2,257 karma · joined September 25, 2014
My initial thinking is to use a text classifier to assign political alignments, or lack thereof, to HN comments. From there I can compile statistics about each of the comment classes. This entire process can be done in an executable jupyter notebook for visibility and verification of results.
I offer you a challenge:
I will provide you with quantitative data that I am correct on this issue, and in return you remove the restrictions on my account.
I beg to differ. Any attempts to contain the mass violence were decried as "Trumpist fascism".
True, but if you relax your hard requirements of optimality to admit "good enough" solutions, you can use heuristic approaches that are much more tractable. High quality heuristic solutions to NP-hard problems, enabled by ML, are going to be a big topic over the next decade, I think.
Edit: During my schooling I developed a rare form of progressive vision loss. It also makes me extremely light sensitive, and thus it's difficult to conceal. After 1500 job applications I wound up as a construction worker doing hard labor. Since then I have managed to work my way back up to a fairly successful ML engineer, although that never would have happened if I hadn't gone to extreme lengths to conceal the state of my eyesight from my employers.
Don't ever imply I don't know what discrimination is like.