My understanding is that this remains unsettled law. We've added a disclaimer after seeing the big players (OAI, Ant) do the same and focus on ensuring LLMs get things right / outperforms those less specialized tools.
The only way to stop congestion is to get cars off the road.
Cars are the congestion. There is no other definition, and no other solution.
HOW you get cars off the road is worthy of debate. Pretending that there's a "war on cars" is not and never has been. Cars and their drivers have been subsidized for more than a century, never paying the full cost of the damage that they do to the environment, the roads, etc.
1. Car taxes / registrations should be based on weight and energy usage inefficiency and ratchet up quickly for personal vehicles that are larger than some sensible size based on pedestrian safety at no more than 30 kph. (Drivers of modern pickup trucks and SUVs cannot see pedestrians shorter than ~1.7m at ~5m, and the fact that their fronts are basically walls mean that those pedestrians are more likely to be killed on impact or bounced under the tires.)
2. There should absolutely be congestion pricing to enter downtown locations during the day or other times when vehicular traffic will be high. Yes, that means that most deliveries by transport truck would need to be staged into smaller vehicles and delivered to the businesses needing them, or delivered overnight.
3. Uber, Lyft, and Waymo are part of the problem and should be treated as such.
4. There should be more investment in public transit. Paid for by increased registrations and congestion pricing, preferentially. Fares should be reduced, ideally to free. More use of public transit makes it safer and incentivizes making it faster.
5. There should absolutely be more bike lanes in pretty much every city. It's called induced demand and once the infrastructure is safe, you will see increased usage by all sorts of alternative mobility users (but mini e-motorcycles should be licensed just like regular motorcycles and should not be using the mobility infrastructure).
6. Pedestrian and wheelchair users matter first and foremost over any other road user (bikes, scooters, cars, etc.).
We've had 50+ years of Robert Moses being proved wrong. It's time to design cities around people, not around cars. Call that a war on cars if you want, but it's not: it's a war for peoples' quality of life. Cars can be part of the solution, but they need to be treated like the problem they are first.
For the life of me, I could never get electronics. I used to love the idea of me coming up with electronic circuit designs, but the arcane art of electronics never really clicked for me because I just couldn't intuitively grasp the maths no matter which book I read (AoE, I'm looking at you). But then it hit me, I don't need maths, I just need a formal language to represent the circuits. So over the past few weeks, I worked on a code your own spice (the electronic simulator). So now, for the first time in history, I finally understand how circuits work and how they are designed. And I did this all by coding circuits in python and making my own functional spice (which used to seem impossible at one point, it's surprising how easy it is though).
depends on where you are. In some parts of California you are out $10-20k just for the permitting and panel upgrade.
Then there are proposals to charge homeowners based on insalled capacity. Part of NEM 3.0 in California was a $8/month per KW, but it was defeated. Who is to say it wont be back in 4.0? that would be about a $1000/yr fee for a 10 kw home system.
When using Pi, one way to significantly reduce input tokens it yields is to ignore common bookkeeping "dot directories", such as `.git`. How to do so can be found with the following interactive Pi prompt:
How do I configure Pi to ignore git related artifacts, such
as the project's .git directory?
Other local assets to consider ignoring are `.pi`, `.agents`, `*.md`, and language specific output directories such as `__pycache__`, `bin`, `obj`, `target`, etc.
You can put the `cargo clean gc` command on a daily/weekly schedule then, or configure the `gc` option in your Cargo config. Or set `CARGO_UNSTABLE_GC=true` in your shell init. Or all of the above.
Another thing that this does not measure is philanthropic effectiveness. If there are lessons to be learned in terms of what kinds of philanthropy works better than others, then those learnings may cause other philanthropic endeavors to have greater success. I've had a bit of experience working with several big foundations and seeking out effective ways to spend lots of money is an extremely hard problem and they're all aware of it, and they treat each investment as an experiment.
Of course, the effectiveness of each type of investment is not fixed, and will inevitably drop off. I believe the whole 'just buy malaria nets' first-order thinking was part of discourse for a while before people wised up.
Then lucky you, most of America already caters to exactly what you want. You just have to understand that the desire to not live in a dense community is what causes traffic. You can’t have spread-out housing for most people, and low traffic.
If you want low-density housing and low traffic for yourself, the best thing you can do to make that happen is support and encourage high-density housing and mass transit for as many _other people_ as possible.
I struggle with terminology so I made a little Gnome utility for easier LLM-based terminology lookups from a highlighted word/term + contextual screenshot. So far it's working pretty well, kinda like a better version of the Mac OS or Kindle ones.
> Yes, get-childitem is a little weird to remember,
That's not the problem. The problem is that it's way too much typing, and with the default tab completion behavior it's often way too much backspacing if I didn't get what I want. A good command-line shell designed for interactive use should be terse enough that it's clearly faster than using a GUI. And the OS should ship with such a shell out of the box; building your own on top of a weird scripting language is silly and not at all the same thing as the OS having a reasonable shell built-in. (Most of the value in PowerShell comes from having it installed by default on all Windows systems, and it's sad that Microsoft decided to just freeze it at 5.1.)
I’m not trying to dismiss the investments she made (both time and money), but more point out that she was not driven by the same corrupting motivations you get with typical billionaires. If she was, she’d have taken her wealth and done more capitalism with it.
All of that is intentional. There isn't enough money to be made connecting people. Revenue streams are driven by ads attached to media interaction. Connecting people is low engagement. We, as consumers, have rewarded this revenue model by proving that it works. If you don't want to contribute to it then don't use these large platforms.
>>Humans have been thinking long before writing was invented.
>And look how poorly they did it until they invented writing.
Yeah, look how poorly they did it until they invented writing.
[peers into the dim ancient past]
Sorry, my time-machine spectacles are on leave or on strike today.
And congrats on being a multi-millen(ium)arian, and a mind reader, who could both 1) exist so far back in the past, and 2) be able to know the thoughts of people who didn't write them down, because they didn't know writing.
Hence who I used both HDI and median household income.
And even with GNI my argument still holds. I'm surprised how so many HNers don't remember how poor Ireland used to be until 20-30 years ago, especially given how most HNers are in their 30s to 40s.
Ireland, South Korea, and Israel are the poster children of developing countries that successfully escaping the middle income trap and climbing up the economic value chain back in the 2010s.
The reason the term "singularity" was used when that term was coined was in analogy to a black hole, which has a point on its radius called the event horizon where the gravitational pull is strong enough to stop light from escaping, thereby making it impossible to see beyond that horizon. "Technological singularity" itself here refers to the technological event which causes this, like a black hole is sometimes explained as a gravitational singularity
This is important to the framing of the idea: "the technological singularity" refers to a point in the future past which further developments can't be meaningfully foreseen or maybe even understood. This is often associated with particular paths people think will cause this, but not defined by those. As such, the prior comments are making the claim that you can get there not just by accelerating technological progress, but also by reducing the ability of everyone to comprehend what's happening, which I agree is unfortunately a plausible outcome in the current world
> But does it take 3 years or 4 years to get the engineering degree, after completing vocational training?
You will have the ordinary amount of time at the university.
As I wrote: the time that you save is getting "(fachgebundenes) Abitur" via evening school, which was historically necessary. So you do save time, just not during attending the university.
In the vocational training, one learns practical skills, while the university is there to teach you the theoretical grounding. So, basically nothing that you learned in your vocational training transfers to university. There is a reason why vocational training vs university were historically two completely different and separated kinds of tertiary education.
Code World Models in Simultaneous Move settings like Capital Markets. DeepMind's CWM approach relies on standard MCTS/IS-MCTS, which assumes a single active player at each node.
This doesn't work in simultaneous-move settings like Orbit Wars (or order-book markets), converging to an exploitable pure strategy rather than a Nash equilibrium.
LeCun's JEPA, by contrast, is a learned neural world model, which lacks the determinism, speed, and debuggability of a code-based simulator. Thus, it can drift or predict illegal states, and you can't inspect why it made a prediction the way you can trace a Python function.
TL;DR: The benefit is better auditability and easier RL-like training.
The SM-MCTS extension fixes the first problem (decoupled UCB per player approximates Nash equilibrium instead of a pure strategy) while keeping the second advantage intact (a deterministic, inspectable code simulator).
