A nuclear power plant (and its control room)?

I would recommend reading The World Without Us, which examines what would happen to the relics of our civilization if we, humans, suddenly disappeared (i.e., not extinct via war or disease, but just hypothetically got raptured away). Nuclear power plants don't fare so well. In fact, without human attendants to control them, cleanly power them down, and then decommission the plant several decades later, there is every likelihood that some nuclear plants would experience catastrophic accidents within weeks to years that would spew radionuclides all over the countryside.

Before launch you have connector to which you connect a computer and you can do a self diagnose on the satellite using that connection to the on board system. There is no reason to dedicate leds and leds wiring for that especially that you will need to check for many fail conditions.

Having spun a number of boards in my career, I can tell you that it is trivial to add an 0402 LED indicator, just as an indication that the 3.3 V logic rail is powered. And because it was easiest (via inertia) to keep it in than to cut it out (even as a do-not-populate instruction to the board house) that little LED stayed in the design, even though in production no one would ever see it.

Given the complexity of most satellites, I would be deeply surprised if there wasn't at least one LED on one of those boards.

The poorest drivers probably own the lion's share of them. Individuals are likely even aware of their vehicle's condition.

You may be correct, but when I think of automotive air pollution, I think of the ass-hats in jacked up pickups that go down the highway coal-rolling Priuses. Those kinds of trucks may be shit-kickers, but many of them actually have thousands of dollars in after-market parts and modifications. They may, ironically, also be driven by very poor people who ought to prioritize their spending better, but there you go.

Note TFA has a redshift(z) scale that is backwards. They have z=1 at 6 billion years, and z>20 at 200 million years.

I puzzled over that for a moment, too. What the time scale shows is age of the universe, or (as the scale is labeled) years since the Big Bang. So z>20 = universe at 200 million years old, not years ago. It's confounding and, to my eyes, counterintuitive, but perhaps that's how cosmologists work.

parts per trillion doesn't make for much of a problem in any case

There are plenty of contaminants in water that would be a serious problem at the parts per trillion level. Whether these chemicals are or not is, I think, not yet demonstrated.

One 4th grader calls another a "doody-head." Doody-head calls the first 4th grader a "poopy-face"

Most every GPS watch produced by Garmin, Suunto, Polar, etc. has exposed gold contacts on the back face or edge for charging and syncing (usually over USB). Such watches are used by active people doing sweaty activities in all kinds of conditions: triathletes, hikers, bikers, etc. Gold contacts embedded in plastic does not present a water ingress problem (nor, except for a few isolated cases, adverse skin reactions). Doing it properly requires good design and manufacturing control, but Apple has made a brand out of doing just that.

Contrary to some peoples belief: The sun still exists during winter. Panel efficiency falls by half during cloudy weather, but you will still get some power. Just means you pull more from the grid during winter.

Actually, panel efficiency (Watts of electrical output divided by Watts of insolation) tends to increase in winter, because photovoltaics are more efficient when cold (conversely, the efficiency falls when the solar cells are hot). The output of the panel will fall in cloudy weather, because the total insolation will be less, but the efficiency may well increase.

We have to spend billions to upgrade the grid, to handle "Green" power sources that are more expensive than their competitors

Perhaps you haven't noticed, but the U.S. power grid has been having all manner of problems as it is: single-point failures that affect whole cities or entire regions, mismatches between supply and demand that allow Enron-style speculators to manipulate markets, deferred maintenance tallying tens or hundreds of billions of dollars, externalities associated with conventional generation sources that aren't properly taken into account (e.g., nuclear and fly ash disposal).

And those are the problems that we have today, with the grid as it presently exists. Even if no further renewables were added to the grid and Tesla closed up shop tomorrow (both of which are miniscule sources and sinks compared to the grid), we would still need to invest hundreds of billions just to keep things from getting worse. If we're going to gradually rebuild the grid, we should be rebuilding it sensibly: for increased robustness, efficiency, and flexibility. Yes, that means that it can also accommodate renewables and electric vehicles, but that's a secondary motivation.

How many Prius would you need to carry 7 people, plus 7 suitcases of stuff, plus tow an 8,000 lb trailer?

I own a Yukon XL, which is the GMC version of the Suburban.

I am not attempting to troll by asking this question, but I am curious: what percentage of the miles driven in that Yukon have just 1-3 people, and little luggage? What percentage of the miles are driven with 7 people, 7 suitcases, and an 8,000-lb trailer?

That was my reaction, too: what advantage does putting it on the ISS get you, balanced against any added costs / difficulties / constraints, compared to a standalone satellite? On the plus side, you have an abundance of solar power available, along with cooling loops, attitude control, etc. What is more, you may be able to make use of astronauts to provide debug and repair capabilities. On the other hand, you are limited to the ISS' orbital inclination and altitude, and probably a variety of added design constraints and safety standards.

the probe is not designed to transmit while aiming its instruments (to save money; contrast with Voyagers)

Voyager has most of its instruments, including the cameras, on a movable platform. This allowed the positioning of the spacecraft and its high-gain antenna (the dish) to be decoupled from the positioning of the sensors. That made it very versatile and capable but, as you mentioned, more expensive. It also increases technical risk. What if the scanning platform jams up? (Some instruments could end up forever pointed back at the spacecraft! There are only so many multi-spectral selfies you would ever want to have.)

New Horzions is, for all intents and purposes, a single solid body. For 98% of its operational life, it's spin stabilized with its dish pointed squarely back towards Earth. That won't suffice for the intensive observations it was built for, so it will stop spinning and tilt itself this way and that to point its sensors at Pluto during its close encounter. Of course, when it is tilting this way and that, it is no longer pointing its main dish at Earth, so there can't be substantial communications. There is still the low-gain antenna, which is much less directional, which will allow for continuous commanding and telemetry, but has too little bandwidth for much science data to be beamed back. (more info here)

Well, I can already predict what the final sentence of this novel is going to be:

Nosaer tnerappa on rof dna gninraw tuohtiw pu welb noom eht.

That was my reaction, too.

[reference]

Why not use the grid as a reservoir..like a battery or capacitor?

The primary reason is because there is no appreciable storage or capacitance in the grid. Every watt that is generated is balanced by a watt of consumption, second to second. That is what the purpose of an Independent Service Operator (ISO) is - forecasting supply and demand on a minute- and hour-basis, and ensuring that balance is maintained. When instantaneous supply does not match demand, the grid becomes unstable. This usually first manifests as frequency drift - it'll rise or fall from its nominal 60 Hz (50 Hz in other countries), followed by voltage drift. Eventually you get bad things like failing transmission lines and transformer fires.

There is some flexibility on both the supply side ("dispatch" power that can be ramped up very quickly) and some on the demand side (certain large users - factories and such - can temporarily throttle their usage in exchange for payments. Residential customers can have their air conditioners temporarily switched off by the utility - again in exchange for payment). There is a small bit of storage - pumped water hydro, flywheels, batteries - but it's pretty localized, and minuscule compared to the total energy utilization. There is also some dispatchable demand to take up excess supply - ice generation and hot water generation - but that, too, is pretty small peanuts.

Some efforts have been made to make a bigger business out of grid-tied storage, but they haven't been ringing successes just yet due to the large capital costs. If someone could cut the capital cost of storage in half, you could expect to see more large-scale deployments, probably in conjunction with intermittent renewables. This may well be Elon's master plan.