How do you make a small fortune?

Start with a large fortune and buy a boat.

Newer macOS is not Unix certified.

It’s UNIX 03 compliant https://en.m.wikipedia.org/wiki/Single_UNIX_Specification

One or two Linux distros were (are?) UNIX certified, though.

Sounds like he was a mantis and was posting while copulating.

Haha yeah that was the counter example I was thinking of. I agree completely — you could make a Gentoo from source beginner distro, and I think you could make it reasonably “idiot proof,” but it would still be a bad user experience most likely (too much time spent compiling).

If your distro can’t be forked into a “beginner distro” then it’s fundamentally flawed IMHO.

To be clear, I’ve used Arch as my daily drivers for a while, and while it’s not the best fit for my needs (I use Debian mostly), there’s nothing that I experienced that was incompatible with a “beginner” distro.

No love for us Dvorak users :(

You can also drop cache for debugging by running something like echo 3 | sudo tee /proc/sys/vm/drop-caches

But remember that the kernel knows best — this RAM will automatically be freed up when needed and you should never run this except for debugging (or maybe benchmarking).

Sounds like it was a 2 petawatt pulsed laser, with picosecond pulses, so 2kJ/pulse. Staggering amount of power and energy for a pulsed laser!

Note that it’s not CW, so the average power will be much, much, much less than the pulsed power. Too lazy to find the rep rate to see average power.

I’ve been super happy with it. Knock on wood it’s been super reliable. I have a single ZFS drive, take snapshots with various retention policies, nothing fancy.

Another fun thing is to set up a reverse proxy on it as an endpoint for services on your local (home) network which can only be accessed by VPN. For example, my Jellyfin service isn’t public facing, but I didn’t want e.g. my parents to need to set up WireGuard. So instead they can point their TV to a raspberry pi on their network to access the service — even a first gen RPI can handle Jellyfin reverse proxy over WireGuard for moderate bitrates!

Remind me again, what color was Obama’s scandalous suit?

WireGuard, and an external HDD. Run at a remote location for off-site backup.

I do this with a raspberry pi 3 at the in-laws. I copied the data over locally before setting it up, and after that it’s just nightly incremental rsync, which is fine even over my slow (35Mbps) upload.

You can turn it off, at least for ext4: https://lwn.net/Articles/784041/

Although you can use case insensitive filesystems with Linux, and case sensitive filesystems with macOS. I believe the case sensitivity is a function of the specific filesystem — but yeah, practically, the root for Linux is always case sensitive, and APFS ain’t is only if you ask it to be ( https://support.apple.com/lv-lv/guide/disk-utility/dsku19ed921c/mac ).

No, that’s not really a useful way of modeling it for the case of light traveling through a linear medium.

The absorption/re-emission model implicitly localizes the photons, which is problematic — think about it in an uncertainty principle (or diffraction limit) picture: it implies that the momentum is highly uncertain, which means that the light would get absorbed but re-emitted in every direction, which doesn’t happen. So instead you can make arguments about it being a delocalized photon and being absorbed and re-emitted coherently across the material, but this isn’t really the same thing as the “ping pong balls stopping and starting again” model.

Another problem is to ask why the light doesn’t change color in a (linear) medium — because if it’s getting absorbed and re-emitted, and is not hitting a nice absorption line, why wouldn’t it change energy by exchanging with the environment/other degrees of freedom? (The answer is it does do this — it’s called Raman scattering, but that is generally a very weak effect.)

The absorption/emission picture does work for things like fluorescence. But Maxwell’s equations, the Schrödinger equation, QED — these are wave equations.

Dispersion and nonlinearities would like to have a word ;)

*in vacuo

Not sure why you’re saying Python forces everything to be object oriented…?

I particularly like the truck/engine correction.

good enough simulations that you can’t tell the difference.

This requires us having actual conversations with those dead people to compare against, which we obviously can’t do.

There is simply not enough information to train a model on of a dead person to create a comprehensive model of how they would respond in arbitrary conversations. You may be able to train with some depth in their field of expertise, but the whole point is to talk about things which they have no experience with, or at least, things which weren’t known then.

So sure, maybe we get a model that makes you think you’re talking to them, but that’s no different than just having a dream or an acid trip where you’re chatting with Einstein.