Sorry you’re getting down voted — lots of replies from folks unclear on what the diffraction limit means, atomic resonances, etc.: https://www.newscientist.com/article/2161094-a-single-atom-is-visible-to-the-naked-eye-in-this-stunning-photo/

https://www.newscientist.com/article/2161094-a-single-atom-is-visible-to-the-naked-eye-in-this-stunning-photo/

Parent didn’t say resolve, they said see — you can’t resolve stars but you can most certainly see them.

Light up a single atom enough and you can see it (unclear if this works with a dark adjusted naked eye or if a long exposure is required): https://www.newscientist.com/article/2161094-a-single-atom-is-visible-to-the-naked-eye-in-this-stunning-photo/

No, they’re too small to resolve. You can see small things if they’re bright enough: https://www.newscientist.com/article/2161094-a-single-atom-is-visible-to-the-naked-eye-in-this-stunning-photo/

A single atom of gold is far too small for any photon in the visible spectrum to interact with.

That’s incorrect — single atoms can, and do, interact with optical photons.

https://arxiv.org/abs/2410.19671 https://www.nature.com/articles/ncomms13716

And the entire field of super resolution microscopy relies on small things (e.g., molecules) interacting with light.

The energy from nuclear reactions can be astonishingly large (compared to, say, chemical reactions).

But atoms are really, really, really small.

Most Linux filesystems, being case sensitive, won’t find the SUDO command.

Yep, you’re right — I was just responding to parent’s comment about fiber being best because nothing is faster than light :)

That’s…not really a cogent argument.

Satellites connect to ground using radio/microwave (or even laser), all of which are electromagnetic radiation and travel at the speed of light (in vacuum).

Light in a fiber travels much more slowly than in vacuum — light in fiber travels at around 67% the speed of light in vacuum (depends on the fiber). In contrast, signals through cat7 twisted pair (Ethernet) can be north of 75%, and coaxial cable can be north of 80% (even higher for air dielectric). Note that these are all carrying electromagnetic waves, they’re just a) not in free space and b) generally not optical frequency, so we don’t call them light, but they are still governed by the same equations and limitations.

If you want to get signals from point A to point B fastest (lowest latency), you don’t use fiber, you probably use microwaves: https://arstechnica.com/information-technology/2016/11/private-microwave-networks-financial-hft/

Finally, the reason fiber is so good is complicated, but has to do with the fact that “physics bandwidth” tends to care about fractional bandwidth (“delta frequency divided by frequency”), whereas “information bandwidth” cares about absolute bandwidth (“delta frequency”), all else being equal (looking at you, SNR). Fiber uses optical frequencies, which can be hundreds of THz — so a tiny fractional bandwidth is a huge absolute bandwidth.

80% of the USA lives within urban areas (source). Urban “fiberization” is absolutely within reach.

Agree that running fiber out to very remote areas is tricky, but even then it’s probably not prohibitive for all but the most remote locations.

So the irony is

I see what you did there…

I think you mean more scrupulous, not less.

Bonus points: use non-qwerty keyboard for added obfuscation (but keep the qwerty key caps of course).

To each their own though? I can’t imagine why anyone would want something other than i3 (or similar), because almost by definition the DE is not the program I fired up my computer to interact with, and i3 “gets out of the way better” than most others in my experience.

But…that’s just my use case. It’s a horrible UX for most people, just happens to work well for me.

I feel old…when I was learning how to run Linux I started with an old 386 (maybe 486?) my dad wasn’t using. I think it had 32MB RAM, which was fancy for those machines.

We had dial up at the time, so only one machine could be on the Internet. So, I set up a modem on the x86, plugged into an Ethernet hub (switch?), and learned enough ipchains (this was before iptables) to share a connection. It also ran Samba, an AFP server, and probably FTP and HTTP (just for local access) — but it worked for filesharing.

It could also run MP3 streaming software which amused me because the machine itself was too slow to decode MP3 (but that’s not necessary to stream).

Compensation for engineers in the Bay area will average much higher than $200k, and that’s not counting benefits (medical, etc.). So cost to the company will be way higher than 200k/employee.

For a project that has hardware, there will be large expenses associated with that — custom silicon has huge setup costs, for example.

Maybe a dumb take, but I think milking customers for all they’re worth is much better option than what HP is seemingly doing — which is milking them for all they’re worth this quarter.

Like, there are companies with a cult like following (Valve comes to mind) and while they could probably increase profit for a quarter or two, they seem to be playing the long game fairly well. Which is ultimately better for everyone — they get more money over your lifetime, and you get a product that you’re happy with.