If nobody cared about civilians why are we killing less of them then in any previous campaign?

I consider myself a leftist, but one of the things that is simply true about military policy is we don't get it. We run in at the last minute, notice a lot of death happening, and condemn the death as unprecedented Un-American evil without bothering to learn any of the precedents. We spend half our time condemning them for doing things that may or may not be stupid/evil/pointless/etc. (we really don't know), and the other half shoveling money on military retirees. And it's dumb.

But we've got a lot of Professors, so we make it sound smart. To people who can't figure out we're clueless, that is.

That's not a lot of dogfights for an operation involving hundreds of aircraft and 100k sorties. Moreover most of the Coalition aircraft involved are F-15s. The F-15 has been replaced by the F-22, which implies that a doctrine based on having one good dogfighter and everyone else works. F-35 is the everyone else replacement.

And the F-16 is such a terrible design that it only sold 4,500 units to 24 countries. Much worse then the twin-engine F-18 sales of 1,480 to 8 countries.

Moreover it's hard to do twin engine VTOL, and the Marines insisted. They probably should not have been humored (or should have been humored by being allowed to by their own, special, Marine Corps plane), but once that decision was made twin engines went out the window.

Good luck detecting it at 30,000 feet with your eyeballs. Particularly in bad weather.

If the electronic warfare suite they have works like they're planning, then they'll have blown you up before you get into position.

Probably pretty badly for the JSF. F-22 is supposed to be the one doing the dogfighting.

But a) they're only planning on buying 150 (and only 50 by 2020), while we already have 115 and will end up with 2,400+ and b) if the plan works dogfighting is irrelevant.

The thing that sucks about spending $1 Trillion on combat aircraft is it's a fucking lot of money you could have used for any number of other things, but the advantage is that you have a $Trillion worth of combat aircraft. At 16-to-1 odds, with better trained pilots (we typically have several hundred hours on-type, the Russians don;t have that kind of budget), you don't need a bette plane to win.

The civilian to combatant casualty ratio is lower for the drone war then for literally any other method invented.

As a non-pilot I can't tell you why those weddings were picked. I suspect there could be a number of factors -- maybe the target only leaves hiding to go to weddings, maybe the strike was considered important enough to do immediately regardless of the collateral damage, maybe this was the only time three bigwig terrorists were all going to be together, maybe some idiot fucked up the coordinates (this happens in warfare), etc. Hell maybe somebody else bombed the wedding and counted on everyone to assume it was the drone war (this kind of thing happens in Pakistan).

I can tell even those wedding strikes tend to be fairly low carnage. I was at a wedding on Saturday, and you could have killed dozens with a single grenade on the dance floor, and yet casualties from wedding strikes are almost always limited to one or two cars. Moreover there haven;t been any particularly recently. The last I could find was December of 13.

Because they know they just made a fairly normal teenage girl a homeless orphan. If they have kids they probably sympathize with the poor guy who is being deceived by his daughter even if his day job is terrorist mastermind.

Fighter pilots and bomber pilots don't actually need to dehumanize the enemy. They simply never bother to learn that enemy is, in fact, human. Those people are lines on a targeting screen that are only occasionally even human shaped (most of the time they're tank-shaped, or house-shaped, because you don't get close enough to see make out details like people). Drone pilots have to do it 24/7.

So you're arguing that it's not going Mach 1 because it's designed to only go Mach .99868593955. Apparently you don't believe in rounding to significant digits. As for typical Jet Speeds, those are below Mach .9, and typically in the Mach 0.8 range. Mach 0.15-0.2 is a pretty significant difference.

Moreover you are talking about a different G-Force measurement then either me or the standard Musk has designed. When taking turns the relevant measure is the lateral G-Force measure, and Coasters don't go too high on that because if they did 6 Gs laterally a 200 lb dude would be exerting 1200 pounds of pressure on the side of their roller-coaster, and it probably can't handle that. They top out at 0.5 Gs lateral. Aircraft don't do a lot of lateral Gs either, partly because they have to go straight really fast or fall from the sky, partly because they have even less structural strength to resist extra pressure on the fuselage, but mostly because people start throwing the fuck up once you break 0.2 lateral. They try to stay below 0.15 laterally unless something has gone horribly wrong.

The whole plan is vintage Musk -- equal parts brilliant engineering, pooh-poohing the standards literally everyone else uses (0.2 Gs laterally is what everything in transportation, except Hyperloop, does), unwarranted financial optimism (his $700 Million tunneling budget won't get him all the way through a single mountain, and Cali has a couple ranges of the damn things), and excellent spin (for example, his planned route technically does not go from LA to San Fran, it goes from 30 miles from LA to the San Francisco Bay, which reduces his costs by roughly 75%, but nobody calls him on it). Given some development the engineering could be very interesting, but you really need a lot of that development.

Current planes are much superior at fighting World War 2, where dogfighting at speed was the norm.

For future wars it's not quite so clear. The last time we engaged in any real dogfights was Vietnam. The Iraqis, who had the planes to dogfight us in the first war, fled to Iran because they figured they'd die of massed missile fire before they got into cannon range.

The theory behind F-35 is that it's virtually impossible to detect, and we have the electronic warfare capabilities to detect anything anyone else actually has. That means that F-35 should be able to fly around at Mach 1.6 with being targeted by the enemy (who don't even know where it is), while firing off it's missiles whenever an enemy aircraft gets into range. It's more a submarine or cloaked starship then a fighter craft. If it works it'll revolutionize aerial warfare and instantly make every Air Force in the world obsolete. Especially the one belonging to Vladimir Putin.

Drone pilots more removed from the action than infantry? Hell yes.

More removed then the rest of the Air Force? Hell no.

The way a drone strike works is a drone loiters on station for weeks on end. During this time the drone's pilots figure out who is in the house when, so they can avoid blowing it up when the local equivalent of the Girl Scouts are in the living room. Which means drone pilots know when their target takes the trash out, whether the teenage daughter has a boyfriend who sneaks in sometimes, etc. This makes for attacks that are much easier on the civilian population then normal bombing, because you can skip the night when the girl and her boyfriend are enjoying themselves, but it makes for very stressed out drone pilots.

OTOH, an F-16 would only be able to loiter on target for a half-hour at a time, and the pilot would be spending his time there focussing on the attack, so he has no fucking idea that the terrorist mastermind he's about to attack has a daughter up to hijinks. He'll drop the bomb, write on his paperwork that the building was totaly destroyed, and dance the Dance of Successful Combat Missions.

Textbook case of disproving a minor point. Whether it's the speed of sound in the air around the tube, or the speed of sound from a table doesn't really change the argument.

My point is that this thing's going significantly faster then anything we let your grandma ride in today, that it's also supposed to make turns in the same radius as a car, all three dimensions (20-100 ft elevation means that you're going up and down relative to the roadbed, which itself goes up and down), etc. It's designed to have a lateral g-force of 0.5 Gs, which is comparable to roller-coasters.

This is not an idea that's likely to actually work without a whole lot more development.

Dude, the document isn't the problem. It's its relation to the real world.

In the real world there is not currently a tunnel through that mountain 20-100 ft in the air above the highway. To get through the mountain they either have to bore a completely new tunnel, 20 ft above the existing tunnel; or they have to bring the Hyperloop down to highway level. Neither is trivial.

Eyeballing the cost of the tunnel, $700 million should be enough to bore a new tunnel through the mountain, but that's about it. The https://en.wikipedia.org/wiki/Eisenhower_Tunnel">Eisenhowere Tunnel cost the equivalent of a half-billion$ in today's money. The H-3 in Hawaii cost $80 Million a mile. In 1980s dollars. I included a lot of viaducts and tunneling, and guess which proposal is a 354 mile viaduct interrupted by the occasional tunnel though inconvenient mountains?

And their total cost including developing a totally new mode of transportation, and buying the vehicles, etc. is $16.9 Million a mile for passengers-only, and $20 Million a mile for one wide enough to handle cars.

So I think it'll work eventually. And I think it'll be incredibly cool. But version 1 is not gonna be a 354-mile run in California for peanuts.

The banking thing is interesting, but you totally destroy your credibility with the 3.0 Gs thing.

Hyperloop is not supposed to be a mode of travel grandma can't use. 0.2G is roughly what the competition is doing, and Musk's proposed 0.5 G would actually be comparable to a roller coaster.

This is an idea that'll work great and be near or under cost at scale, but that's because people won't be in it. I am not saying it's a bad idea, or that it's not possible that we'll be zipping our way around on Hyperloops in 2100, but this is a new technology. It's not software. It is pretty much impossible for it to be all three of faster, cheaper, and better without a few more years of aggressive development.

Viaducts never save construction money. The reason is pretty simple: the road-bed you build on the viaduct has to be equivalent of the one you'd build on the dirt, which means the only way to save money is on digging the hole the road-bed would go into. And that's only cheaper if you're in a magical place with easily diggable dirt patches located precisely where you need them.

In the long-term they tend to save money as manageable grades reduce fuel consumption and the need for specialized equipment, but Hyperloop isn't in that phase yet.

See, this sentence is really the reason that I sincerely doubt any of hyperloopies has any clue how much this will cost.

To use the tunnel Hyperloop will have to a) come to the ground (which means that the Hyperloop's vacuum tube is no longer protected from car crashes by being 20 ft above them), and b) convince the government of California to give it a lane through the mountain. Neither is a trivial task. And you're just assuming they'll both happen.

For free.