ggdG

6 ms·

1.
by ggdG 5d ago
Would model values like 'openrouter/pareto-code' or 'openrouter/auto' be of any help?
2.
by ggdG 6d ago
OpenAI trying their best to put the Navier-Stokes episode behind them by making the GPUs go brrrr. NYT: https://archive.vn/lWzkk > In its Wednesday night statement, OpenAI said: “In addition, since the completion of Navi
3.
by ggdG 17d ago
I asked my clanker to summarize it for me: https://rentry.co/2ttr46r9
4.
by ggdG 3mo ago
>That night my family watched a war movie on the VCR, maybe saving private ryan or something similar, and the helicopter struck me. If there was a helicopter in the movie, then we can confidently exclude Saving Private Ryan.
5.
by ggdG 3mo ago
> As the patient's temperature approached 22 °C (72 °F), low-frequency and low-amplitude sinusoidal electrical deflections were noted on his electrocardiogram. Dunno about you, but this does things to me.
6.
by ggdG 8mo ago
I think this fits into some "Commoditize The Complement" strategy. https://gwern.net/complement
7.
by ggdG 9mo ago
> I said build on top of wlroots, not DWL. Turns out, the wlroots API is so volatile atm that even the developer of the super small compositor DWL has to throw in the towel for now. > DWL is more interesting as a learning exercise tha
8.
by ggdG 9mo ago
Current status: > 2025-08-16: dwl IS CURRENTLY UN-MAINTAINED. AT THE PRESENT TIME, I (@fauxmight) DO NOT HAVE THE TIME OR CAPACITY TO KEEP UP WITH wlroots CHANGES. https://codeberg.org/dwl/dwl
9.
by ggdG 1y ago
FTA: > [Einstein's General Relativity] tells us that the universe is expanding Does GR really tell us that though? The way I understood it, GR's differential equations will produce solutions for many different constraints and i
10.
by ggdG 2y ago
You're ignoring the elephant in the room: frozen mammoth dung. To restore the gut health of humans, there is this thing called "fecal transplant" [1]: use another person's poo to bootstrap the growth of your own gut micr
11.
by ggdG 2y ago
> the assumption that every non-white person hired is a "diversity" hire before even looking at their qualifications is incredibly racist. That's not an assumption that GP makes.
12.
by ggdG 3y ago
So this is about epidemiologist Martin Kulldorff being fired by Harvard for doggedly sticking to the evidence regarding covid immunity. Some excerpts: > With a genetic condition called alpha-1 antitrypsin deficiency, which leaves me with
13.
by ggdG 3y ago
>You can find a meta-analysis to prove anything you want. I'm not trying to prove anything. I just rely on the judgement of domain experts. In this thread I cited Cochrane, The Lancet, SciAm and Science Magazine. If you have more re
14.
by ggdG 3y ago
>You sure? > https://jamanetwork.com/journals/jamanetworkopen/article-abs... That's just one RCT. The Cochrane meta-analysis looked at a bunch of them. >When the medical field phases out masks because
15.
by ggdG 3y ago
>literally quoting from that meta-analysis, which does not include many clinical trials that have demonstrated an impact: Yes. To their credit, they only looked at randomized controlled trials. >"Key messages We are uncertain whe
16.
by ggdG 3y ago
>This isn't complicated. You can just look at any COVID thread and see what a shitshow it is. I hardly see any covid threads here. I happened to see the one of this week. It got 8 comments before being flagged into oblivion. >Tha
17.
by ggdG 3y ago
It's "self evident" yet a large Cochrane meta-analysis finds no benificial effect of masks whatsoever: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...
18.
by ggdG 3y ago
Here's some serious research, spanning one year. Note how the confidence increases throughout time. You can't blame nvm0n2 for taking for granted what is already well-established since three years. --> May 2020: "How Coron
19.
by ggdG 3y ago
>The six foot rule was based on what the best understanding of the experts was at the time, and probably saved thousands of lives. You can't just make up the beneficial effects of something as you go. Can you cite some randomized co
20.
by ggdG 3y ago
> It seems to me that the quality of any public discussion tends to increase when it’s relevant to the expertise in the room, and decrease when it involves people’s casual reads of complicated stuff about which they have vague but emotio
21.
by ggdG 3y ago
> I'm not interested in expending extra effort to rescue marginal stories with a low likelihood of generating a good conversation. I didn't ask you to expend effort in rescuing stories. I took issue with the way you expend ef
22.
by ggdG 3y ago
> lots of people flag stories based on their experience of what the threads are like IMHO story submissions should be judged based upon their own merits. Toxic commenters can be downvoted/banned but the story submitter shouldn'
23.
by ggdG 3y ago
> The top comment complains that the title submitted to HN is both not the original headline, and not an accurate characterization of the content of the article. What do you mean: "not an accurate characterization of the content of
24.
by ggdG 3y ago
> I think reading the top comment on that post provides plenty of explanation why users would flag that post. That top comment complains that the HN title is WSJ's informative subheading instead of its clickbaity headline.
25.
by ggdG 3y ago
> In fact, I don't see a single story that I personally feel should not have been removed. I don't understand why this story was removed: "It turns out the six-feet social-distancing rule had no scientific basis", ht
26.
by ggdG 3y ago
> They vaccinate for all sorts of diseases. Covid is no different. But the covid vaccines ARE different: Vaccines typically take 10-15 years of research and clinical trials before approval [1]. The covid vaccines OTOH got developed, &quo
27.
by ggdG 3y ago
> I don't understand your goal. Why are you doing this sort of comment? Taking the time to dull peoples minds against whats happening to the world? > Is it for yourself? Are you scared and it makes you feel better to try and cont
28.
by ggdG 3y ago
Thank you so much for this!
29.
by ggdG 3y ago
>Yes, and that retraction was itself political in nature: > https://www.wsj.com/articles/i-cited-their-study-so-they-dis... Indeed. The chronology is important: 1. Cesario et al [1] received some criticism [2],
30.
by ggdG 5y ago
> That study (which was still a pre-print, not reviewed) was retracted. That study wasn't retracted. It's still here: https://www.medrxiv.org/content/10.1101/2021.08.30.21262866v... Maybe you're

More ›