necovek

31 ms·

751.
by necovek 11mo ago
I've actually had ChatGPT admit it was wrong by simply asking a question ("how is X achieved with what you described for Y"). It responded with "Oh, it's a great question which highlights how I was wrong: this is wh
752.
by necovek 11mo ago
Because we generally define intelligence by what we do, even if it's not clear how we do it.
753.
by necovek 11mo ago
There is a huge risk with that type of usage: for some percent of cases (unknown how much, though), it will get things seriously wrong. You'll be accountable for the decision though. This is what RMS is flagging, though not very subst
754.
by necovek 11mo ago
This seems to be a complaint against general use of "Artificial Intelligence" term: none of it is "real intelligence" as we don't really have a definition for that.
755.
by necovek 11mo ago
I am complaining about the language (phrasing) used: a Python, TypeScript or Java program might be truly portable across architectures too. Since architectures are only brought up in relation to dynamic libraries, it implied it is otherwise
756.
by necovek 11mo ago
This seems to imply that Go's binaries are otherwise compatible with multiple platforms like amd64 and arm64, other than the issue with linking dynamic libraries. I suspect that's not true either even if it might be technically po
757.
by necovek 11mo ago
And Serbian, so I'd guess it's proto-Slavic word.
758.
by necovek 11mo ago
For something in the financial space, I don't see much (or really, any) tests in the code repository. CI also only has ESlint and prettier running. How are you ensuring the application will remain maintainable in the future, you are no
759.
by necovek 11mo ago
I've generally switched to management when I realised I was doing both great, complex technical work and management work my managers were failing at. I fully subscribe to the Conway's law, so I love creating system architecture th
760.
by necovek 11mo ago
Hardware can do any "magic" software can, which should be obvious since software runs on it. It's just not as cost-effective to modify it after shipping, which is why the tech sector is moving to more sw less hw (simplified,
761.
by necovek 11mo ago
All the American cars (Ford, Chevrolet, GM...) are much cheaper in Europe than eg. German cars from their trifecta (and other Europe-made high end vehicles from eg Sweden, Italy or UK), and on par with mid-priced vehicles from the likes of
762.
by necovek 11mo ago
Sounds a bit like when UI/UX was standardized and improved for GNOME 2.0 with the introduction of HIG: it was 2003 when it was released.
763.
by necovek 11mo ago
That was my thought too, but on second look, I am guessing it's Home Assistant instead.
764.
by necovek 11mo ago
Generally legal system has negligence, and once someone is provably informed of the negative consequences but keeps being negligent, "willful negligence", which is much closer to intent (I see it defined as "intentional disre
765.
by necovek 11mo ago
Gives them proof they did their best to "protect minors" even if they circumvented the GeoIP rule: someone trying and realising it still does not work might get X percentage to not bother further thinking there was something smart
766.
by necovek 11mo ago
When that happens, most VPN providers will face similar destiny. Which means that we'll all have to run our own VPNs, possibly masquerading as HTTPS traffic, if that remains viable against government interference (eg. they might ask to
767.
by necovek 11mo ago
Is there a mesh wifi system that can run open source firmware? I imagine that might be the best bet for VLAN tagging too in the "affordable" sense too.
768.
by necovek 11mo ago
In that particular example (drunk neighbour damaging your property with reckless driving) is really — in most of Europe, I would guess as Serbian laws are largely copied over from different EU country laws — handled by the insurance. Basica
769.
by necovek 11mo ago
> Image based is the future. While this is the direction many are going for particular use-cases (IoT in particular), I am very much conflicted. Yes, inconsistent updates between components have caused a couple of nights of fixing my RPM
770.
by necovek 11mo ago
While I agree we do a lot of our problem solving with symbolic languages (streams of images), even if we define "thought" as symbolic language processing, I believe many great experts in philosophy and law do internalize the relat
771.
by necovek 11mo ago
But taking a step back, this process of converting reasoning tied to experienced consequences into words that have relatively stable meaning and interpretation over generations is what is "academic". Without that, one does not lea
772.
by necovek 11mo ago
The top-level comment tried to distinguish betweeen symbolic processing — verbal and non-verbal — as really being "thought", and other cognition/reasoning not. I believe many of the things you bring up still involve symbolic
773.
by necovek 11mo ago
It happens because cybersecurity teams do not design for efficiency and believe that security trumps everything else. If they understood that security, just like anything else, is there to drive the business, they'd perhaps sit down wi
774.
by necovek 11mo ago
I have to disagree on the tests not potentially helping here. Finding the right abstraction layer is hard, but there was obviously no integration test that tested wherever the original query was being constructed and where the output was be
775.
by necovek 11mo ago
This bug might not have, but others would. Formal verification methods still rely on humans to input the formal specification, which is where problems happen. As others point out, if they didn't really ship fast, they certainly would n
776.
by necovek 11mo ago
No disagreement there, but opportunity costs are present and unregulated everywhere: eg. a bad traffic light design (timings) might increase congestion and greenhouse gasses emissions 10×, but nobody is losing their traffic engineering lice
777.
by necovek 11mo ago
I wouldn't go so far: just 50 or 100 years ago (even today in some areas), we had arranged marriages, male dominance (women seen as sexual servants to men) etc. But the study really does not show anything of interest: it seems to have
778.
by necovek 11mo ago
While a couple of significant limitations are acknowledged directly in the article (study only on single adults, education evaluation based on remembering it), the most significant does not seem to be called out: it's really measuring
779.
by necovek 11mo ago
I did not say all regulation is bad, just why big companies like it and push for it.
780.
by necovek 11mo ago
While the original question is not explicit, the mention of "remote jobs" seems to indicate they are talking about YC company employees and not where the company is based out of.

More ›