shiredude95

there's some nuance to this. people might want to spend minimal time hacking on their shell and more time hacking on things they find interesting that are not related to shell setup (and also not webshit). besides even if its webshit, what makes you say shell setup hacking is more or less interesting compared to webshit hacking. the term webshit itself implies you view it as less interesting than shell setup -- fair if thats your pov, but doesnt make it intrinsic.

how does this service deal with a coordinated advertising campaign -- most likely also driven by LLM's over a period of say X months. Moderators on subs can be bought out or marginalized, while youtube reviews can also be bought out. In other words, how is an aggregated source a better and more trustworthy source of information than a single blogger who people can ascribe some amount of trustworthiness to over a period of time.

A workaround i've found is to always reply with "no you're wrong" for each seed reply. This forces it to either add more detail or in some cases give a completely different answer. If the replies are in the same ballpark, there is a good chance its not hallucinating. If it gives a completely different answer you are better off rewording the prompt or scrutinizing the answers more than you typically would.

Will be interesting to see what the post-AI movement will look like. I imagine some subscription service by a company that provides "authentic human generated consumable media" which has everything you listed from songs/tv/movies/news etc created by humans. I can see the early adopters of AI will be the early adopters of this type of service as well having gotten tired of the AI golden age. Truly human made content will become a status symbol and beyond the reaches of the average joe as AI generated content screws the supply v demand equation. I can see arts and adjacent fields becoming as in-demand as STEM in around 50-60 years.

nice! sounds very close in style to TSN soundtrack.

so good! added to my bookmarks for future visits.

binary search is standard imo.

in other words dca is the most effective strategy for your average investor?

value is in aggregation across the board, not at an individual level. a single signal is hardly a reliable metric

I built a similar service a while back, with a small modification to the common algorithm.

You can improve contextual summarization by splitting the x sentences into x/n buckets. Then based on the percent of article to be summarized (eg return 60% of the article), pick the sentences ranked in the top 60% of each bucket. Then do this for all the x sentences, ie top 60% across buckets, and combine them together.

This prevents the bias rising from picking a sentence with a lot of critical words.