[jkepler]

Probem is, at least in terms of the climate debate, climate alarmism and mainstream media says that there's scientific consensus, but at least when I read the Intergovernmental Panel on Climate Change's 5th report in 2015, they didn't even predict how much doubling atmospheric CO2 would warm the planet (as they had in their previous reports)---and that with no clear explanation. However, following their footnote to the scientific detail report, they noted a lack of consensus between the computer-based models ("catastrophe immenent!") and the latest observation based models.

When I combine that with prominent climatologists like Judith Curry changing her position when presented with observation-based evidence that countered her climate predictions[1], it leaves me deeply skeptical of any and all alarmists.

I haven't taken time yet to read the IPCC's latest report, but I've heard that it has the same lack of consensus between the computer models (full of alarmist assumptions in how they were programmed) and observation-based studies.

I firmly believed we must follow observational data above computer simulations... Its just better science.

[1] https://reason.com/2023/08/09/this-scientist-used-to-spread-...

[happytiger]

Creating and following the data is science.

When we arrange the science to fit the narrative we deny science in whole and corrupt the scientific method. And unfortunately, when funding is tied to donors or programs this is exactly what we incentivize.

Science is only as good as the freedom of the scientists to ask unencumbered questions.

[agoose77]

Judith Curry is interesting case-in-point; most science should not be conducted in such a manner as to associate it with a particular science; the cult of personality should by virtue of a preference for rationale treatment of evidence be avoided in favour of the facts. This includes a systematic treatment and discussion around uncertainty, something that she is criticised for abusing. It should be noted that she is a > prominent climatologist

precisely because she is a contrarian, i.e. it is not the case that she represents dissent from a prominent researcher. It is to be expected that she is prominent given that it is hard to find many scientists who agree with her thinking. Politically, she is one of the few people available to provide credibility to climate-denialism (even if "credible" does not apply within her own field).

I really want to engage with you on this, because there is an alarming trend in the alt-media to try and depict scientific communities as a cabal of politically minded sell-outs. But the reality is pretty far from this. It's not perfect; peer review needs to be more accessible, journals should be open-accessm, etc., but to depict it as a cult is deeply misrepresentative.

The source you referred to (https://reason.com/2023/08/09/this-scientist-used-to-spread-...) has all the hallmarks of this. It writes as though "establishment science" (which I would be interested in having the authors define) is a person you can speak to, and receive a single opinion. I've seen a number of these kind of "free speech" publications pop up recently, but from what I can see, they are really just contrarian outlets; they seem more interested in being "other" than actually pursuing a particular theme with any kind of rigour.

Part of the problem is of course science communication. The media is problematic in the UK (at least) because of its poor treatment of complex issues like Brexit, climate change, economics, etc). In general, the media tend to fail in communicating how science is conducted, what risk looks like, and how uncertainty is described. You can see this in cinema; science is portrayed as esoteric experts who defy the establishment in some good-wins-out triumph (see Oppenheimer). But, that's really, that's just good entertainment.

We know that computer models aren't perfect. No-one would take you seriously if you gave a modelling result without a description of the uncertainty and how it was derived. But they're also the best thing that we have besides "wait and see", where we walk without agency into whatever future awaits us.

Observational data can disagree with models without invalidating their conclusions; it depends upon the questions that you ask. A specific example; the nucleus is often modelled as a bunch of protons and neutrons moving around independently. We can answer some really interesting questions about why some atoms have very long half-lives using this model. But, if you want to know why a particular nucleus shows vibrational behaviour, the shell model can't help you. Despite this, the shell model is considered one of the most successful theories in physics.