> Second: Colossus' plan - and its ultimate actions - assume everyone else on earth is a nuclear-disarmed-rational-actor

The plan would still work. Colossus couldn’t be destroyed with nuclear weapons and would retaliate against any attack. It could force compliance of conventional forces as well, and force automation on them, also force populations to rearm it.

In the end, the population would appreciate the eradication of poverty, hunger, disease, and the surplus from not maintaining military capabilities. Colossus could afford democratic institutions while acting as a guard rail against humanity’s worst impulses.

I wish it expanded on the //e hardware rather than becoming something else. I get the II video hardware was pushed to its limits and something better was needed, but the new video processor with an entirely different approach didn’t feel like a II. Fill-mode was cool, but I don’t remember it being used for anything other than demos. A blitter and sprites would have made more sense, as would colored text modes, redefinable character sets…

The sound chip was so capable it contrasted with the rest of the machine.

It’s not really an 8-bit computer, stuck in a weird place between the //e and the Macs. It kind of is a II, but only in a half hearted way, kind of a new Apple ///, but designed for graphics and sound, and nicer at pretending to be a //e. It’s not even the fastest II - that would be the //c+ - nor the last to be discontinued, which was the Platinum//e.

It’s a nice machine, but I find it uninspired, a bit like the C128, which instead of improving the VIC II, added a VDP with garish RGBi colors. Both look like they tried to check all boxes and, in the end, made computers unable to follow on their immediate ancestors legacy. Both disappointed me because they could be so much more.

> I really wish they'd crammed a SID chip into the Amiga alongside Paula

This is something the Apple IIgs had. It had an extremely capable synthesiser with good graphics and performance capped so not to compete with Macs. It was a weird machine, a sharp contrast with the minimalistic Apple IIs that preceded, over complicated and trying to be too many things at once.

For the same reason I prefer the design of the ST over the Amiga’s. Amiga made lots of assumptions about the use that ended up tuning it well to platform games and NTSC video editing, but nothing else.

I don’t think the flash storage would keep data integrity for that long without a lot of redundancy. Also, processing capacity would be limited as well. Is there a problem that can be solved with a LEO server that can’t be better solved with the same amount of money sitting in a datacenter somewhere?

What would be good cases for data-heavy applications in expensive places?

To me it’s clear adding the ability to express intent to parallelise is the Right Thing. This is the only way the compiler can actually know what you want it to do.

One of the smallest sculptures in the universe.

I believe the rationale is that they are so much denser than they will compensate the price difference over the entire lifetime of the system. A 10 TB install is a full rack otherwise and this is 5% of the space. Colocation for HFT tends to be expensive and using 5% of the space for the same amount of data might make sense.

IIRC, metals aren’t great shields because of secondary particles emitted after the first one hits the shield.

What is the designed lifespan of a Starlink satellite?