finebalance
finebalance on Hacker News
lol. 16 years after the first protests started being coordinated over social media, Meta, the largest social media company in the world whose tech folks are probably paid in the millions at the highest level, still hasn't figured out a way to avoid trampling over legitimate dissent for purely 'innocent, technocratic reasons'?
I wonder if these free-speech absolutists will rail about this? Something tells me probably not.
They are different mediums. While the underlying story might be Tolkien's, every frame is an artistic choice and while LLMs can make a choice is many situations, they are unable to 1) keep it coherent b) make it meaningful because art, to me and most, in an outcome of human experiences and thought, which by definition an llm cannot do.
> This war was unavoidable btw, it was going to happen sometime this year or next.
Iran was, as per the latest reports I've read, complying with terms and not enriching uranium to weapons-grade or close to weapons-grade. Are there credible reports suggesting otherwise?
"What's the point of having fuck you money if you never say fuck you."
They are all horrible. If there is ever a reckoning, I hope this entire class of spineless shits get destroyed.
"evolution (fit to environment) than engineering (fit to purpose)."
Oh, I absolutely love this lens.
It can be both, I think? Politicians/powerful people take advantage of divisions in society, but often those divisions do exist in a fully-fledged or nascent form for them to exploit.
I do it more or less this way - except I keep the root intact until the end. It keeps the onion structurally intact until I'm done with the dicing. At which point, the root takes a single chop to lop off, and then the whole thing scatters into tiny, mostly uniform dices. It's quite satisfying.
Google exposes their reasoning. You can use their new gemini python sdk to get thought traces.
> not that a single user uses a lot of energy for a simple chat.
This seems like a classic tragedy of the commons, no? An individual has a minor impact, but the rationale switching to LLM tools by the collective will likely have a massive impact.
That's ok. AI will kill those off soon enough, and like all winners, rewrite history enough so that that inconvenient theft never happened anyway. It's manifest destiny, or something.
I can't speak to Rust, Go or C, but for me LLMs have greatly accelerated the process of learning and building projects in Julia.
Not a clue.
I'm a decent engineer working as a DS in a consulting firm. In my last two projects, I checked in (or corrected) so much more code than the other two junior DS's in my team, that at the end some 80%-90% of the ML-related stuff had been directly built, corrected or optimized by me. And most of the rest that wasn't, was mostly because it was boilerplate. LLMs were pivotal in this.
And I am only a moderately skilled engineer. I can easily see somebody with more experience and skills doing this to me, and making me nearly redundant.
So, tech workers paid >200k couldn't anticipate that hot places are going to get a lot hotter due to global warming and all the trends we've been observing over the last decade or so? Interesting.
... like women are all special and interesting by virtue of being female? Most of humanity is bland and forgettable. Gender is hardly a factor there.
The purple prose is exhausting.
Expecting emotional support from my mother when I feel like I support her in pretty much everything, and have for over half a decade now - finances, physically in the house, her work from the big to the small, and even encouraging her socially at the expense of my own social life. According to the therapist I should have been setting boundaries a long time ago, and shouldn't have the expectation that she should necessarily make an effort even tho I begged her to, and this driven home by the fact that my fiancé has returned the ring and left, largely due to this situation at home. Life feels like a cruel joke. Protect yourself, always.
> And in all of this no-one gives a crap about the children, since they rarely spend even a word talking about tracking down the uploaders or creators of said porn (i.e. they say a site "has child porn", but not how fast it is removed, or if the site gives the IP of child-porn uploaders to the police, or anything beyond trying to establish in the viewer a bad site <-> child porn association).
Aren't there legal requirements in place for this stuff, at least in the US?
Collective fucking madness.
It's tiring to be subject to the constant narcissism. The human toll of the incompetence of this government is incalculable. They've set us back decades, if not forever, because I have no clue how our dumb bureaucracy and spineless civil society are going to deal with destruction that climate change is likely to visit on the subcontinent without completely fragmenting.
I swear. But it's not just an issue with the cheap, knockoff stuff: my Bose Soundsport Earbuds often fail, too, with the left one connecting maybe two thirds of the time. It's so infuriating. The only upside is that they were a gift rather than a purchase, so at least I don't have to suffer through the fact that I paid for this crap.
In so many of these papers it seems to be almost a given that the global south is fucked so, holding that constant, they do a cost-benefit analysis on various levels of warming for already (mostly) rich countries.
It's kind of astounding how rapidly the series has disappeared from the mainstream. Couldn't have bungled it up worse had they tried.
Slick pretentious nonsense unfortunately makes the world go round these days.
"Dost thou think, because thou art virtuous, there shall be no more cakes and ale?"
Possibly: low income earners lost jobs in substantial enough numbers to affect the median (that to some extent is robust to outliers).
It matters, in certain contexts.
For example: a large number of clustering methods boil down to matrix factorization, with variations in the constraints. If you have both domain understanding and a general understanding of what kind of output these variations are likely to result in, you can often narrow down the list of methods you need to try.
Also, interviews.
The point the poster above is trying to make is that given the same amount of data, improvements in technique is leading to significant improvements in accuracy.
An illustrative example comes from the first lesson in fastai's deep learning course: an image classifier that would have been SOTA as late as 2012/13, can be built by the hobbyist in like 30 seconds.
That said, I don't disagree that this is all narrow AI, at best.