bbor

49 ms·

751.
by bbor 2y ago
Yup, well said. If you think we engineers have it tough, think of the poor AI professors — the go-to book for AI survey courses ( Russel and Norvig ) is maybe 75% symbolic approaches, and students are surely up in arms about that now that e
752.
by bbor 2y ago
To be fair, and in case it isn’t obvious: this is kinda this guy’s whole schtick. And has been for decades: The inability of standard neural network architectures to reliably extrapolate — and reason formally — has been the central them
753.
by bbor 2y ago
I'm glad I'm not the only one initially drawn in by the title! As the old meme goes; > If you can't describe your job in 3 Words, you have a BS job: > 1. "I catch fish" Real job! > 2. "I drive taxis&
754.
by bbor 2y ago
Wow, fascinating stuff and "grokking" is news to me. Thanks for sharing! In typical HN fashion, I'd like to come in as an amateur and nitpick the terminology/philosophy choices of this nascent-yet-burgeoning subfield:
755.
by bbor 2y ago
A) Wow this is just incredibly impressive for a solo dev - well done! The feature list just keeps going and going, by the time I got to kanban boards I was in disbelief. I was incredibly dubious based on the title that any “Show HN” could r
756.
by bbor 2y ago
startup success across stages of a venture’s development - from initial fundraising to exit. … we know how pointless this all is, right? Not the research (which seems like a solid confirmation of intuitive assumptions and advice) bu
757.
by bbor 2y ago
I'm just a rando but here's my unprompted solution: treat your fellow citizens as fallible epistemic machines instead of unified, rational members acting freely within a social contract. The solution to flat earthers (to pick a sl
758.
by bbor 2y ago
I. This is a well written, engaging article, thanks for sharing. I absolutely agree that the thesis is true to an extent, though I would phrase it more broadly (perhaps "American culture is unusually individualistic" or "Amer
759.
by bbor 2y ago
A) fair, b) I think this common distinction is a little overblown. Authorization is just a particularly straightforward CRUD feature, perhaps with some inheritance logic — authentication seems to be where 99% of all security sadness comes i
760.
by bbor 2y ago
Oo my favorite topic! Great writing and the right themes are there, but I think they’re missing a lot by not taking a more historically-holistic view. Aka wondering what all the people who’ve been criticizing psychology think, from Chomsky
761.
by bbor 2y ago
Why are you so confident that Philosophy isn't the superclass to "science"? How could you hope to start on any science without philosophy, much less arrive at a definition for the term? I could maybe see mathematics without
762.
by bbor 2y ago
Interesting point, fair enough! Orion is indeed thicker than normal glasses now that I look ( https://about.fb.com/wp-content/uploads/2024/09/05_whatcomes... ), tho I can't access this article for a d
763.
by bbor 2y ago
I mean, this is awesome, obviously, but the framing is odd — this is very, very far from the SOTA, no? 30° FoV and ~30% image degradation with monochromatic images seems way less capable than the Orion Glasses. But I’m sure that’s just jour
764.
by bbor 2y ago
…are we complaining about people living in India, or American/Australian citizens that have Indian national heritage? The former seems more like an outsourcing issue than an Indian issue. The latter… didn’t we learn in the 90s not to s
765.
by bbor 2y ago
I’m gonna dip into politics a bit for my assertion that “fight racism with racism” isn’t a great move, either for the societal long term or the personal/legal short term.
766.
by bbor 2y ago
Idk, just looking at it now Barbour seems much, much more rigorous. The linked article is more “using scientific terms to muse about philosophy” than physics, IMHO. For example; In essence, therefore, we experience time because of the i
767.
by bbor 2y ago
Hmm I think your question is a little vague in terms of what exactly is meant by “interface”, and I’m not an expert on LLM deployment myself, but to help you out with the big picture: LLMs are statistical models that take in some big array
768.
by bbor 2y ago
Very cool, and it’s already on version five ! I’m impressed. Only one question for now, since I’m don’t yet have experience with these specific data viz techniques: Skew-ey? Skoo-ey? Squee?
769.
by bbor 2y ago
Basically all of this is based on the scientific principle of “eh it just seems like it should be like that”, which I’m not a huge believer in. Like; by the time children being born today are in their eighties, if they live that long, a
770.
by bbor 2y ago
Plus the sync only works between clients AFAICT — literally all the listed backends are also “coming soon”. Which is troubling, I’d guess that almost all local-first applications need some sort of trusted verification server, at the very le
771.
by bbor 2y ago
Well Mr. Dang, we've got a new edge case for you lol: An article with substantive content before the paywall, but that doesn't have any workarounds for the rest of the article. I would love to be proven wrong, but AFAICT we'd
772.
by bbor 2y ago
Amazing article, thanks for posting! I don't know this author, but they've definitely got a solid understanding of the relevant facts, IMO/AFAICT. That said: Altman could now get equity in OpenAI—around $10 billion worth
773.
by bbor 2y ago
I’m glad to know a new article that “everyone knows”! Thanks for pointing out the age. And, at the risk of intentionally missing the metaphor: they do in fact make automated tenderizers, now ;) https://a.co/d/hybzu2U
774.
by bbor 2y ago
Oh man, this one hits close to home! I for one am deadset on setting foot on Luna one day, so I need this to work. Ever since reading the incredible treatment of this issue in the Red Mars trilogy, I’ve been on board! And with intuitive
775.
by bbor 2y ago
The notion that it’s a platform for supporting creators in general, rather than merely for news, could help justify its valuation — the company was reportedly valued at $650 million in a 2021 funding round — and open the door to a poss
776.
by bbor 2y ago
It’s for the dreamers. The crazy ones among us that do not think of themselves as experts in programming machines to solve business problems, but rather novices in cajoling machines to think like humans do.
777.
by bbor 2y ago
The core use case is as a small part of larger programs. It’s just computer vision but for words :)
778.
by bbor 2y ago
Most LLMs are first trained on a broad dataset that informs the model’s overall knowledge and capabilities, and then are customized for a particular domain. That wouldn’t work for Rufus, since our aim was to train it on shopping data f
779.
by bbor 2y ago
Anyone a fan? I mean, it’s creative, certainly. It’s a bit weird/a shame that the logotype doesn’t have anything to it now other than bold sans serif letters, and they certainly were fighting an uphill battle by changing the terribly c
780.
by bbor 2y ago
Partisan political rhetoric, no reporting here IMO. “Acting like a president should” includes telling desperate people that FEMA is not coming to help, despite being immediately contradicted by the Georgia governor and FEMA? It includes le

More ›