The assertions that underlie a variety of government behaviors are often quite weak; but what would it even mean for a document to be 'not open to interpretation'? Short of a superhuman AI that is the authoritative interpreter of itself, or a not-necessarily-finite document that manages to address all questions within its remit, without ambiguity or contradiction, neither over nor underdetermined, there is no separation between 'reading' and 'interpretation'.

This doesn't mean that all interpretations are valid, or that some aren't trivially bullshit; but there is no such thing as a 'non-interpreted' conclusion from the constitution. Your 'originalists' (allegedly, their adherence to this is sometimes...questionable) attempt to interpret the document as much like one of the people who wrote it would as they are able to, while other judicial schools do not endorse this as an objective; but 'interpret the constitution while pretending as hard as possible to be Thomas Jefferson' is 'interpretation' just as much as any other technique.

There's still a fairly big gap between the interpretive capabilities of the neural networks we manufacture with unskilled labor and anything the computer scientists and computational linguists have been able to achieve.

For very, very, large datasets, that's not terribly relevant because you have no choice; but for comparatively constrained ones(like Netflix's catalog), this makes throwing meat at the problem rather more attractive...

Generally, when somebody is paying for what it sounds like they could get for free, or even get paid for, there is good reason to suspect that the job description is either underplaying the exact level of difficulty and/or boredom involved, or that somebody has already learned the hard way that what they can get for free isn't exactly what they want.

In this case, I'd be inclined to suspect that the job is closer to being a 'machine vision' substitute for stuff that machines can't yet see or which it wouldn't be cost-effective to have an expensive analyst cobble together a ruleset and then cheap labor check for mistakes when you could just have cheap labor classify it (eg. 'movies set in space' is probably something that you could achieve reasonable accuracy on, if you do some futzing with detecting starfields and common flavors of "rocket thruster jet of flame"; but you'd have your false positives and false negatives from things in space that happen mostly inside spacecraft, and things not in space that happen to involve looking at the sky more than usual, and so on).

It's probably a hell of a grind, actually, given that (unlike, say, being a film critic or some film-studies culture critic type) Netflix is going to want everything ground through and tagged on a variety of parameters, not just the stuff you happen to be a geek about, or the stuff that's worth watching, or what have you. It wouldn't much surprise me if, for efficiency's sake, they have you monitoring more than one stream at a time, or working in faster-than-real time, or a combination. You can probably extract the data they want rather faster than you can enjoy the program, even if it is one you like.

Ah, of course. How could I have failed to consider the 'my betters know better than I do, though what they know and how they know it is a holy mystery, I shall not doubt, nor let any scurrilous disparagement of state entities, especially if true, dent my faith' hypothesis...

Is there any action on the TSA's part that couldn't be 'justified' under this...elastic...standard?

Thankfully, one needn't even go to the trouble of designing your own benign-looking, low power, circuit board. Dell's 'Latitude ON' product was not a wild success; but it is a low-power ARM SoC capable of exhibiting highly plausible 'booting up and doing stuff' behavior, neatly integrated into a deeply prosaic business-traveller laptop.

Can you think of any political development, short of global thermonuclear war, that would make 'the appropriate response to a sizable number of potential threats' go away quickly enough to qualify as a 'short inconvenience'?

What I find curious (honestly, both from the TSA's side, and from the terrorists' side, to the degree that they aren't simply far less common than popularly believed), is how dead-set everyone is on fighting the 'last war' so to speak.

Given the (mostly low-lethality, albeit with occasional exceptions that really sucked for a specific hostage) history of aircraft hijacking, being the first to radically change the game before anybody knew that the game had changed (strictly speaking, the attempt occurred across 3 planes simultaneously; but with limited cross-communication, each was essentially 'first' for the purposes of that aircraft, and the one where that information isolation broke down was the one that was forcibly crash-landed and never made it to target) was a ruthless and clever move. The historical rule had always been 'Hijacking, that sucks; but within a few days, and with the death of very few passengers, the matter will be wrapped up', and so heroics simply didn't make much sense.

Now that everyone knows that that isn't the case, you pretty much have to be confident that you have the manpower to overwhelm an entire aircraft full of people who expect you to kill them even if they do cooperate, as well as national air-defense assets that expect you to kill everyone, and worse, if they don't shoot you down. Aircraft are now largely targets that are only as useful as their direct destruction is.

Given that, it's downright weird that both the TSA, and at least the dumber terrorist types, have remained fixated on airplanes, despite the fact that there are far softer targets, vastly more numerous and harder to secure, all over the place. At this point, hitting a TSA security line, rather than trying to pass through it, or just skipping that entirely and turning a good, honest, domestically available, AR-15 on a little-league crowd somewhere in Iowa would be at least as scary and way easier...

Li-Poly often doesn't (one of the major perks is being able to use 'pouch' style designs with limited packaging overhead or rigid shape constraints); but that just means that Joe Jihad faces the (trivial) challenge of finding a device that still uses classic row-o'-metal-cylinders Li-ion packs.

Oh, don't worry, citizen! If you own a phone that still has a swappable battery, it's probably ancient junk anyway, so throwing it away is just helping you out!

Now... about keeping your AppleCare after the TSA's tech experts remove your iPhone battery...

Unless a given widget has heroic inrush current on start, if you can power the thing up there is no way to tell whether it's the genuine article or the genuine article with a teeny li-poly cell(or even a higher-density lithium primary cell, no need to recharge in paradise after all...) providing ~10 minutes of runtime and leaving most of the battery volume for even more energetic contents...

Apropos of this; (but absurdly NSFW, so cube drones who aren't also the IT guy who controls the local censorship appliance watch with caution) TSA Gangstaz!

Takin' Suckaz Assets.

Short of 'eh, just buy a display model on ebay and pack it full of semtex, the TSA won't notice...' slacker-terrorist stuff, how useful is the 'turn it on' test?

With the relentless demand for miniaturization and battery life, most consumer electronics should be able to get enough power to boot-and-display-innocence out of a battery pack markedly smaller than their real one, even without further clever surgery. In the case of products that have substantial spec variations available in the same chassis (like most 'workstation' laptops) or very similar ones(most cellphone flavors that have a high-end and a cheap-seats variant that share a design language, and often a number of parts), the slightly more adept attacker has even more room, literally, to build a low-drain device and its teeny battery into the chassis designed to run a fairly firebreathing set of components for a couple of hours.

Does the TSA expect that most of their enemies are as dumb as they are, or is this a 'well, it isn't worth much; but it's easy to impose so it's probably worth what you pay...' measure?

Or, y'know, that cost-of-living varies by location and that salaries are a function of the relative strengths of the parties' positions (which are in part influenced by how difficult a job is; but only in part)...

The people 'getting so worked up' are doing so because the company's best interests are directly at odds with their own, and if they had any lingering doubts about how the story ends, they can just ask those once-practically-middle-class manufacturing workers in rustbelt hellholes. Its...totally unshocking... how touchy people get when you threaten their continued economic viability.

If you have to ask, you'd probably find getting linux running on a randomly selected phone (especially if you don't mean 'Android/bionic libc' when you say that) to be a bit of an adventure...

The Broadcom part in the rPi isn't actually all that open, even by the low standards of eccentric arm SoCs(graphics support, especially if you want X rather than Android, isn't a pretty picture on the ARM side; but the 'VideoCore' graphics system is a particular oddball, and dominates the rPi's chip); but it has the advantage of Broadcom in a helpful mood, as well as a large userbase working with exactly the same hardware.

Assorted phones and tablets of the world...less so. At the higher end, people who buy them tend to be disinclined to tear them open looking for GPIO connectors (if there even are any, which isn't necessarily the case), and the low end of the market churns fast enough that knowing what hardware you are getting can be a bit of a challenge.

The use case for swapping the CPU/RAM module yourself looks pretty weak; but it would appear to make sense for the manufacturer: This 'hummingboard' appears to be their existing 'MicroSOM' product attached to a fairly rPi-like breakout board.

If they already make and sell those, they'd likely have to churn out a lot of hummingboards before the savings on connectors makes it worthwhile to integrate the CPU directly with the board.