VHRanger

10 ms·

121.
by VHRanger 2y ago
I use Yazi with all of those - it also has ripgrep integrated Between all of those it's hard to think of a better experience navigating a filesystem.
122.
by VHRanger 2y ago
Correct. This effect is known in economics since forever - new technology has - An "income effect". You use the thing more because it's cheaper - new usecases come up - A "substitution effect." You use other things
123.
by VHRanger 2y ago
So what? Not every product needs to be at the 100m user scale to be a success. The nice thing about this era for a creator is that having 10,000 dedicated fans is enough. Dedicated fans will happily monetize on you. 100,000 low engagement f
124.
by VHRanger 2y ago
I'm also curious about gradient-less algorithms For non deep learning applications, Nelder-Mead saved my butt a fees times
125.
by VHRanger 2y ago
Also note this rebuttal: https://singlelunch.com/2023/09/13/the-bad-economics-of-wtfh... With related discussion here: https://news.ycombinator.com/item?id=39144867
126.
by VHRanger 2y ago
How would shrinking the processor size help if the current GPU generation is already completely bottlenecked by VRAM bandwidth saturation? We aren't seeing nearly the same gains on VRAM bandwidth as we are on compute bandwidth
127.
by VHRanger 2y ago
Correct. But military spending only *grows* GDP if the money is invested properly. In, say, research/infrastructure/training, etc. Malinvested money will pump GDP for a few years before growth stops and potentially reverses.
128.
by VHRanger 2y ago
> Google spent heavily to index websites without a clear path to profitability Uh, they were advertising in 1999 already?
129.
by VHRanger 2y ago
I can recommend the yazi CLI tool for filesystem traversal: https://yazi-rs.github.io It has support for fzf and zoxide for these uses. As well as a panoply of other goodies (ripgrep, file previews, commands without exiting the
130.
by VHRanger 2y ago
I agree, moving the earth's axis with respect to the sun seems actually feasible compared to getting a standards body to agree on coordinated change.
131.
by VHRanger 2y ago
Xi Jinping without a doubt. Arguably followed by other dictators like Putin, and backdoor political operators like Peter Thiel. Trump and Musk would be somewhere in the top 20, maybe.
132.
by VHRanger 2y ago
I mean the concept of distance from 3 points introduces a mess of metrics or even measure theory. 2 points always have a shortest path between each other, so the constant is about this fact. For 3 points you have the whole universe of possi
133.
by VHRanger 2y ago
Correct. The word here is not "regulate" it's "enforcement". People doing illegal things should face personal responsibility on the actions. So do managers who approved it. The issue with enforcement is the same thi
134.
by VHRanger 2y ago
SRAM is extremely hot, it's the very opposite of dark silicon
135.
by VHRanger 2y ago
> There must be some law of software engineering that all engineering teams eventually become run like assembly lines where management disallows ground-up thinking. Its a law of incentives. At some point protecting the existing processes
136.
by VHRanger 2y ago
> Is this potentially a part of why Apple doesn't want to support Vulkan? Because they don't want to implement common Vulkan features in hardware, which leads to less than ideal performance? Yes, it's a big reason. I tried
137.
by VHRanger 2y ago
Geoffrey Hinton wins the Nobel prize in physics for giving physics postdocs more reasonable job market options
138.
by VHRanger 2y ago
1. Absolutely correct, it works "sometimes", but not in the general case, and especially not in real life in the general case given how it skews the incentive structure of the firms. The point remains that in general if a foreign
139.
by VHRanger 2y ago
Krugman won his nobel prize in economics for his work on international trade, and especially protectionism of infant industries: https://www.nobelprize.org/prizes/economic-sciences/2008/kru... His NYT column
140.
by VHRanger 2y ago
> I don't think that macroeconomics is an empirical field If that's your opinion it's pretty clear your engagement with the field of macroeconomics is several degrees removed from the actual research. Assuming you're
141.
by VHRanger 2y ago
The proper response to a foreign country subsidizing something is to just buy a lot of it. They're losing money on the unit economics, you can profit from it by buying. We've known for decades that protectionist policies and subsi
142.
by VHRanger 2y ago
Or, you know, one of the 1350 dependencies deep in that stack hardcoded something stupid and now you can't update another critical dependency.
143.
by VHRanger 2y ago
Looks like LLM written dreck to me? Why is this posted here?
144.
by VHRanger 2y ago
I mean... There was a behind the bastards series on the Iceman: https://open.spotify.com/episode/46v0Yz2qu603DLnX8RVcfS And even among subjects of that show I'd give him a "top 5 worst fathers", which re
145.
by VHRanger 2y ago
Modern cars are heavier, but that's not why I'd think they're shit. I use them to go from point A to point B, not to race. 1. Modern cars are shit because in the pursuit of profit growth, cars have tacked on a bunch of shitty
146.
The pawpaw, America's forgotten fruit(atlasobscura.com)
2 points by VHRanger 2y ago|1 comments
147.
by VHRanger 2y ago
Thanks for the comment, very insightful. As for LLM embeddings being inferior: the quality of embeddings really is dependent on the usecase. Embedding is a lossy form of compression. In more than half the cases I've seen, a 32-dimensio
148.
by VHRanger 2y ago
Tmux does suck pretty bad though? It conflicts with the clipboard and a bunch of hotkeys, and configuring it never works because they have breaking change in how their config file works ever 6months or so. These days I only use it to launch
149.
by VHRanger 2y ago
I've talked about this before on my blog: https://singlelunch.com/2020/12/28/why-im-lukewarm-on-graph-... Basically, language models are already graph neural networks. To understand why, go back to Word2
150.
by VHRanger 2y ago
Helix certainly won't give you a 10x improvement. It tends to convert a lot of people moving "up" from VS Code, and still a decent chunk, but certainly fewer neovim users moving "down". Advantages of Helix are prett

More ›