binarymax

8 ms·

241.
by binarymax 1y ago
But what does the study show? There was no control for anything. None of the data is valid. To clarify: how does the research team know the bots were interacting with people and not other bots?
242.
by binarymax 1y ago
We’ve seen lots of posts about DOGE get killed. It’s not far fetched to think they are actively working to keep news like this off HN’s front page.
243.
by binarymax 1y ago
Just last month I finally moved on from my iPhone 6, which had been working great for 10 years, because some critical apps stopped working unless I upgraded, but couldn’t upgrade because apple no longer released iOS updates. It needed a new
244.
Can you control lightning with sound?(youtube.com)
3 points by binarymax 1y ago|0 comments
245.
GitHub Copilot auto enabled itself on all workspaces without my consent(github.com)
19 points by binarymax 1y ago|2 comments
246.
by binarymax 1y ago
Funny, I had that experience with Docker - mostly due to c++ dependencies - that were fine in Nix
247.
by binarymax 1y ago
Yes, I know that. But when talking about reproducible packages, we can and should learn from existing techniques. Is Fedoras system capable of this goal? Is it built the right way? Should it adopt an alternate package manager to achieve
248.
by binarymax 1y ago
I often see initiatives and articles like this but no mention of Nix. Is it just not well known enough for comparison? Because to me that’s the standard.
249.
by binarymax 1y ago
He certainly decried the use of “all-natural” labels for products, because “everything comes from nature!”
250.
by binarymax 1y ago
Contrarily, I love the mini timer, because it’s something to beat. The mini is too easy of a puzzle. The NYT crossword is more for the thinking type - it also has a timer, but its less in your face and I don’t feel the pressure.
251.
by binarymax 1y ago
Hard rules are fallible, but we can lean on precedent in the supreme court for an adjacent topic (obscenity): https://en.wikipedia.org/wiki/I_know_it_when_I_see_it
252.
by binarymax 1y ago
The rates were likely a bad GPT prompt. I wish I was joking. https://www.theverge.com/news/642620/trump-tariffs-formula-a...
253.
by binarymax 1y ago
With what I’ve seen I don’t believe the administration is acting in good faith. Everything about their behavior suggests otherwise.
254.
by binarymax 1y ago
Pandas uses numpy which uses C. So if numpy used CUDA then it would benefit.
255.
by binarymax 1y ago
This is from Allen Ginsberg's Howl. A must read.
256.
by binarymax 1y ago
I’m interested in learning more about this. Are these APIs documented anywhere and are there server side implementation examples that you know of? EDIT: this is the closest I could find. https://developers.google.com/chrom
257.
by binarymax 1y ago
Also a perfect opportunity to link to one of my favorite comics https://onegianthand.com/post/188730414621/silica-gel
258.
by binarymax 1y ago
Even without your own rack or colo, The math with AWS stops working as soon as you no longer fit in the free tier, since providers like Hetzner are 40% cheaper.
259.
by binarymax 1y ago
They talk about some examples in their research. > “Specifically, we initialized the DeepSeek-Prover using the DeepSeekMath-Base 7B model (Shao et al., 2024). Initially, the model struggled to convert informal math problems into formal s
260.
by binarymax 1y ago
And as we can see from DeepSeek this clause means nothing, outside the realm of OpenAI blocking your access to its models.
261.
by binarymax 1y ago
Seems like the author really wants to build a search engine and not use something that exists already. This post https://www.parse.ly/lucene/ provides a great insight into the value of file system use when writing inve
262.
by binarymax 1y ago
This is one of those interesting mathematical attempts to formalize a very human “I know it when I see it” quality. So maybe k-means is a poorly defined task if you don’t know k . Then you end up with the meta task of defining k, which h
263.
by binarymax 1y ago
Let’s look at this like food, instead of knowledge. You ate a donut, and you got energy. Maybe you’ve eaten 5 donuts per year, for 20 years…so a hundred donuts. Maybe you didn’t pay for all of them, you got some at work or at a friends hous
264.
by binarymax 2y ago
I just tried this, thinking that I was "holding it wrong" and no - it still doesn't work. It says it doesn't know where I live.
265.
by binarymax 2y ago
On Lambda cloud, an 8x H100 is $14.32/hr. H100s are better than 4090s, So if you count setup time it’s probably about $100
266.
by binarymax 2y ago
The personal context issue could have been solved long before LLMs appeared. My foil for this has always been the simple request from Siri: "Take me home". Apple and my device know where I live. I have even registered "home&
267.
by binarymax 2y ago
This is really cool. You might be interested in this talk which shows you can incrementally add some preference vectors as a way to improve the recommendation. https://haystackconf.com/us2023/talk-20/
268.
by binarymax 2y ago
The article is too high level to figure out exactly what they are doing.
269.
by binarymax 2y ago
If they sell you a defective product, and the support is also defective, then you need to take a different route.
270.
by binarymax 2y ago
Sometimes you have to hack their support script to get a replacement or a refund. After the first support call if you don't get what you want and it happened again, Call back to open a new support ticket. Pretend to walk through thei

More ›