A key piece of advice in performance optimization is to profile first. The same should apply to environmentalism. Some interventions have orders of magnitude more impact than others, and we shouldn’t pretend otherwise
This is one of those times when programming analogies don't work. Ecosystems are diabolically complex and nonlinear. A small perturbation might be okay one year and collapse everything the next, depending on a thousand other factors. It's good to try, but it would be a terrible policy to mandate.
That sounds like an excuse to completely ignore any evidence and just try whatever interventions you feel like. A far better approach would be to quantitatively evaluate the different possible options, and then take the one most likely to have the biggest impact. Maybe you pick the wrong one due to some complex effect you didn’t account for, but it is still better than picking at random!
I'm not sure how you got from "it's a good idea to try, but terrible to mandate universally" to "do anything you feel like", but it's a very uncharitable interpretation of what I said.
My point is that restricting yourself solely to the particular set of things you can understand quantitatively with all the real world practicalities that entails is limiting. Other than that one word, we're in complete agreement.
It occurs to me that large computer programs can also be diabolically complex and nonlinear. Changing a single byte of code can result in a huge change in process behavior.
I generally disdain the temptation to use computer analogies for natural processes too, but in this case your objection isn’t strong.
Humans tend to limit the complexity of their systems and declare tech debt bankruptcy long before they're comparable to an ecosystem. You certainly could build something distantly comparable (e.g. the internet as a whole), but it's obvious how difficult studying those sorts of systems is.
Exactly how to understand and intervene in complex systems is an active area of multidisciplinary research, not something trivial or remotely well understood.
I repeat that wisdom about performance optimization more frequently than my coworkers probably like, but it can still be taken too far. If you have two approaches that are equally clear and equally easy, take the one that you expect to be more performant.
In this case, we have two lubricants that are apparently interchangeable. One we know has potential downsides (petroleum based oils are known to be harmful to the ground when spilled; this is part of why we have laws against doing so); one does not. Pick the one that doesn't, since all else is effectively equal.
Even in that situation it is still worth asking what order of magnitude the performance difference will be. If a junior engineer codes up the slower one, should they rewrite it or is it not worth the time?
I just don’t understand why environmentalists sometimes get so indignant when asked about the effect size of the interventions they call for. There’s a million low effort lifestyle chances that would be better for the environment, it is a totally fair question to ask if we want to maximize impact