ecnal

Many countries in Europe, even. Tuition in the UK is actually higher than it is in the US, though the terms of the repayment are quite a bit more forgiving.

https://fred.stlouisfed.org/series/MEPAINUSA672N

The vast majority of economics is not macroeconomic forecasting. That's somewhat akin to saying that computer science is about making email work.

Those are the same thing. You may as well say that there is no shortage of food in Venezuela, just a shortage of cheap food.

The reason companies can't get STEM grads at the prices they're willing to pay is because those STEM grads can simply work for a company willing to pay more. That's not a problem for the grads (you bet your ass I'm not complaining too much, haha), but it is for everyone who wants to hire them.

You heard it here first, everyone: economic indicators are actually just a plot by the manufacturing sector.

https://fred.stlouisfed.org/

Can you take a look and help me pick which of those 421,000 pieces of time-series data are the conspiracy ones so that I know not to look at them any more?

If there were no talent shortage, companies wouldn't have to offer high wages to hire people. You're looking at two sides of the same coin.

It's going to be a spectacular, automated race to the bottom.

Excellent! More efficient shipping benefits a lot of people. The problem isn't the increase in efficiency — are faster CPUs a race to the bottom because worse chip makers go out of business? — it's poor policy that fails to adequately compensate those who lose from this. We should focus on electing officials and supporting policy to turn these Kaldor-Hicks improvements into Pareto ones.

(I know you're not taking a stance against automation, I just think it's important to stress that the correct response is to make everyone better off by using some of the gains to compensate the losers)

http://www.sciencedirect.com/science/article/pii/S1053535700...

http://www.sciencedirect.com/science/article/pii/S0148296301...

While tipping is not an enormous driver of service quality (primarily for the reason you mentioned—that tips are expected even when service is poor), it does appear to be correlated with better service to some degree.

Eh, the F-35 is more an example of how truly infinite time and money can eventually overcome even the worst development processes—at this point the LRIP 10 35As are coming in around the mid-$90m range flyaway with engine, which puts them solidly under the Rafale/Eurofighter price point and on par with the still-in-development Gripen NG. The Danish type selection report is worth a read.

Here's some actual contenders for a modern/Cold War list, though it should be noted that most of these aren't fatally flawed in some fundamental fashion, just designs that didn't turn out as hoped:

- the F-111: much too expensive and only good at long-range strikes - the original Hornet, but not the Super Hornet: not enough fuel and too immature a design for its intended role - the Eurofighter: mind-blowingly expensive, its AESA radar development program is older than the F-35 project and still isn't available, unfortunately timed (becoming a mature platform as VLO aircraft begin to enter service in many air forces) - the F-16/LWF as it was originally conceived: range-only radar, no real payload, more or less a target drone - the MiG-29, a point defense fighter that never really could be molded into a true multirole

Correct: it is fantastic at being vulnerable to air defenses, leaving it unable to engage the Iraqi Republican Guard in 1993.

http://www.afhso.af.mil/shared/media/document/AFD-100927-061...

I am seriously struggling to understand the rationale for that.

From personal experience, CP Optimizer beats the crap out of other constraint solvers. Its API is... well, a bit rough, but it's incredibly fast and the support is top notch.

I can't really speak for other IBM products, but I suspect they generally align with your perception here.

Who needs brave pilots in fighter seats when you can get the same results with slightly less brave drone operators?

Primarily because we can't get the same results with slightly less brave drone operators. High-performance UAS are significantly more expensive and still far less capable than their manned counterparts in contested airspace; their primary advantage is in loiter time because they don't require a pilot and can be optimized for efficient flight in low-airspeed regimes (at the cost of their medium and high airspeed performance). "Like a manned aircraft, but with more latency (or autonomous controls and planning techniques we don't yet know how to implement well against an opponent actively attempting to make your life difficult), a bigger price tag, and a kill chain enormously vulnerable to electronic warfare" is not a good selling point.

Your arsenal ship idea isn't a bad one (except for the bit where it renders all civilian shipping a valid target in times of war and is also a war crime, though without knowing that I can see why it would be appealing :), but those ships already exist. Those shipping containers are called VLS cells, and the autonomous expendable relatively inexpensive point-and-click drones that accept sensor and targeting input from external platforms are the missiles that go in them. E-2D and F-35 CEC already exists and has been demonstrated with midflight SM-6 guidance handover to strike OTH targets; man-out-of-the-loop sensor fusion has been a part of the USN's ABM program for a while now, and they've been launching SM-3s with cueing from external platforms for years and years. The USN quad-packs ESSMs, too, though ESSM is purpose-built point defense against super and hypersonic maneuvering targets, not an AAW tool.

Drone operators can control entire formations of combat drones (no need to actually fly these things — point and click interface should be enough with modern tech). Autonomous refueling, sensor fusion, battlespace mapping — the only need for humans in the loop is giving "go / no go" for kill requests.

This largely ties into the USN's view of the future (cf. various whitepapers), except that the USN views it as a sensor/shooter divide where the drones are extremely LO sensor platforms datalinking target information to arsenal platforms capable of engaging at extreme range with huge numbers of weapons.

numerous uniform modular cheaper vessels instead of a few big expensive carriers surrounded by combat support

Small ships are awful. No, really: they're horribly inefficient, because as soon as you run up against their tiny volume, weight, or power generation limits, you need to build another copy of them with all of the same supporting crew and systems to get more capability. That means a ton of inefficient small turbines generating power for duplicated systems, significantly more crew (who are actually far and away the most expensive part of operating a ship across its lifetime), and significantly less capable individual combatants. Compare the Sachsen-class (5800t displacement) with 32 VLS cells (each cell representing one missile to use for offense and defense) to the Arleigh Burke-class (up to 9800t displacement), which has 96 VLS cells, each storing one missile for offensive purposes (SM-6 in anti-air or anti-surface mode) or four for defensive purposes (quad-packed ESSM).

The three Sachsen-class ships you need to purchase to have the same offensive and defensive capabilities of the Burke (minus ABM capability, which the Sachsen lacks) cost $3.18 billion to acquire, vs. the $1.8 billion for a Burke. You're also paying, training, and feeding 729 crew members for the Sachsens, but only 329 in a Burke. Additionally, groups of smaller ships are much more vulnerable to attrition in saturation attacks unless they're sitting right on top of one another, as you can defeat any given one of them by engaging it with slightly more missiles than it's capable of taking out (i.e. 1 to 3 + 32 anti-ship weapons, assuming your Sachsen has a perfect defense), whereas to sink one Burke you need to strike it with 1 to 3 + up to a hypothetical maximum of 384 missiles, if you've quad-packed every VLS cell and have a perfect defense.

Smaller ships are just not worth it. If you're willing to spend the money it takes to have a world-class navy, you're better off taking the far more efficient route by building arsenal ships with hundreds of extremely long range weapons and networking them with sensor platforms.

Every ship can (and should) be fitted with ECC and offensive electronic warfare options, which prevents attacking electronic support first

This is already the case.

Try to jam 200 small ships simultaneously while they are constantly on the move in the large area — nearly impossible

That's simply not really how electronic warfare works. 200 small ships facing a one-axis attack are no less vulnerable to noise jamming than one ship. If you want to force your enemy to expend a lot of emitters on various more advanced techniques, the most efficient way to do that is by simply having more of your own emitters to jam, and the most efficient way to do THAT is not to build a bunch of ships with them, but by putting them on aircraft or in large quantities on larger ships.

No designated C&C ships — all battlegroup members can take multiple roles as needed, including command and control.

This is desired irl, but the training overhead is difficult (not everyone is good at C3) and it tends to rely on the assumption that your opponent is not capable of messing with your datalinks (which is not a safe assumption to make).

Alternately, remove the need for highly skilled workers to get visas entirely, thereby completely preventing companies from abusing the system.

Because, for all their software and development delays, they significantly outperform 4.5th gen offerings and (the A, at least, which is the majority of what people are buying) don't cost nearly as much (~85-95 million for an F-35A flyaway vs. 110-120+ million for a Eurofighter or Rafale).

There is no reason to keep the A-10 around if its only role is a Maverick/Paveway truck chilling at altitude and plinking targets with Sniper/Litening. There are plenty of aircraft that can do that /and keep doing it when there are angry little men on the ground who don't want them to/, something the A-10 can't do.

The A10 is basically a flying tank meant to be hit and survive.

Like Dale Storr's OA-10, which took a single Strela hit and went down? The A-10 is certainly more likely to be mission killed by MANPADS and AAA than other aircraft (which are liable to be destroyed outright), but those that survive are not useful for combat--Kim Campbell's A-10 was grazed by a Shilka and rendered combat ineffective; after she landed it in manual reversion, it was shipped back to the US and required months of intensive repair. That's great if your objective function is maximizing the number of airframes that are eventually flyable again after being damaged, but godawful if your goal is to actually use them to perform CAS.

The A10 is simply put the most effective CAS tool the US Military has or ever had.

I completely agree with this, but tools only benefit you /if you can use them/. A-10s cannot survive in contested airspace, and if the not-at-all-fearsome Iraqi Army spanking them so hard that they ended up with a 10,000 foot hard deck didn't clue politicians in on that, maybe a few videos of modern SHORAD like the Pantsir which are available for $peanuts will.

just because they might be against an enemy with better anti-air doesn't negate the need for boots on the ground, which require CAS for maximum survivability

The CAS you get is better than the CAS you don't. If your primary CAS tool is the A-10, then being against an enemy with even the slightest anti-air capability means that you don't get any CAS at all.

Most likely the Stealth B2s or RQ-170 Sentinel would be sent in first to quickly take out as much of the AAA as possible

The RQ-170 is a phenomenally expensive unarmed recon drone that the US operates a handful of. The B-2 is useful for striking an enormous number of fixed targets, but is not designed or intended to be employed in a SEAD/DEAD role against anything more mobile than theater-level air defenses. No SEAD weapon in the world is capable of striking air defenses that it isn't fired at, and as a result anything flying at low level is vulnerable to SHORAD whose operators use the advanced tactic of parking it in a treeline and waiting until they hear a jet to engage it.

Well, it's fair to take issue with the idea that ZH provides any information whatsoever, but I get the feeling that that's not what the parent commenter is upset about.

Multiroles like the F-18, F-16 (excluding the original worthless LWF proposal), Su-30, Su-35, (today's, though not the original) MiG-29 and Su-27 variants, Typhoon, Tornado, Mirage, Rafale, and nearly every fixed-wing fighter employed by every air force today definitely do not have bad histories--some less stellar than others, but none bad.

Multirole capability is primarily a function of the weapons an aircraft can employ. This wasn't true in the 70s when targeting pods were binoculars and effective employment of aircraft in strike roles required decent performance in low-and-slow flight regimes, but we've been living with pods and PGMs coming out of our ears for 25 years now. Design tradeoffs for multiroles today are primarily those of cost and maintainability. You can pay out the ears and get larger twin-engine multiroles that strain logistics and wallets, opt for a single-engine F-16alike multirole that'll do everything nearly as well except for flying really high and really fast, or inexplicably go Eurocanard to pay F-22 prices for sub F-35 capabilities. Anyway, the point is that it's next to impossible not to make a multirole these days even if you're trying; if you can employ weapons with your jet, you're multirole.

Progressive consumption taxes based around this enjoy a great deal of support among economists and very little support among politicians, unfortunately.

Those articles (and the data behind them) suggest the opposite of your point: the middle class is shrinking, yes, but it's shrinking because incomes are increasing and people who were previously in the middle class are moving into the upper class.