This week: ideologically-biased walks through the garden of forking paths, the decline of the humanities, people who are hard to compress, Claude = Homer, tell me again what “emotional regulation” is, the most impactful forecast of all time, and more.

I ask because I’m not sure I can be bothered. It’s not a ton of work to ID a bunch of new TT ecology hires and compile various bits of public information about them. But it’s not a trivial amount of work either, and to my surprise Claude isn’t much help. It’s also pretty boring work at this point; the data look the same every year. And unlike last year, I haven’t been triggered by any insane ecoevojobs.net threads.* So I’m not really feelin’ it this year.

On the other hand, I know that every year in which I don’t compile these data is a year in which my body of work on this topic starts to slip out of collective awareness. I’d rather that not happen. Many ecology faculty job seekers have found my job market posts useful in the past, and I hope they’ll continue to do so in future. But they won’t find them useful if they’re unaware of them. I can of course repost my old posts, but that tends not to draw much traffice, and the traffic it draws tends to come mostly from long-time regular readers who already saw the old posts.

And while I find these posts boring, I don’t know if readers feel that way. Are there lots of people out there looking forward to new ecology faculty job market posts based on data from last year’s market?

So here’s a little poll. If there’s a bunch of pent-up demand for me to compile data on last year’s N. American ecology faculty job market, I’ll seriously consider it.

*Never underestimate the powerful desire to prove people wrong on the internet.**

**Please do not take this as your cue to post insane claims about the ecology faculty job market on ecoevojobs.net, in the hopes of goading me into proving you wrong by compiling a bunch of data. 🙂

Here is author and former financial analyst Dan Davies, being snarky about people in financial analysis–possibly including him!–who used to claim that you shouldn’t download financial data from Bloomberg because you didn’t really understand any data you hadn’t typed into a spreadsheet yourself.

I link to this because I could be similarly snarky about how some people used to say–in all seriousness–that you didn’t really understand ANOVA unless you’d done all the calculations yourself by hand. Indeed, I may once have been one of those people!

I know better now.

This week: RIP Jeff Houlahan, one academic’s AI workflow, a new review of my book, and more.

Two years ago (almost to the day!), our own Brian McGill asked if ecology is ready for “Big Code” and the errors it hides. That is, increasing numbers of ecology papers are based on hundreds or thousands of lines of code, which peer reviewers and editors don’t ordinarily check for errors, because it’s not traditionally their job, it would take a huge amount of time, and they’re mostly not trained to do it well. Brian discussed various possible solutions to this problem. The last one on his list was this one:

[S]omeday tools like ChatGPT may be able to help here. I am skeptical that it is good enough to find subtle errors in complex code to address today’s big code errors. But I figured it is worth speculating about the future.

No need to speculate any further. The future is here.

Note from Jeremy: Welcome to our latest interview post! Today’s interview is with Bethann Garramon Merkle and Stephen Heard. It’s about their new book on teaching and mentoring writers in the sciences, though as usual with our interviews it ranges further afield. Come for advice on teaching scientific writing that even a lazy schlub like me can implement, stay for the discussion of whether the pedagogical literature is really all it’s cracked up to be, and stay longer to find out what it’s like to write a book with someone you (sometimes) disagree with…

The interview was conducted by email. The questions and answers have been lightly edited for brevity and clarity. Thank you to Stephen and Bethann for taking the time to do this!

This week: Wildlife Photographer of the Year, what makes for good exploratory research, perfect AI alignment achieved, “the world has put up a lot more resistance than I thought to be standardized and quantified,” Fast Slow Fourier Transform, the hedgehog knows one big thing (namely, how to steal an election), and more.

I’m an editor at Am Nat; before that I was an editor at Oikos. An increasingly challenging part of an editor’s job is finding reviewers. Traditional methods all have drawbacks:

• For many mss, the topic’s sufficiently far outside my wheelhouse that I can’t think of any good reviewers off the top of my head. And if I can think of good reviewers off the top of my head, they’ll probably be famous senior people who are unlikely to agree to review (in part because they’re probably already doing another review).
• In my experience, authors often suggest good reviewers (not always). But I’m not allowed to use more than one author-suggested reviewer, and wouldn’t want to use more than one even if I was allowed to.
• Am Nat has a big reviewer database. But the keywords describing each reviewer’s expertise are way too broad to be useful. It takes a long time to find suitable reviewers by first searching the database and then googling the candidates.
• The DiversifyEEB list often doesn’t include any suitable candidates. Even when it does, it’s time-consuming to identify them because the keywords are too broad to be useful.
• Reviewers who decline invitations to review occasionally suggest alternative reviewers, which is very helpful. But most reviewers who decline don’t do this.
• The authors of key papers cited in the ms can sometimes be good candidate reviewers, but often they’re not (e.g., because they’ve co-authored with the ms authors).
• It’s time-consuming to search for recent papers on the topic of the ms, and then check whether the authors of those papers might be suitable reviewers.
• It’s time-consuming to dig through the lab websites of mid-career and senior researchers who work on the topic of the ms, looking for postdocs and senior PhD students who might be suitable reviewers. Plus, mid-career and senior researchers working outside N. America often don’t have lab websites listing their postdocs and grad students.

Recently, I’ve come up with a new method for finding reviewers, that complements traditional methods while being quite fast: ask Claude for suggestions.

Note from Jeremy: Trying an experiment today. Usually when I bang out out a post when I’m annoyed, I rewrite it once I’ve calmed down and then publish it. Instead, here’s the Annoyed First Draft.

Many years ago, back when this blog still had that new blog smell, I complained that I no longer knew what “major revisions” were. Back then, it seemed to me that revisions that once would’ve been considered “minor” had become “major”, and that revisions that had once been considered major were now grounds for rejection. Except, not really rejection, because rejection was often accompanied with invitation to resubmit as a new ms. Except, not really new, because the “new” ms needed to be accompanied by a cover letter detailing the revisions.

Since then, I feel like all those trends have only gotten worse. I’m now seeing and hearing about papers getting rejected with resubmission permitted for what once would’ve been minor revisions. I don’t know if this is all down to journals gaming their metrics, rejecting lots of mss quickly in order to have high “rejection” rates and quick times to “final” decision, or if there are other reasons too. But whatever the reason(s), it’s all become a pointless charade. I’m sorry to say, but most journals no longer convey useful information to authors by classifying their decisions as “minor revisions,” “major revisions,” and “rejection with resubmission permitted.” Those decisions no longer tell authors anything about the nature or magnitude of the requested revisions, and no longer have any discernible relationship to the odds that the ms will eventually be accepted.

So let’s end the charade.

This week: Claude-shaped science, the Nothing Works Clearinghouse, remembering the population bomb singularity, philosopher vs. opera, a new review of my book, and more.