As an engineer but not a climate scientist, it is really difficult to make sense of these models. They come in at least three levels of complexity, where simpler ones are used to verify the output of the more complex ones.
Fun connection—"3 Idiots" is a Bollywood film about the journey of three engineering students at an Indian college, inspired by Wangchuk's (the engineer's) life!
"In a 2020 journal article for the American Society on Aging, Kushel wrote that of all the homeless single adults in the early 1990s, 11% were aged 50 and older. By 2003, she says that percentage grew to 37%."
When I read something like this I have to ask—Where is the data for the past 20 years?
It looks like a combination of static analysis and live execution of what they call "pure" parts of the code, i.e. code that doesn't affect state outside the current function.
The more plug-ins you have, the more likely it is that ChatGPT will call one in unintended ways. This is also why plug-ins should not be granted permission directly for potentially destructive actions.
I'm not the dev but I find the code logical since each selector is arbitrary and based on a completely different HTML page. Abstracting it would be a huge hassle—would be interesting to see an attempt though!
(You need to sign in, then you have unlimited generations.)
Source code: https://github.com/gitwitorg/gitwit-server
It's still in beta and pretty buggy, but happy for feedback as well!