Far Side cartoonist Gary Larson* wrote a brilliant book about the “prehistory” of the Far Side, including a whole chapter of sketches of ideas for cartoons that never made it into print. In the same spirit,** here are a bunch of ideas for posts I will probably never write.
Many of the ideas have been in my notebook of post ideas for years, which suggests I’m never going to get around to writing them. Usually because whatever fleeting inspiration I had has long since passed. Sometimes for other reasons.
In the comments, you’re welcome to to try to talk me into writing some of them. Or out of writing them. Whichever. 🙂
Also, see if you can guess which ones I still think are really good post ideas, and which one I will never write because I now realize it’s a really terrible post idea. 🙂
In no particular order:
- Something on evolutionary implications of the “theory of the second best”, drawing on the famous econ paper on this. If a constraint prevents you from satisfying some optimum, then it’s not true that things that would’ve gotten you closer to the optimum in the absence of the constraint necessarily (or even likely) do so in the presence of the constraint.
- Community ecology doesn’t need more descriptive metrics of “diversity” or “structure”. We have too many already, thank you very much.
- Ecology journals should have contrasting invited commentaries (~400 words each) on key papers, the way some statistics and philosophy journals do.
- What did we learn from the biodiversity-ecosystem function bandwagon, when all is said and done? Was it worth it? How would ecology be different if BEF had never become a bandwagon?
- Teaching post: what’s the portfolio effect? How does it relate to mean-variance scaling?
- Teaching post on what “limiting factors” are, in the context of species coexistence. See classic Levin paper.
- Something on the evolution of bet hedging (and risk-mitigating strategies more generally), and if/how it’s affected by Knightian uncertainty. See this old Noahpinion post.
- Zombie ideas in ecology: selection and complementarity effects. My very own zombie idea (Fox 2005)! Plug Ian Carroll’s Ecology paper undermining the standard interpretation of selection and complementarity effects.
- The most unkillable zombie idea in ecology: limiting similarity.
- Science as a social enterprise (Matthew effect; need for diversity of perspectives/viewpoints/experiences; collaborators with complementary skills; etc.) vs. the desire for meritocracy for individuals. Discuss via analogy with evolution when genes have epistatic effects.
- Something on how different ways of filtering the literature, and of measuring “impact” or “importance”, aren’t actually estimating anything real or “objective”. There’s no “true” or “latent” underlying fact of the matter as to how “interesting” or “important” a paper is, either to the field as a whole or even to any particular individual. Just because different measures of “importance” or “impact” are correlated doesn’t prove that there’s some real underlying variable that they’re measurements *of* (that’s the error of reification). It’s all just professional judgment calls. But not arbitrary, purely subjective ones (this is the hard part to articulate). Sufficiently-shared professional judgments regarding what questions are worth asking and how to go about answering them are a crucial component to scientific progress.
- Something on how the decision of where to publish your work starts when you’re conceiving and designing it, not after you’ve done it. The simulation models one occasionally sees on how to optimize the decision of where to submit a paper of *random* quality are irrelevant. They’re modeling a scenario nobody is ever faced with. The quality of your paper, and the target audience, aren’t exogenous random variables that fall out of the sky, at least not usually.
- Comparative study of how some classical verbal hypotheses of the ’60s and ’70s were subsequently developed, refined, and tested. HSS counts as a success, I think. The IDH and the humped diversity-productivity relationship, less so or not at all. Diversity-stability?
- How, and how often, are recent meta-analyses of the local-regional richness relationship being cited? Have they killed off the ideas that local-regional richness relationships are predominantly linear, and that you can infer something about the strength of local species interactions from the shape of the local-regional richness relationship? Compare to how, and how often, people cite older review/perspectives papers about local-regional richness relationships.
- The process-pattern two-step. How attempts to infer process from pattern morph into muddling of process and pattern. The pattern comes to be seen as *diagnostic* of the process, or even as *synonymous* with the process, rather than as merely *one possible symptom* of the process that might also arise for other reasons. Species sorting in metacommunities as an example? Another possible example: facilitation. Is it a process that might (or might not) cause non-random co-occurrence of two species? Or is it *defined as* such co-occurrence?
- Why do ecologists continue to think of “no disturbance” as the baseline against which “disturbance” should be compared, not “continuous, constant mortality at the same long-term average rate as in the disturbed case”? Why do they continue to confound the mean mortality rate with the temporal variance of the mortality rate? Because they don’t confound mean conditions with the variance around the mean in the context of *spatial* heterogeneity.
- Something on teaching what you don’t know. For instance, I don’t know much about algal biofuels, but I read bit on them to teach them as an example in aquatic ecology. Quite possibly, this means I missed a lot of nuances, made some errors, possibly even taught some zombie ideas. How does one guard against this?
- Are controversial papers disproportionately likely to be interesting/influential? (and if so, is that because they’re “clickbait”?) See Medawar’s famous comment that the best papers are the ones that some reviewers love and others hate. And see Dan Bolnick’s recent tweet that HSS was rejected from Ecology before Am Nat published it.
- Compare the covariates used in recent meta-analyses in evolution vs. ecology. Does the latter more often use “theory-resistant” covariates like latitude, taxonomic group, biome, continent, etc.? Does the former more often use as covariates quantities that could also be parameters in theoretical models, such as population size? And how often do ecological vs. evolutionary meta-analyses claim to be testing theory, or paving the way for new theory? Seems like an interesting follow-up to this old post arguing that meta-analyses in ecology rarely test theory or “pave the way” for new theory, even though they often claim they do. (Aside: might also be worth checking how often meta-analyses in ecology claim to pave the way for new theory. Do they really do so all that often?)
- Compare ecology and evolution papers for what sort of hypotheses they’re based on. Are evolutionary papers more often based on hypotheses drawn from mathematical theory, as opposed to verbal hand-waving or “conceptual models”? Would make a good companion piece to this old post showing that leading evolution journals publish more and different theory than leading ecology journals.
- What common practices from other fields should ecology import? I can think of many that we *shouldn’t* import, and have blogged about some of them (e.g., the social scientific practice of hiring people based in part on where they got their PhDs). But surely there’s something that’s common practice in some other field, that ecologists ought to adopt?
*Man, I’m getting old. I hope most of our readers aren’t too young to know who Gary Larson is. NARRATOR: Most of them are too y– [narrator trampled by herd of cows]
**Except much lazier and more boring.