep Your Identity Small by Paul Graham, but of course, goes into more detail. I found the mentioned stories to be a solid illustration of the key points and will dwell on these more.
5. People close to Julia’s work have heard much of this before, but maybe half or so seemed rather new to me.
6. As a small point, if the theme of the book is about the benefits of always being honest, the marketing seemed fairly traditionally deceiving. I wasn’t sure what to expect from the cover and quotes. I could easily see potential readers getting the wrong impression looking at the marketing materials, and there seems to be little work to directly make the actual value of the book more clear. There’s nothing up front that reads, “This book is aiming to achieve X, but doesn’t do Y and Z, which you might have been expecting.” I guess that Julia didn’t have control over the marketing.
Review/Overview Thread
I think in a more effective world, “Digital Humanities” would just be called “Humanities” :)
That sounds right. However, I think that being properly calibrated is a really big deal, and a major benefit compared to other approaches.
On the part:But I’d guess that more explicit, less “black box” approaches for predicting what Y will say will tend to either be more robust to distributional shift or more able to fail gracefully, such as recognising that uncertainty is now much higher and there’s a need to think more carefully.
If there are good additional approaches that are less black-box, I see them ideally being additions to this rough framework. There are methods to encourage discussion and information sharing, including with the Judge / the person’s beliefs who is being predicted.
Thanks for the attention on this point.
I think I’m very nervous about trying to get at “Truth”. I definitely don’t mean to claim that we were confident that this work gets us much closer to truth; more that it can help progress a path of deliberation. The expectation is that it can get us closer to the truth than most other methods, but we’ll still be several steps away.
I imagine that there are many correlated mistakes society is making. It’s really difficult to escape that. I’d love for future research to make attempts here, but I suspect it’s a gigantic challenge, both for research and social reasons. For example, in ancient Egypt, I believe it would have taken some intense deliberation to both realize that the popular religion was false, and also to be allowed to say such.
Fair points. I think that the fact that they can predict one’s beliefs is minor evidence they will be EV-positive to listen to. You also have to take into account the challenge of learning from them.
All that said, this sort of technique is fairly prosaic. I’m aiming for a future much better; where key understandings are all in optimized prediction applications and people generally pay attention to those.
This sounds roughly right to me. I think concretely this wouldn’t catch people off guard very often. We have a lot of experience trying to model the thoughts of other people, in large part because we need to do this to communicate with them. I’d feel pretty comfortable basically saying, “I bet I could predict what Stuart will think in areas of Anthropology, but I really don’t know his opinions of British politics”.
If forecasters are calibrated, then on average they shouldn’t be overconfident. It’s expected there will be pockets where they are, but I think the damage caused here isn’t particularly high.
This looks really promising to me. I hate to say though, I can’t currently afford this. I think I can solve this by launching a new paid Substack for all of QURI’s upcoming content.
I’d probably recommend this strategy to other researchers who can’t afford the costs to the LessWrong Substack, and soon the QURI Substack.
Introducing Metaforecast: A Forecast Aggregator and Search Tool
I feel mixed about this.
My guess is that Anna means something fairly specific by “honor”, but there are many cases of people using honor or similar abstractions to justify some really terrible things (lots of violence, for example). So if you were to tell most people to “maximize honor instead of do PR”, I could see this going quite poorly.
https://en.wikipedia.org/wiki/Culture_of_honor_(Southern_United_States)
For one thing, for many people, in many important situations, “not saying anything at all” is a really good thing. Think of prisoners who don’t plead the fifth, or many other legal cases or otherwise. Arguably Trump and Elon Musk have been fairly damaging on Twitter to themselves.
I think a lot of PR professionals are quite bad, but this is true for most professions. I imagine in a lot of (good) cases their advice is “don’t say really stupid stuff”, and much of the time their clients really could use hearing that.
Forecasting Prize Results
Happy to see enthusiasm and initiative here. I tried starting a few web startups in college and after; almost none worked, but the experience was useful.
I think you’re massively underestimating the challenge of this. There were many products like Kickstarter before Kickstarter. To be successful you often need a combination of technical talent, marketing, and luck. This isn’t to say that it’s not worth a more intense endeavor; just that I wouldn’t be as optimistic as your post sounds.
You sound quite similar to other young entrepreneurs I have known. Some wind up being successful, sometimes in part because of their overconfidence. Entrepreneurial claims often work better in other markets than rationalism / EA; when posting in places like this especially I suggest keeping things a bit more humble.
So overall I think this is quite a bit more challenging than I think you think it is, but I don’t want to discourage you from trying (though I really hope the process doesn’t involve you burning over-hyped colleagues). I imagine more people making attempts at these things could be quite good overall.
Fixed
A bit more info;
I lived at 20Mission, which was technically an SRO. I enjoyed the setting quite a bit, though I’ve heard they’ve had trouble recently with COVID. That said, most of the other SROs I know of nearby (in the Mission, SF), are really not nice places. (lots of drugs and some violence).
https://www.20mission.com/
There’s been discussion of having “Micro-Units” in SF, but they’re heavily regulated. It seems like small progress is being made.
https://socketsite.com/archives/2012/11/microunits_approved_for_san_francisco_capped_for_market.html