And there have been other failures in space based nuclear programs which have caused detectable leaks. The RTGs themselves are pretty safe, but that's because they have a pretty small amount of material (and generate very little power). Reiterating the point a couple of posts up, yes, there are current systems using RTGs--but that says pretty much nothing about the safety of, or tolerance for, nuclear propulsion.

I'm not sure what they have to do with deep space probes or with nuclear propulsion (they were used for electrical generation just like RTGs, but at a much higher power output and with much greater weight.) But yes, they do exist and are a good example of why there's so much skepticism about nuclear power in LEO.

There's a little bit of a difference between thermal decay generators and nuclear propulsion. All deep space probes have an RTG, solar panels don't work when you're far from the sun. Even with the RTGs there's been concern about what to do if there's a launch accident. But direct nuclear propulsion is inherently dirty--in Project Pluto it was actually considered a feature that in addition to dropping nukes the platform could just be flown around irradiating the target. AFAIK nuclear propulsion is only envisioned for extra-atmospheric use, but even getting the fuel up there is a tricky question if every launch runs a risk of dirty bombing a launch facility.

Probably 1. The research suggests that having a human doing nothing except watching and waiting to override a failing autonomous system basically never works--so it probably isn't a big factor in how many accidents google has.

And a safe lane change (signaling, plenty of room) followed by a distracted driver not noticing that there was a car in front would be the bus driver's fault. Now that we've established the basics, we'll have to see what actually happened instead of speculating wildly.

Well, when I said "just apps" I meant "apps and their associated data, everything except pictures, music, podcasts, etc." (which I keep on the SD). So on my current 16G system I lose upward of 6G to the OS. Then there's about 6.5G "used", 1.5G "cached" and a little under a gig "misc". That leaves about a gig free, and I just deleted 512M worth of samsung text to speech bloatware yesterday. So it's not completely full, but it's at the point where I'll occasionally have a large application update fail because of low disk space--and I'd rather just not ever deal with that. And all that is without much in the way of games or movies or other really big space hogs. It's possible to get by with less space but at some point I guess you'd kinda just be better off with a dumbphone if you don't want to do anything on the smartphone.

My wife's 32G 5x has something like 26G available after the OS, so I'd assume a 16G 5x would also have about 10G after the OS. More than enough for her, but the kids these days burn through storage like I can't even fathom.

My current phone fills up 16G just with apps (to the point I have to delete some to manage space). And I've got a 64G SD card for podcasts & pictures & maps & such. (I spend time in places with poor or metered connectivity, and having a local cache is extremely useful.) But everyone has their own needs, and if you don't need the space the 5x is a great choice. I'm mostly just annoyed when manufacturers 1) charge so much for memory, 2) use memory sizes as a way to artificially segment their product lines, and 3) don't offer SD or other removal expansion options.

The 5x would be just about perfect except that it doesn't have an SD card and they artificially limited the internal memory options to segment their market. It's a shame. If 32GB is enough, I agree it's a great value.

I've stood behind an 87 tercel and I've stood behind a new toyota. There is a difference you can smell. 1975 was one inflection point in emissions with the addition of catalytic converters, 81 was another point with the addition of three way converters, 94 was another with the Phase 1 standards, etc. The high mileage tercels were still carburetted, and there's a reason nobody tries to meet current emission standards with a carb. The new cars are definitely cleaner, and most of the emissions controls reduce MPG. (Other things have improved MPG, like aerodynamics, variable displacement engines, etc., so it's not a straight line downward.)

The only pi the beaglebone is over twice the price of is the zero, which I don't consider in the same class. (No network or storage.) Unless you're seeing pi 1 or 2s for $25? (The beaglebone black is $50.)

If you're looking for open why on earth are you looking at the raspberry pi instead of the beaglebone? Graphics are basically the only advantage the pi has over the bone, so if you take that away you've basically got a bone with fewer I/O options and a lousy network interface. I don't get it.

Note: it looks like ass

Why on earth would you do that? These devices are not exactly memory starved.

Old imports got incredible fuel economy. They also weighed as much as a go-kart, had zero safety features, and spewed emissions like they were going out of style (which, in fact, they were). Apples and oranges.

This kind of thing comes up a lot, and seems to come from people with no clue how roads work. Repairing the surface is the easy part. Repairing the subsurface is hard, and putting some glass on top isn't going to change the fact that you've got a major repair involving a lot of earth moving equipment if the subsurface of a road is compromised. Just throwing a new piece of glass over a sinkhole isn't an option.