WanderPanda

Yeah I was wondering the same! I fondly remember coding swift almost 10 years ago but large part of it was the good integration with Xcode I believe. Seems like they have an OSS LSP so probably that experience translates to other IDEs nowadays

Cargo culting the stock market

Interesting thought experiment! How would they best take advantage of the weights and what would be signs/actions that we could observe that signal it is likely they have the weights?

I wish Symbloics.jl ‘s rewriting/simplification rules were as good as SymPy’s/Matlab’s. The composability of the Julia ecosystem would be really great for symbolic math

You’re enumerating all kinds of failure cases and how the most important thing about the product is how fast/easy to rectify them but you should really consider the probability times the effort (plus some risk penalty for increased risk).

Also driving to the store (plus keeping excess stock of everything) to buy something must be much worse than just shipping it in a minimal box.

That’s my mental model as well. The incompressible fluid-based explanations never made much sense to me

How does it compare to MuJoCo?

For many systems the optimal PD can be seen as the right answer (LQR)

What do you mean by non-deterministic hardware? cuBLAS on a laptop GPU was deterministic when I tried it last iirc

It will probably look like a keyboard. Keyboards are so lindy at this point they didn’t even adjust the layout (from that path-dependent mess we are dealing with), let alone the form factor

I was wondering if it is possible to formalise this and cast it into a book or lecture so that new devs don‘t have to grind out ten years of hard-gained experience. But at this point I think the only real way to really learn this is to go e.g. through dependency hell yourself at least once.

Author here. We did not pre-programm basic control or feedback parameters. The trained policy (fully connected neural network) takes the state estimate (position, orientation, linear and angular velocity) and a history of the previous actions as the input and directly outputs the PWM/RPM setpoint (uint16 that sets the pwm interval).

The code is open source, you can verify it yourself ;)

https://github.com/arplaboratory/learning-to-fly

Author of the referenced work here. I have to disagree with this, at least with the motion tracking system part. The optical flow deck [1] works like magic and can maintain low drift positioning even when doing relatively fast and agile maneuvres.

For an example you can have a look at our video [2] (e.g. the disturbance rejection tests at 3:00 and after are solely using the optical flow deck)

[1] https://www.bitcraze.io/products/flow-deck-v2/

[2] https://www.youtube.com/watch?v=NRD43ZA1D-4&t=198s

What they are highly optimized for is mixed-precision GEMM (like all other accelerator manufacturers). What distinguishes Nvidia for now (imo) is that CUDA cores are also quite good at normal code (with control flow etc). I used to think that being close to optimal in one of them would contradict being close to optimal in the other but it turns out they share a lot of resources (SRAM) and the overhead in chip surface if one or the other is laying dormant seems negligible. I'm pretty sure that AMD et al will be sufficiently successful at blatantly copying the CUDA API that we will see serious competition in the next years. The bigger source of uncertainty might actually be fabbing capacity.

I find it hard to argue that this mode supports a 1.7T valuation. I find it hard to believe that for a couple of billions + TSMC credits no one would be able to recreate the CUDA ecosystem + hardware in the medium term.

IIRC about 40% of us politicians are lawyers, unfortunately I’m sure they will find a way to gatekeep these revenue streams for their peers.

Do the sims include functioning VOR instruments these days? That would be quite helpful during the night :D

I highly doubt a human can beat the market consistently looking at the timeseries of some ticker alone.

They probably mean the manifold of the universe

Why is that in turn? I’ve heard people refer to low vitamin D levels but I’m sceptic if that is the reason (would be so easy to fix)

Anyone confused about the horizontal axis’ tick labels? Usually the range befor e.g. “5” is 4-5. But here it would be 2003-2004 in case of “2005”. I never consciously thought about this but is this way of indexing the standard?

I remember chasing one of the iridium flares and it was very bright, even during the day

This sounds like motivated reasoning to me. Having an above-average personal tutor that doesn't get mad or tired 24/7 every time seems like a big multiplier

You raise a very good point. If you join a company with the goal of retiring in the same company after a a 40-year career, the incentives are much better aligned than with contractors. Maybe the demise of the 40-year career in modern companies is one of the bigger elephants in the room

Why don‘t they scale up? They seem to have all the regulatory capabilities to get a commercial jet on the market already. I also never understood why e.g. China doesnm‘t just buy Embraer (or take over all their employees) and builds out from a functioning product?

FTC should consider splitting the commercial and military branches of Boeing. The military part is hopelessly entangled in lobbyism etc. but the commercial branch should be as removed from the gov as possible

They didn‘t maximize profits at all. Why do people operate on the principle „if it is bad it had to be because of maximizing profits“? This seems to be much more about corruption and misaligned incentives as well as bigcorp bullshit-job rot. There is really only one thing that can cure this and that is more fierce competition

I would assume margins are razor thin in the music streaming business, there are only very tame moats

Is that changing of global processor flags a x86 feature or does it hold for arm as well?

dns clients come preinstalled on virtually any os