Monday, May 20, 2024

Friday links: RIP Daniel Dennett, rewilding the internet, and more

EcologyFriday links: RIP Daniel Dennett, rewilding the internet, and more


Also this week: reviving cybernetics, Spinoza vs. Meghan, no one buys books (because they don’t want to), no one buys future onions (because they’re not allowed to), McSweeney’s makes fun of Jeremy, and much, much more. Oh, and 1957 called, it wants its panic about public trust in science back.

From Jeremy:

Public trust in science is alarmingly low today. We can and should fix this with more rigor in study design and statistical analysis, open data, de-emphasizing statistical significance, being honest about sources of uncertainty, valuing replications and incremental advances over purportedly novel “breakthroughs,” communicating our results more clearly…[listens to voice in earpiece] Hold on, I’m getting a report…Correction. Everything I just said wasn’t about today, it was about 1957. My apologies for the error.

Science news article about a new paper (Lamb et al. 2024 Ecol Appl), showing that wolf culls are the only management intervention known to reverse declining abundances of southern mountain caribou populations. YMMV on the broader lessons to draw, if any. My own reaction (well, one of them) is that this is an illustration of the general principle that management interventions can fix a problem even if they don’t address the “root causes” of the problem (here, deforestation). Another example would be my contact lenses, which fix the problem with my vision (nearsightedness) without addressing its root cause (the shape of my eyes). I leave it to commenters to discuss whether, why, or in what circumstances public policy, and the research that informs it, should focus on “root causes.”

Students at (private US) colleges and universities should (be made to) study more.

Philosopher Daniel Dennett has died. His most influential work among philosophers concerned consciousness. Readers of this blog will probably know him better for his popular book Darwin’s Dangerous Idea. I also really like his little essay about the dangers of studying “chmess”–artificial puzzles of no lasting importance. I was fortunate to meet him years ago, when he gave the annual Darwin Lecture at my university.

Helen De Cruz, on whom Dennett was a formative influence just by being himself, uses his passing as an occasion to reflect on how just being yourself can be part of a good moral life. I liked this a lot. Resonates with Meghan’s posts on the value of saying “no.”

Maria Farrell and Robin Berjon on the need to rewild the internet. I have mixed feelings about the policy prescriptions for which this essay argues. I found it interesting mostly as an example of ecological ideas getting taken up and applied outside of ecology.

Henry Farrell reviews Dan Davies’ new book on management cybernetics, The Unaccountability Machine. I loved Davies’ last book, on financial fraud, which has applications to science. So I’m really looking forward to this new book.

Here’s why there’s no futures market for onions in the US, the way there is for (say) lean hogs. Long but interesting history.

The economics of US fiction publishing. Welcome to the world of very right-skewed distributions.

UC San Diego has a new points system for choosing among continuing students who want to switch into a selective major (life science majors, public health, data science, computer science, some engineering majors). Students with the most points get to switch. I found this interesting because the points system doesn’t just give points for the things you (well, I) would expect, such as GPA in intro courses. It also gives points for California residency, Pell Grant eligibility, and first-generation college status. I’m not aware of other institutions that use measures of socioeconomic status, or first-gen status, as criteria in major program admission (as distinct from admission to the institution as a whole). Which may just illustrate my lack of awareness, happy to be pointed to other examples by commenters. I should also hasten to add that I don’t know nearly enough about UC-San Diego to have any opinion on this policy. I don’t know anything about, e.g., how many students this policy affects, or what other criteria could be used to choose among students wanting to switch into a selective major, or how this policy does or doesn’t affect students’ initial choice of major, or whether there are any plans to increase the number of available slots in selective major programs so that access to them doesn’t have to be rationed, or etc.

Alex Usher’s commentary on what the 2024 Canadian federal budget means for universities and scientific research. Like him, I’m glad to see the big increase in the value of major federal graduate scholarships (and postdoctoral fellowships). I was visiting UFlorida last week, where two different people excitedly shared news of this increase with me. I think I slightly disappointed them by pointing out that those federal scholarships and fellowships are very competitive; most graduate students don’t get them. And the other sources of STEM graduate student funding in Canada aren’t keeping up with inflation. So if you want to substantially increase per-grad student funding for all Canadian STEM grad students, I’m not sure how you do that except by having fewer STEM grad students. And as the linked piece points out, yes, the government plans substantial increases in federal scientific research funding down the road. Some of those planned increases could be spent on increasing the stipends of all STEM grad students. But those plans likely won’t come to pass unless the government wins the next federal election. Which at the moment it is very unlikely to do.

Can a planned economy work, if only the planners have good enough data, good enough optimization algorithms, and good enough computers? Maybe it couldn’t work back in the 1950s heyday (if that’s the right word) of the Soviet Union, but now it can, thanks to advances in computer hardware and software? Here’s Henry Farrell interviewing Red Plenty novelist Francis Spufford about those questions. And here’s statistician and polymath Cosma Shalizi reviewing Red Plenty, and explaining why the answers to those questions are “no” and “no.” You would think a novel about linear programming and economic policy would be boring, and that reviews and interviews about it would be even more boring. But you would be wrong!

Stephen Heard argues that PIs should set up their research teams in such a way that more junior people deserve co-authorship on papers, according to traditional authorship criteria. Rather than relaxing authorship criteria so as to give more co-authorships to junior people. Agree.

Liam Bright on how academic disputes about “truth,” “objectivity,” etc., actually have much less to do with politically controversial issues than many of the disputants like to think. I think this is right. Now I’m wondering if similar arguments apply to other abstract academic disputes that are claimed to have relevance for everyday concerns.

Speaking as someone who likes smoothie sours and pastry stouts, I feel personally attacked by this. 🙂

Check out our other content

Most Popular Articles