Maybe we need more benchmarks that involve human collaboration instead of solo task solving.
Since they have become so capable the new bottleneck is what they can't know. The stuff that's inside people's brains who work in real companies with products and processes absent from any training set.
100%, it's not like adults are immune to misinformation, social media addiction or scams. Those platforms should be safe for all people regardless of their age, nationality, sexual orientation, gender or religion.
We could just start making it illegal to advertise to children. Without age verification, it's just a burden of proof with the advertiser that their ad isn't targeting children and a ban on age specific target groups for ad platforms.
Reading this as "why stop with ad restrictions there" rather than "why only stop ads there":
Salami-slicing, and appeal to a more appealing proposal and target group. Adverttising-to-children is inherently controversial, already banned in many countries / jurisdictions, and is a relatively easy sell. If there's going to be a reform in how advertising is perceived and regulated, it's less that this is where to stop than where to start.
Because it's not preinstalled on every machine. Bash is a good target for portability reasons, but it's a shitty language to write in. If ever I get in the position again to have to write some bash, like for an installer or so, I'm going to be looking into Amber again.
Portability is dead now anyways. Something can have "sh" but all you can depend on are sh semantics and built ins. The whole point of a shell script is to compose a larger workflow out of the command line tools preinstalled on a system. I can't guarantee a sh script written on macOS is going to work on FreeBSD or Linux anyways, so there's zero reason to limit myself to sh.
Given that you reuse Postgres' tests and LLM have been clearly trained on Postgres' three decades of contributions, this may constitute a license violation. Unless you include the original license of course. From a human level I also understand how you ended up with a less permissive license.
It's a fork from Maps.me, the streamlined map app popular with normies. I myself use and love OSMAnd, but in the travel communities I am active in, most people react badly to OSMAnd as something arcane and nerdy.
The performance advantage is small to nil, IME. The UI is admittedly cleaner, though partly (as you say) that's simply because it has a ton less features.
The difference mostly comes from the default detail levels and shape optimization. You can make it as fast in OsmAnd as both are GPU-accelerated, but OsmAnd renders the map more detailed by default.
Same in OsmAnd though. Someone recently posted a screenshot of an area they thought was badly mapped (with AI or at least not using the "make square corners" tool). Turns out to be Osmand's coordinate rounding as part of the map data compression
Comparing a building in my city between the two maps by flipping between the apps, it looks identical to me. Not a scientific test though
Yes, you can toggle a setting to use the network colors. I use it mostly for hiking. The UI is a learning curve, 100%, I am curious to check another tool
Is that going into effect globally? It's hard to believe this will sit well with Japanese customers where second hand gaming markets are, dare I say it, culturally relevant.
reply