Notices by Ramkumar Ramachandra (artagnon@mathstodon.xyz)

This 1997 paper "Why g matters" is quite disturbing, and paints a very dire picture for "low IQ" individuals. I didn't realize that the correlation was so strong, and am disturbed by the borderline-eugenics theme.

https://www1.udel.edu/educ/gottfredson/reprints/1997whygmatters.pdf

I thought about it for some time, and the study is absolute rubbish -- Wikipedia really should not link to such a sketchy study. The problem with this study and similar studies is:

1. They try to show some kind of weak correlation between two variables without varying all other variables sufficiently. I found inconsistencies, faulty experimentation, and implicit caveats in similar studies, with the conclusion that IQ is not _correlated_ to anything -- humans are not FLOPS-rated CPUs, and the studies are highly misanthropic.
2. The studies don't have anything actionable, and serve no purpose other than to reinforce eugenics views.

Something like "The effect of early intervention in African American children" is a much more credible study, with a clear control group, and clear actionable items.

This Wikipedia page is highly problematic, making wild speculations, citing sketchy old studies and missing citations for many things: https://en.wikipedia.org/wiki/Intelligence_and_education -- please fix it if you're able. I marked it as a hoax for now.

My edits were reverted, and my request for deletion was challenged. It requires a lot more time and effort to get this fixed, and I'm exhausted. Insidious questionable content remains on Wikipedia, and will continue to mislead the general public 😢

This isn't the first time I'm contributing to Wikipedia, but the first time I'm having a sharply negative experience. I painstakingly checked the article and verified that:

1. It states facts from the cited works, and draws wild conclusions that don't follow from the facts, that are not mentioned in the original work.
2. The text is completely orthogonal to the citation link that follows immediately after.
3. It cites terrible sensationalist works that shame academia.

The references are obscure dated documents that are analog in some cases, and most aren't open-access.

The article is not a work of neglect or omission, but the work of a malicious entity.

I explained this in my deletion note with examples, but the editor in question clearly doesn't care about Wikipedia.

Thanks to Florian's endless follow-up work, the TypeSanitizer finally landed, and you should be able to do `clang -fsanitize=type` (should be live on Godbolt tomorrow)! 🎉 #LLVM

https://discourse.llvm.org/t/reviving-typesanitizer-a-sanitizer-to-catch-type-based-aliasing-violations/66092