moab

Strikes me as very anomalous. 5% attendance? Anyway, bad idea to draw conclusions from one university…

These kind of prognostications /opinions are easily falsifiable by talking to your coworkers that went to other universities. Most of the top-25 schools have upper-div CS classes that are nothing like what you're taking about, with both faculty and TAs putting in enormous effort to make sure that courses are accessible and intellectually stimulating.

"During our conversation, I was continually struck by the degree to which Hotz and his company are anti-mimetic. Like many founders of tech startups—thanks to the influence of Peter Thiel – Hotz has a passing familiarity with René Girard’s theory of mimetic desire. The theory, now supported by a trove of empirical evidence, posits that our desires do not originate in us but are always learned from models."

Perhaps not so relevant, but here is a nice read about Girard: https://arcade.stanford.edu/rofl/deceit-desire-and-literatur...

I found your comment hard to follow, but I think I got the general idea. I’m not sure I buy the last extreme disjunction you’re making though. Surely we can have some middle ground between anarchic individualism and a totalitarian thought-police. All long-term societies of the past and probably of the future are examples. The extreme societies historically implode or revert back to moderate patterns that compromise between diktat and individual speech, usually sometime after the death of the god-leader (Mao, Stalin, Hitler, etc.)

I like your theory about local goods/thought products.

It looks like the English translation of Haffner's memoir is titled "Defying Hitler: A Memoir". Thanks for the recs!

Your post reminded me of this: https://www.youtube.com/watch?v=_o7qjN3KF8U This is fantastic satire :-)

This is a great idea, and I hope they will be able to scale it beyond UCB/Stanford/UCSF. A little cynically, I wonder whether PIs at these #1-ranked universities in CS/Medicine actually need an additional funding program since they are probably doing just fine with NSF/NIH and other existing funding agencies. On the other hand they are offering more services than your typical funding agency (planning to help with spin-outs, etc). Also being geographically close together and being near the valley all make sense from the POV of quickly pushing research from academia to industry.

Can you expand more or provide a concrete example for the second point? What kind of database-like searches are you thinking about for spatial data? Things like range-queries can already be (approximately) done. Or are you thinking about relational style queries on data associated with each point?

No doubt there is a lot of good advice in your post. Being independent is of course a key aspect of being a successful scientist and researcher, whether in industry or in the traditional pathways in academia (tenure track asst. prof, etc). It's just unrealistic to expect in new PhD students. Of course at places like Berkeley/CMU/MIT/Stanford you will find many such people, who come with "batteries included". As a result, their advisors can sometimes just let them do their thing and they will churn our fantastic results. But that is not really the norm at most of the top-25 or top-50 graduate programs, and the students there will need a lot more guidance and collaborative work to turn them from merely strong students to great independent researchers. So I absolutely agree that being able to work relatively independently will make it easier to get a PhD. I just think that that quality is something one develops during the PhD journey.

Re length/rambling: It's not a personal attack on you, just a suggestion that the information you write can probably be compressed and made much more effective. Yes, this is not an English class. And yes, you're trained as a mathematician and trying to be rigorous. But often times dropping some details to convey the general ideas is sufficient, especially on a fairly technical forum like this one. Your writing would be more effective if it was less prolix.

No offense but your post is rambling and does not contain great advice for prospective phds. Most advisors I met during PhD were extremely generous with their time and their ideas, and their students would not have found appropriate problems and made rapid progress towards solutions without the intuition or help of their advisors. The story you tell about “2 week discoveries” and rapid independent progress is not the norm.

A little off-topic, but I find it strange to brand the company around the "Eigen" name when it's already a super well-known C++ numerical library: https://eigen.tuxfamily.org/index.php?title=Main_Page

Obligatory on the topic of long-burning bulbs: https://www.tildedave.com/byron.html

The comment is about turkish tenses that are "similar" to the kind of knowledge conveyed in the tenses OP points out. It seems relevant to me.

Thanks for sharing this. The only book by Sebald I have read is "The Rings of Saturn", a wonderful book about a walking trip through the south-east of England. Sebald mixes personal reminiscence with the history of the largely forgotten villages he walks through, places that saw their best times in the early 1900s and have experienced a steady and inevitable decay in the century since. I know I was aware of the Nazi past of his family when reading this, but it doesn't come across strongly, at least in this book. In any case it's incredibly moving and I'm now looking forward to reading it again (perhaps after Austerlitz, which I haven't read).

Reading about bar-ends made me think about the Brompton P-type, which (alas!) Brompton recently discontinued: https://www.bikefolded.com/brompton-s-m-p-h-type-handlebar-d...

Although you might have a point with regards to vocational training, there is basically no research coming out of places like lambda school and the recurse center (please correct me if I am wrong). Things will change only when researchers can find stable (and reliable) sources of funding outside the traditional sphere of institutions like universities.

We have been using objects of the kind described in your last paragraph, called compilers since the 1950s, and with the increasing number of portability-focused high-performance DSLs / frameworks like tensorflow or OneAPI, we are only going further down this direction. But yet 70 years after the advent of compilers, there are still people who know how to open-up the machine, improve it, and fix it, and there probably always will be.

I don't see how machine learning, at least in its current non-AGI state, will be any different. It's just that your average end-user will have no idea how to "open up the machine", but that's also true for compiler technology today.

Thank you for posting this. It's an excellent talk that covers history, personal stories, and great anecdotes. For the amount of success he's had, he comes across as incredibly down-to-earth.

Snark aside, I can't see what good maintaining such a public list and encouraging people (presumably frustrated graduate students) to submit complaints. When working on the experimental section of a paper, maybe half of the papers you try to obtain some source code for or get help from the authors will end up being duds, and so you typically just say in your paper "we contacted the authors but unfortunately were not able to reproduce their experiments / obtain their source code". The existing approach seems like a more polite and equally effective way of encouraging authors to build reproducible research without resorting to McCarthy-style public punishment.

Seems about time that woke-tactics are applied to the high-stakes playground of academic reputation.

Just FYI you probably meant equate instead of equivocate. Two other reasons to add on to the GP's post: (i) improving our species fault-tolerance to astroids, and other existential risk is a good thing (ii) efforts like SpaceX make science "cool again" in a way that it has not been since the Apollo program, and generally since the end of WWII

I agree that in principle anyone can do what they want, and I hope that in our long term our society can actually realize this ideal, but in practice an individual's decisions are informed by the mores and attitudes of the culture they find themselves in. So the conflating factor is our society's history of discouraging women from pursuing technical paths, the (older) insistence that they be homemakers and take on more traditionally feminine roles (like K-12 teachers). Yes, we no longer overtly say such things, but the effects of these attitudes persist across large subpopulations and ethnic groups. It seems very likely to me that traditional gender roles and attitudes about what jobs a women is capable of doing has caused fewer women to pursue math/physics (e.g., see the statistics here: https://math.mit.edu/wim/2019/03/10/national-mathematics-sur...).

Also consider the very recent history of breakthrough results coming from female mathematicians like Lisa Piccirillo and Urmila Mahadev. How much human potential are we leaving on the table, untapped?

I see no "easy fix" for improving representation in higher education other than affirmative action. It's one of the few knobs that we have to turn without much more significant changes to society like implementing UBI or making the average work week much shorter. While it may hurt people like me who happen to be well represented, and prevent me from joining the institution "I deserve" to be in, well maybe that's better for society in the long run.

Thanks for the datapoints and level-headed reply, and apologies for the snark.

I didn't know about the Cornell study, although I am not at all surprised to hear that. How do you suggest representation imbalance in CS faculty gets somewhat closer to 50/50 without having some kind of bias towards hiring women, though? As someone of Asian descent, I totally understand where the lawsuits against Harvard/Yale were coming from, but at the same time I see no way to fix representation in undergrad/grad-school/faculty outside of some kind of "discriminative" hiring procedures (outside of fixing much broader and challenging systemic issues).

I recently applied to the faculty job market and wrote a diversity statement. I have also served on the hiring end (last year), and have heard enough to understand that the content in this statement will neither make nor break your case, but is an additional datapoint that the committee takes into consideration.

I hope the OP understands I was being mostly sarcastic, but really they wrote a (imo) uninformed post about the university life which is far from the reality at the research-oriented institutions I've passed through in the past 10 years.

Thanks for the additional datapoint. Speaking seriously, I am on the same page about bloat and there is a very real danger of it especially in the UC and other state school systems. Maybe private institutions are more robust to bloat since there is less protection for non-tenured employment. Generally the trend to hiring XM$/yr administrators as university heads is worrying, but I am not sure what the fix for it is in the endless competition for endowment and donor money.

Last edit: check out "Bullshit Jobs" by David Graeber if you haven't already. You may not agree with his politics or his prescriptions, but the book is a funny and highly alarming wakeup call to the proliferation of useless jobs (part of the book discusses bloat in university administrations).

It sounds like you have not been in an R1 university (ever?) and are pulling your critiques out of a hat after listening to right-of-center podcasts like the Rubin report. I would love to hear about a _single_ R1 university that is hiring based on identity criteria, where the majority of the staff are "career bureaucrats who contribute nothing to research". The way you get grant money is by doing research that gets published at top-tier conferences, and at least in CS it is pretty hard to bullshit your way into those. Try not to make authoritative sounding posts like this without any knowledge of what you are talking about.

> and they are absolutely in every way possible better off than a bison wondering the Serengeti

Why is that, though? I would reckon that most animals on the Serengeti (not Bison, mind you) live out their average lifespan. Sure, some percentage of them are eaten or die to drought, but there is a reason that their average lifespan is 15--20 years, or whatever it happens to be. Compare that with the 5--6 year average lifespan of a factory farmed animal. It is hard to give you a more detailed argument without understanding the real reasons behind your value judgement.

At what cost, and with what kind of externalities? Individual rewards that come at the expense of social / public health, the inability to have nuanced political conversations, and depressive symptoms for a broad segment of the userbase sounds like a "social ponzi-scheme" to me.

I am sure we would still have some number of dairy cows like the lovely highland cows that this person takes care of: https://www.youtube.com/watch?v=YmBYs53quP8

Their population would be greatly reduced, but is having a billion consciousnesses in a mediocre state and potentially suffering somehow better than having a few orders of magnitude fewer consciousnesses that are properly cared for and living more "natural" lives? It is hard to say without getting into the details, but from a high-level utilitarian perspective, one feels that the latter is somehow more preferable, although doing any kind of arithmetic with souls is a risky business.

Well, if you take the time to learn about dairy farming (even in free range dairy farms), you would quickly find out that cows have a natural lifespan of ~20 years, but cows used in dairy farms have an average lifespan of ~6 years. From the perspective of giving entities that we believe are probably conscious and probably feel suffering and pain a fair chance at life, minimizing the amount of dairy and cheese we consume would seem like a good idea. I'm looking forward to a world where lab-grown cheese substitutes are as good as (or better) than the real thing.

Choice quotes like "Kanye is the only candidate to my knowledge that has talked about agriculture." abound. Funny read.