4,566 karma · joined August 27, 2014
Into woodworking and a (new) homelab.
jameshtimmins {at} gmail
In LA I pay $105 for "supposedly" 2-300mbit, but this week I've been seeing 30.
I keep looking for alternatives but haven't found any in my area.
Looks interesting, but I'm skeptical that a book can feasibly stay up to date with the speed of development.
I don't understand how or why this is hard for people to grasp? It's no different than Radio Free Europe being secretly funded by the CIA, except it's even more powerful.
10/10 book.
Surprisingly often, it's both.
A trope of many action/thriller movies is groups of top-notch professionals becoming disenchanted with democracy and forming terrorist organizations. While movies aren't reality, it's striking that even watered-down versions almost never seem to happen. Maybe the military is just that good at filtering out those types during psychological testing, or maybe belonging is far more important than ideology.
But that may get back to your point about discipline, motivation, intelligence, etc.
Re SEAL/Delta-level: I guess I'm asking two distinct questions.
1) To use your basketball analogy: while NBA-level skills may be an unreasonable expectation, why do the videos seem to show middle school or JV-level competence? I'd expect that D3 college-level competence wouldn't be super difficult, but evidently that isn't a correct assumption.
2) What is the practical requirement for a world-class tactical unit (or near that level, e.g. D1 basketball or a bad pro team)? I wouldn't expect current militias to be at that level, but what about less developed nation-states?
Practically, why are there not militia groups with Navy SEAL/Delta-level tactical abilities? Or at least near to that? Is it personnel selection effects or bc that level of training requires time/money investments that are out of the reach of non-professional organizations?
I assume the FBI or whomever has automated this to some degree already, and I really hope someone does a great writeup of how LLMs/agents can do even more.
But "just get rid of 80% of the bureaucrats" isn't how you start that conversation.
But if anyone is curious about the idea that we have too many government bureaucrats, and whether this increases costs, I encourage them to read the findings of the Transit Costs Project. In looking deeply into why transit is so expensive in the US, they found that a major cost was that the US has too few bureaucrats, and instead has to hire extremely expensive consultants.
https://transitcosts.com/about/
This aligns with my understanding from friends/family members who work at state and county agencies: they are rarely able to provide the quality of service people want because they're severely understaffed, and are killing themselves in a sisyphean attempt to meet people's needs.
I've looked for other booking sites that helped me determine where I want to go, but haven't found anything else similar sadly.
I plan to get a regular 16, for the reasons you mention. What regular features would you want to see?
I've been trying to find the best next step and what seems fruitful from my vantage point are:
1. Cohere's LLM University - Seems to go more in depth into terms like embeddings that are still pretty unclear to me. 2. promptingguide.ai - For similar reasons, that it covers terms and concepts I see a lot but don't know much about. 3. Reading survey-level papers.
I'm including this info just in case it's useful to you, as I've really appreciated all the content you've put together.
One specific thing you or someone else could do that is simple yet high value is to create a list of "the first 20 LLM papers you should read". I've looked for this to build out more base knowledge, but have yet to find it. Suspect it would be helpful to others as well.
Hardly scientific, but shows the risk of using Hacker News to draw overly strong conclusions of language popularity.
I seem to have noticed that Rust has gotten more common than golang in the job postings, but it's hard to verify without code because there are so many false positives for "go" on any post.