What are LessWrong's thoughts on Venkatesh Rao, Gregory Rader, and Daniel Lemire?

post by InquilineKea · 2011-07-03T23:00:28.606Z · score: 8 (14 votes) · LW · GW · Legacy · 23 comments

Venkatesh Rao (who is amazing because he literally constantly challenges our definitions of everything):

http://www.quora.com/Venkatesh-Rao/Quora-Portfolio-Year-1

http://www.quora.com/Venkatesh-Rao

http://www.ribbonfarm.com/be-slightly-evil/

http://www.ribbonfarm.com

Gregory Rader:

http://www.quora.com/Gregory-Rader

http://onthespiral.com/

Daniel Lemire:

http://www.quora.com/Daniel-Lemire and http://lemire.me/blog/ are also super-super-super interesting

==

I tried doing google searches of site:lesswrong.com + their names (or websites), but ended up with little. I'd like to see what LessWrong thinks of Rao in particular. I have NEVER seen posts that were as amazingly insightful as his. It's worth it just to sacrifice a day just to see all his amazing posts. The others were also amazing and really make you think about everything.

==

E.g. with Rao, you have

My favorites: 

http://www.quora.com/What-are-the-benefits-to-obtaining-a-PhD-in-the-field-of-Education/answer/Venkatesh-Rao?srid=0WH

http://www.ribbonfarm.com/2010/05/19/intellectual-gluttony/

http://www.ribbonfarm.com/2011/04/07/extroverts-introverts-aspies-and-codies

http://www.quora.com/What-careers-or-industries-are-the-most-meritocratic/answer/Venkatesh-Rao?srid=0WH

http://www.ribbonfarm.com/2008/09/11/how-to-measure-information-work

http://www.quora.com/What-are-some-tips-for-advanced-writers/answer/Venkatesh-Rao

http://www.quora.com/Children/Why-do-some-humans-not-want-children/answer/Venkatesh-Rao

http://www.quora.com/What-is-the-best-way-to-motivate-oneself-to-finish-a-PhD/answer/Venkatesh-Rao

http://www.quora.com/History/What-is-the-most-important-human-decision-ever-made/answer/Venkatesh-Rao

http://www.quora.com/Is-the-following-follower-model-the-new-standard-of-social-connection/answer/Venkatesh-Rao

http://www.quora.com/What-discipline-specific-principles-help-reframe-activities-so-as-to-provide-useful-insights-and-or-improvement

  1. "Premature optimization is the root of all evil" -- Knuth, computer science
  2. Never design a law with the worst case in mind -- law/legislation

Rader:

http://onthespiral.com/stop-wasting-time-and-effort-developing-fragile-capabilities

http://onthespiral.com/unifying-value-universe

http://onthespiral.com/principles-disruptive-learning-environments

Lemire:

http://lemire.me/blog/archives/2010/08/16/working-long-hours-is-stupid

http://www.daniel-lemire.com/blog/archives/2010/02/08/trading-latency-for-quality-in-research

http://www.daniel-lemire.com/blog/archives/2009/09/14/how-things-change-cheaters-are-innovators

http://www.daniel-lemire.com/blog/archives/2008/08/19/the-secret-to-intellectual-productivity/

http://www.daniel-lemire.com/blog/archives/2010/01/13/the-fundamental-properties-of-computing/

http://www.daniel-lemire.com/blog/archives/2007/12/05/formal-definitions-are-less-useful-than-you-think/

http://www.daniel-lemire.com/blog/archives/2007/11/19/directed-research-is-useless/

http://lemire.me/blog/archives/2006/08/09/big-schools-are-not-longer-giving-researchers-an-edge/

http://lemire.me/blog/archives/2011/06/23/probabilities-are-unnecessary-mathematical-artifacts

http://lemire.me/blog/archives/2011/06/14/the-language-interpreters-are-the-new-machines

http://www.daniel-lemire.com/blog/archives/2009/08/01/why-i-hardly-ever-blog-about-my-ongoing-research/

http://lemire.me/blog/archives/2011/06/08/is-wikipedia-anti-intellectual

http://www.daniel-lemire.com/blog/archives/2008/06/05/why-pure-theory-is-wasteful

http://lemire.me/blog/archives/2011/06/06/why-i-still-program

23 comments

Comments sorted by top scores.

comment by knb · 2011-07-05T07:36:10.707Z · score: 15 (8 votes) · LW(p) · GW(p)

Venkatesh Rao is a fun writer, but I wouldn't take his ideas very seriously. He tends to rely on broad associations, and tells "grand stories" using fuzzy categories, without a lot (or usually any) statistics to back up his rather broad and extraordinary claims. He also tends to name-drop famous philosophers, representing their ideas in ways one might generously call "heterodox".

In other words, he seems like a well-read and verbally intelligent "narrative-builder" like Freud or Marx, and about as accurate.

comment by Normal_Anomaly · 2011-07-04T01:05:46.718Z · score: 14 (23 votes) · LW(p) · GW(p)

I don't have the time or energy to read anywhere near this many links. Please replace the majority of the links with a summary of these people's ideas, views, and/or ways of thinking.

comment by InquilineKea · 2011-07-04T03:09:37.863Z · score: 2 (2 votes) · LW(p) · GW(p)

Okay. The one thing is that these writers are so diverse that it's hard to summarize them all in one post.

But basically, they're all contrarians who question the very models that most people (including academics) follow. Basically, they're all about taking risks, short incremental bursts of productivity, and economics based on non-financial principles of values. Are they right on everything? Who knows. But it's refreshing to see what they say.

E.g. here's one of Venkat's good quotes (I can't take too much) - from http://www.quora.com/What-careers-or-industries-are-the-most-meritocratic/answer/Venkatesh-Rao?srid=0WH :

This question is deeply misguided. "Meritocratic" usually indicates some rigid and static notion of merit, usually based on testing/scales/numbers games in mature and professionalized domains. So highest scores, most papers/patents, things like that. Even "highest customer satisfaction scores" is a meritocracy scale.

Being meritorious in this sense has very low correlation with being effective in most fields. "Merit" is entirely about individual qualities, "effectiveness" is about the quality of an individual in a given setting. Merit is neither necessary, nor sufficient for effectiveness. It is often a major source of ineffectiveness.

The clause "where luck, nepotism, etc. are minimized" shows where the question is going wrong.

Luck is good. Hire people who know how to get lucky. People with good radars, peripheral vision, daring and an opportunistic streak. They are rarely "meritorious" in any measurable way, but are highly effective. The meritorious types hate them, because their influence and rewards can seem disproportionate to their effort. These lucky, opportunistic types are often amateurish bumblers when evaluated purely on various scales of professional "merit."

They may even have NO personal qualities that contribute to their effectiveness: they may just happen to know one useful thing by accident. Would you give up a treasure-hunting expedition simply because the person who has the map is a happy-go-lucky drunk who happened to find it in a dumpster by accident while looking for food? Judge people by what they can do, not who they are. As Forrest Gump told Bubba's family, "stupid is, as stupid does."

http://onthespiral.com/wp-content/uploads/2011/06/Value-Universe21.png (from http://onthespiral.com/unifying-value-universe) basically summarizes one of Greg Rader's main points.

Summarizing quote:

The conception of attention economy presented above unites two definitions that superficially contradict each other. In marketing conversations, attention economy refers to the usage of various tactics that convert attention into sales (represented in the image below by the vertical arrow). In discussions of peer-to-peer social media, attention economy refers to analogous tactics that empower individuals to convert weak ties into strong relationships (represented by the horizontal arrow).

And here is one of Lemire's excellent points (http://www.daniel-lemire.com/blog/archives/2009/09/14/how-things-change-cheaters-are-innovators/):

You do not convince existing journals to give more respect to this new field you created. You go out and create your own journals and conferences. John von Neumann did not wait for his colleagues to approve of his work on Computers. In fact, he had to use threats to get what he wanted.

Another one (http://www.daniel-lemire.com/blog/archives/2007/12/05/formal-definitions-are-less-useful-than-you-think/):

There is a widely held belief that shared formal definitions improve collaboration. Certainly, most scientists share several unambiguous definitions. For example, there cannot be a disagreement as to what 2+2 is.

In crafting a research paper, it is important to keep ambiguities to a minimum. You do not want the reader to keep on wondering what you mean. Yet, many highly useful research papers contain few, if any, formal definitions. In fact, entire fields exist without shared formal definitions. One such fields is OLAP: the craft of multidimensional databases. The term was coined by Codd in 1993, yet, as of 2007, I have no yet seen a formal definition, shared or not, of what OLAP is! But it gets more interesting: even the common terms in the field, such as dimension, are fuzzy. What is a dimension in OLAP depends very much on who is holding the pen. Yet, there is no crisis lurking and people do get along.

In short, you do not need shared formal definitions to be productive as a group. A good research paper does not need to introduce formal definitions. Your research papers will be slightly ambiguous.

comment by Bongo · 2011-07-04T14:22:20.434Z · score: 0 (6 votes) · LW(p) · GW(p)

How ironic.

comment by JenniferRM · 2011-07-11T04:44:43.078Z · score: 12 (5 votes) · LW(p) · GW(p)

I just discovered him via a link from Zed on this site. And I appreciate all the links and summary text to help me explore his output.

My initial take after about 10 hours of reading is that he's a incredibly erudite verbal thinker who is awesome for (1) pointing to books that I may actually buy and read, (2) breaking me out of "one model thinking" on topics I didn't even realize I was using only one model to think about. In this respect he strongly reminds me of Mencius Moldbug. However, also like Mencius, he seems to have a very political/social focus which fuels my suspicion that many of his top down "grand theories of how reality works" are false.

Estimating from the hip: there is a 12% chance that I'll buy his book :-)

comment by gwern · 2011-08-15T03:31:04.129Z · score: 8 (4 votes) · LW(p) · GW(p)

In this respect he strongly reminds me of Mencius Moldbug. However, also like Mencius, he seems to have a very political/social focus which fuels my suspicion that many of his top down "grand theories of how reality works" are false.

Oh yes, this is the aptest comparison on the page (a comparison to Robin Hanson would not be amiss either), except I get the feeling that Moldbug is more of a single big idea thinker than Rao (if you used Tetlock's dichotomy, Moldbug is a very happy curled-up hedgehog, and Rao would be the fox, possibly a fox on Adderall).

Both are provocative thinkers who probably are only occasionally truly right. But that's better than most writers, who are never right or provocative.

comment by Zed · 2011-07-14T06:50:05.598Z · score: 4 (1 votes) · LW(p) · GW(p)

His book took me by surprise. I expected it to have the same style and sort of content as his blog (like every other blogger who writes a book) but it was completely different. It consists of half a dozen (unrelated) chapters that end with an exercise. I'm not sure if Tempo is any good (I suspect his solutions overfit the data) but it's short (130 pages or so) so there's no reason not to pick it up if you like the way Venkat thinks.

I'm taking the liberty of adjusting your estimate of wanting to buy the book upwards to 30%.

comment by Davidmanheim · 2018-04-11T14:17:26.717Z · score: 4 (1 votes) · LW(p) · GW(p)

Tempo was a bit more in the style of the blog, but a good read.

comment by Zed · 2011-07-04T10:45:44.900Z · score: 11 (4 votes) · LW(p) · GW(p)

I don't think we should criticize[1] people on a public forum, on the internet where nothing ever gets deleted. For one, this is just not the kind of thread that is going to produce insightful ideas or a healthy discussion.

Second, it's just too delicate a subject matter. I have my thoughts on the authors you list, but I keep deleting whatever I write. I just can't find a way to discuss it in good taste. And really, how can I possibly give my critique on Venkat who came up with the brilliant "I was expecting an inquisition by a panel of Spocks" to refer to the judging nature of the LW populace!

[1] To judge; not necessarily to find fault.

comment by InquilineKea · 2011-07-04T18:45:40.053Z · score: 3 (3 votes) · LW(p) · GW(p)

Hm, well, I'm not asking LW to criticize them - I'm just inquiring about what LW thinks about their ideas (since LW is the only community whose input I would really consider). Right now, I'm pretty much an uncritical fanboy of them, but I'd still like to update my beliefs with some discussion (of course I'll always respect them a lot)

comment by satt · 2011-07-05T08:39:12.875Z · score: 1 (3 votes) · LW(p) · GW(p)

That's quite an extreme degree of self-censorship! Especially with respect to Rao, who's practically a public figure: he's written a book, runs a blog where people can comment on his ideas, and hosts a mailing list.

comment by Zed · 2011-07-05T11:13:54.538Z · score: 6 (2 votes) · LW(p) · GW(p)

I think there's a big difference between discussing a specific idea and discussing the person and everything he writes from Quora posts to the blog, the mailing list and the book. I'd happily voice my opinion about Tempo, or about Slightly Evil, the Gervais Principle or any of his other ideas. They're different works though and they don't deserve to be clumped together in one "what do I think about Rao" kind of post. When somebody is as prolific as Venkat inevitably some of the stuff is going to be good and some of the stuff is going to be not so good.

Maybe you can give a nuanced opinion about a person and his work in a couple of paragraphs. I sure can't. It's just incredibly difficult to do a person's accomplishments justice with some kind of quick summary. If I don't think I can stand behind what I write a couple of years down the road I shouldn't write it in the first place.

Also, this is not self-censorship. We all have (controversial) opinions about a variety of topics and when we voice those opinions we send a signal. Some of these signals we don't wish to send and consequently, on some matters we choose not to voice our opinion. It's just a mundane cost-benefit analysis; nothing more.

comment by satt · 2011-07-06T10:05:37.761Z · score: 0 (0 votes) · LW(p) · GW(p)

Also, this is not self-censorship. We all have (controversial) opinions about a variety of topics and when we voice those opinions we send a signal. Some of these signals we don't wish to send and consequently, on some matters we choose not to voice our opinion. It's just a mundane cost-benefit analysis; nothing more.

I disagree. A decision being the result of a cost-benefit analysis doesn't preclude it from being an act of self-censorship. (Otherwise it would follow that a perfect utility maximizer could never self-censor, which I see as a reductio.) As I see it, avoiding controversial opinions to prevent controversy or an off-putting signal is usually self-censorship, albeit self-censorship that's often benign & helpful.

I think there's a big difference between discussing a specific idea and discussing the person and everything he writes from Quora posts to the blog, the mailing list and the book.

Agreed. But I see all of them as legitimate things to discuss.

They're different works though and they don't deserve to be clumped together in one "what do I think about Rao" kind of post.

Seems alright to me; nothing's stopping a commenter from using this thread to give their opinions about individual works. Besides, people who've read one part of Rao's work are likely to have read others, so it makes some sense to try soliciting their thoughts in one go.

It's just incredibly difficult to do a person's accomplishments justice with some kind of quick summary. If I don't think I can stand behind what I write a couple of years down the road I shouldn't write it in the first place.

That's a high standard. A good standard, but a far higher one than is normal (both on Less Wrong and in general). People on LW post shorthand judgements of others' work quite often (here's a recent example) without being 100% rigorous about it but I don't think it lowers LW's level of discussion appreciably. Someone saying "I'm not a fan of this author, here are a couple of reasons why" or "I really liked this aspect of so-and-so's work on such-and-such" can be very useful even if it's not a careful, judicious balancing of a writer's pros & cons.

[Edited to fix brackets.]

comment by Davidmanheim · 2018-04-11T14:25:08.676Z · score: 9 (2 votes) · LW(p) · GW(p)

I just want to note that the experience of being a resident blogger there was very fun, and it helped me crystallize a lot of ideas - plus his editing and writing advice was VERY helpful. It also led me to think about Goodhart's law, discuss it extensively with Scott Garrabrant and Abram Dempski, and blog about it - and that led in various ways to my recent MIRI paper with Scott.

My posts there: http://ribbonfarm.com/author/david/

So if anyone is looking for a good way to express ideas that are adjacent to, but not exclusively for the rational blogosphere, and/or wants more editing and exposure for their blogging - and already has interesting ideas worth writing about under the broad umbrella of what he'd host - I would say it's worth seeing if you can get into one of his writing workshops / blogpost generating projects - he encourages people to submit a pitch for a piece: https://www.ribbonfarm.com/pitches-and-prompts/

comment by atucker · 2011-07-04T05:24:19.506Z · score: 4 (6 votes) · LW(p) · GW(p)

This post about a LW meetup that he attended is probably relevant.

I personally find Venkatesh's blog very interesting, but am somewhat skeptical of a lot of his particular ideas. He has very cool posts about some historical things, like the importance of modern shipping containers, history of the corporation, and the importance of barbarians. Those are all seriously awesome, IMO.

I also found that his idea of legibility crystallized something I kind of knew before, but not particularly in words. Basically, you have some stuff that follows rules imposed on it (legible) and things that arose organically with lots of local adaptations, that is much harder to understand. Legibility is in the eye of the beholder, and learning thoughts or having particular ideas can make some things legible that weren't before. If you optimize too hard for one particular thing, you get something that's legible but largely useless, if you don't try to impose order its difficult to manipulate.

comment by TAG · 2018-04-12T16:53:15.205Z · score: 8 (2 votes) · LW(p) · GW(p)

Some would say that legibility is James Scott's idea.

comment by Morendil · 2011-07-04T08:03:37.793Z · score: 7 (4 votes) · LW(p) · GW(p)

I'd take issue with the phrase "his idea of legibility". What Rao is doing in that post is summarizing Scott's excellent book on that same topic; it is of course a valuable service to intelligently summarize a book, but it's hardly the same as coming up with the idea in the first place. (I was turned on to that book by a guy named Michael Bolton, who I think is a pretty extraordinary thinker in his own right. Rather than write a post about Scott's book he's only been going around ordering everyone he knew to read the book.)

comment by atucker · 2011-07-04T15:29:20.379Z · score: 1 (1 votes) · LW(p) · GW(p)

True.

So yeah, not his idea. He uses it a lot but didn't originate it.

comment by Jonathan_Graehl · 2011-07-05T19:11:58.673Z · score: 0 (0 votes) · LW(p) · GW(p)

The linked post has prominent first-graf credit to Scott's book.

comment by Morendil · 2011-07-05T19:42:09.792Z · score: 0 (0 votes) · LW(p) · GW(p)

Yup. Which would make "Scott's idea of legibility" more accurate than "his (Rao's) idea of legibility".

comment by Jonathan_Graehl · 2011-07-05T20:47:20.293Z · score: 0 (0 votes) · LW(p) · GW(p)

Yeah, I missed something subtle: atucker, not VR, was claiming too much credit for VR, probably because you went on to contrast Bolton against VR.

We should expect narrative thinkers to riff on each others' themes.

comment by InquilineKea · 2011-07-04T06:53:16.428Z · score: 0 (0 votes) · LW(p) · GW(p)

Oh cool - thanks for the reply and examples!

Which of his ideas are you skeptical about?

comment by Manfred · 2011-07-05T23:28:17.859Z · score: 0 (0 votes) · LW(p) · GW(p)

The lack of posts is pretty much my own opinion too.
Some of Rao's writing seemed interesting from a "making social games explicit" standpoint, which we've talked about on here once or twice.