I think the Navy's primary long-term interest in this is a defensive measure against ASM tech, and it's frankly, a good application.

We're not going to be seeing gently-rolling guided cruise missiles any time soon (and certainly not for $10k) and we're not going to see long-range guided rolling rockets.

Anti-ship missiles are getting faster and faster, and CIWS is getting less and less likely to work. Aegis cruisers can already take out most ballistics that would threaten a carrier group. CIWS needs replacement- mounting lasers on support cruisers to train on a fast cruise missile is perfectly legitimate, and it increases the usefulness of having carrier groups to begin with.

I think the power-delivery capability is going to greatly outpace the defensive capabilities of ordnance. The kind of cooling necessary to stop that kind of energy is massive, and unlikely to work well on a cruise missile, and the only real defense- thicker armor is going to make it harder to keep the missiles as fast as they need to be to have any chance of success.
This isn't a waste of money, it's literally the only hope for keeping carrier groups relevant.
Mirrors will never be credible way to counter a high-power laser threat. Not even internally cooled. Mirrors are simply too easy to damage, and they're not a mirror after that happens.

They've been destroying 60mm mortar rounds from 500m out with 20kW lasers since 2006, what exactly is the basis for your disbelief in the usefulness of laser point-defense?

This is the only realistic protection. Rotation and ablative shielding. Reflection is folly.

Small portions of thin layer of beryllium is destroyed in some small fraction of a second, copper is hit, copper heats up, it and beryllium turn into slag en masse. There's no realistic battlefield mirror scenario that defends against 50kW of light. That's not to say 50kW of light is guaranteed to kill whatever is flying at it, but a mirror just doesn't buy enough time to matter.

Bad idea. The idea is to prevent the mirror from heating up. Covering it in the ablative equivalent of tinder is not a solution to that problem.

You are right- but it takes just a few J/cm^2 to make a mirror no longer a mirror. That's the cost of high reflectivity. A strong ablative armor is a superior route for laser protection. Unfortunately, that's also heavy and not well suited to guided ordnance.

7) Figure out how to design a guided missile that rotates at a high rate

I don't even see it being pitched as ABM. The congressional briefings on the system even mention that many-MW-scale lasers would be required to injure an incoming ballistic RV. This may however be highly useful for carrier protection- especially if not located on the carrier. Cruisers in the carrier group, spaced properly, could very likely take out ASM projectiles, especially since they tend to be cruise-missiles, not rotating warheads, and being able to target the incoming weapon from the side as opposed to head-on would limit the atmospheres disruption of the beam.
Most potential enemies that have ballistic ASMs are using shitty enough ballistics that we can intercept them with standard issue ABM systems (Aegis cruisers).

You're right- though at typical projectile velocities, the atmosphere makes a piss-poor coolant. Why pulse the laser in nanoseconds if you have a laser with sufficient cooling and available power as to hold the beam on the target? Obviously, we'll need MW lasers if we ever want to take truly well shielded and rotating/tumbling projectiles out, but for most things flying through the air, being able to hold 50kW on it for a reasonable amount of time is brutally devastating. Even coated in mirror.
Lockheed has demonstrated in-flight subsonic rocket destruction with 10kW lasers.

Oh, quite simple. Because 2.5kW in a that small of an area is more than enough to make that mirror no longer a mirror. You can try this at home, if you have sufficiently tough conductors. 10kJ is obscenely high for an estimate of what it takes to oblate a mirror surface. Usually a few J/cm^2 is enough to do the trick, and degradation (micropitting) starts happening at the 1J/cm^2 levels. ITER has some publicly available studies on this exact topic. Assuming a focal area of 126 cm^2 (roughly 5 inch diameter) and 2.5kW total absorbed energy, we're talking ~20J/cm^2- enough to turn the mirror into slag in fractions of a second. Now, in the case of a fast-rotating mirror object, this may be a lot more difficult to capitalize on, so ballistic RVs may be out of the question, but something like an anti-ship missile, or a cruise missile would be easy pickins'

From someone named MrL0G1C of all things. Thank you for the laugh

LOL. Are you serious right now?
And that is why you fail.
So, some people that were smarter and more committed than us ended up being genocidal maniacs, and thus the economic theory they started out championing is bunk. Does anyone really need to blow holes in that logic?

Here in the Pacific Northwest of the USA, I can say we had a snowless winter down in the lowlands, which is pretty unusual.
What's really unusual though, is there's no snow on the mountains bordering the lowlands. It didn't even snow much up there. The mountains in January looked like they normally do by about August. It was a pretty weird year. There's talk of drought worries now, since our water supply is snowfall runoff in the foothills.

Agree 100%

All sample sizes are statistically significant, you just need to understand the error bars.
In this case, assuming a true accident rate of 4.5% (human-driven rate, bad assumption- but it's all we've got), a reasonable distribution, and a population size of 48, you have a 20% possibility of getting 4 accidents in the sample (8.3%), which is far too high to call conclusive by any stretch. A better conclusion would be: Very weak correlation found between self-driving cars and increase in accident incidence. Need more data to see if correlation is real.

Sigh. If you took a sample of 48 US human-driven cars, you'd have a 20% chance of getting 4 accidents out of that 48.
I'm not trying to say that's huge, but that sample size of 48 is far too small to draw conclusions from statistics that contain millions of samples. The error bars are massive.