j-pb

20 ms·

91.
by j-pb 1y ago
`0.00034MWH`should probably be kilo not mega, but I think you still did it correctly :D I think I used Btu (thermal units) in my calculation, so I only calculated actual energy expenditure for my 60.000 queries result. But you are right, a
92.
by j-pb 1y ago
Whatever you need to tell yourself to keep eating meat buddy.
93.
by j-pb 1y ago
I'm gonna (mathematically speaking rightfully) consider the conversation an absolute waste of time and resources, until a significant portion of humanity does that, yes. Because you falling over dead and consuming 0 resources makes onl
94.
by j-pb 1y ago
- https://mistral.ai/news/our-contribution-to-a-global-environmental-standard-for-ai - https://www.statista.com/statistics/1201677/greenhouse-gas-emissions-of-major-food-products/ -
95.
by j-pb 1y ago
You can try to misconstrue and ridicule the argument, but that won't change the math that if you have one thing that causes 1 unit of damage, and another thing that causes 100.000 units of damage, then for all intents and purposes the
96.
by j-pb 1y ago
Yes but one with a much much much smaller impact as we just demonstrated. This is exactly the kind of cognitive dissonance in people that I meant. You literally see the math and go "but I like my meat, why should I give that up if you
97.
by j-pb 1y ago
You don't need beef, beef is a lifestyle choice. I use LLMs to do all of my coding these days, it's certainly more essential for feeding me than beef.
98.
by j-pb 1y ago
people downvote your sarcasm, but if you do the calculations you're kinda right. 1Kg of Beef costs: - The energy equivalent of 60.000 ChatGPT queries. - The water equivalent of 50.000.000 ChatGPT queries. Applied to their metr
99.
by j-pb 1y ago
Cost of opportunity and diminishing returns. It's stupid to optimise a 0.01% case when you can spend that time on a 25% case.
100.
by j-pb 1y ago
1Kg of Beef takes as much Energy as 60.000 ChatGPT prompts, and as much water as 51.383.333 ChatGPT prompts. Stop eating meat, driving your car, and taking hot showers. Then we can talk about prompt resources.
101.
by j-pb 1y ago
For me the thing that convinced me was the ability to write so much more documentation, specification, tests and formal verification stuff than I could before, that the LLM basically has no choice but to build the right thing. OpenAIs Codex
102.
by j-pb 1y ago
I never said to use LLMs to generate Uiua. I said that Uiua is an edge case where tacitness is indeed elegance. I wouldn't write anything but Rust via LLMs, because that's the only language where I feel that the static type checki
103.
by j-pb 1y ago
I disagree; Elegant software is explicit. Tbh I wouldn't mind if we got rid of derives tomorrow. Given the ability of LLMs to generate and maintain all that boilerplate for you, I don't see a reason for having "barely enough
104.
by j-pb 1y ago
You return an error specific to that function. If it internally has a `InnerFuncErr::WriteFailed` error, you might handle it, and then you don't have to pass it back at all, or you might wrap it in an `OuterFuncErr::BadIo(inner_err)`or
105.
by j-pb 1y ago
The fun part is when you take that approximation and apply it to 2^n. If you have n bits, your collision probability is 0.5 at generating 2^(n/2) values. Or put differently: a 128bit uuid gives you a "good enough" 64bit distr
106.
by j-pb 1y ago
The current naivite of the scientific community is exhausting. As if the current political climate isn't going to result in the sabotage of scientific infrastructure if some state actor decides that it could provide some economic or mi
107.
by j-pb 1y ago
It's about the functions you _do_ call. Those probably have larger scope beyond your specific use case. Worse they have to support the superposition of various use-cases of their users. Let's say you got a library to do arbitrary
108.
by j-pb 1y ago
This. We finally have a tool that can learn from all the libraries and abstractions that have to fit everybody's needs (and do so badly because there is no free lunch), and extract just the parts that are actually relevant to our probl
109.
by j-pb 1y ago
Most database literature simply uses Datalog to mean the query language fragment of conjunctive queries + recursion/fixpoint-iteration and potentially stratified negation. Yes it started out as a Prolog subset, but the definition as th
110.
by j-pb 1y ago
Given that phase change captures a ridiculous amount of energy (it takes roughly 5 times as much energy to go from liquid water to steam than it takes to bring that same water from 0C to 100C). I've always wondered why we don't op
111.
by j-pb 1y ago
wtf are talking about? the word model comes from latin modulus (“measure, standard”), it has nothing to do with the similar word (which is derived from the same latin word for standard) used to describe someone who shows off clothing
112.
by j-pb 2y ago
Yet they are bad at every one of those points. * An auto increment ID is just as human non-readable as a UUID, it's just easier to get silent collisions from typos. * The Source is metadata that belongs in a metadata system, not into t
113.
by j-pb 2y ago
Scientific publishing is one of the very few legit use cases for block-chains imo. But magnet links and the BitTorrent mainline hash-table are a better DOI than DOI.
114.
by j-pb 2y ago
Their whole organisation should have been a hash function... DOI must die
115.
by j-pb 2y ago
XML is still a huge mistake for most stuff. It's fine for _documents_ but not as a data storage solution. Bloat, ambiguities, virtually impossible to canonicalise. XPath is cute, but if you don't mind bloat, text-only and lack of
116.
by j-pb 2y ago
SD-Cards are a currently widely used physical media. Even my streamlined slimed-down minimalistic MacBook Pro has an SD-Card slot. We can make SD-Cards with music and movies on them a reality, we just have to walk to our nearest convenience
117.
by j-pb 2y ago
Ah damn, now I know why I had the feeling I forgot to hand in an assignment for a while...
118.
by j-pb 2y ago
These guys write really great papers! We implemented a simplified version of their ring index for our data space ( https://github.com/triblespace/tribles-rust/blob/master/src/... ), and it's a re
119.
by j-pb 2y ago
I'm not sure how to make the multithreaded case more explicit. The point is that with ownership the multi-threaded case does look exactly like the single threaded case, just with some machinery to "aquire" all the things you
120.
by j-pb 2y ago
Yeah, the idea in the above code is that you can't do anything with an account so long as it's placed in the "blocked" set, because that means that it's effectively taken out of circulation for other operations. The

More ›