I'm not sold.

Why is wagering cash directly 'gambling; but exchanging cash for nonrefundable WagerBucks that are only used for wagering just prior to wagering not gambling? It's true that being nonrefundable makes WagerBucks no longer a strict representation of cash; but that's sort of like saying that things you buy with gift cards are part of the barter economy because gift cards can't be cashed out.

I'm also unclear on point 2: people can certainly gamble on games that are directly adversarial and have rules with win/loss/tie conditions; but there's an entire category where all outcomes are wins of varying sizes(like the 'ticket' machines where you always get at least a tiny trickle of output per play; but the actual prizes are hundreds or thousands of tickets; a loss state is clearly implied by the valuation but denied by the rules; most lootboxes are this way since they are never genuinely empty.

Point 3 also seems curious: is casino poker not-gambling because the owner/operator only takes the standard house rake each hand, regardless of which players are winning or losing? Is it only ever gambling if the operator deals in money, and as long as there's a level of indirection it's never gambling? What level of separation counts as separation if the operator pays out in something that only 3rd parties buy; but keeps an eye on the secondary market and adjusts the game accordingly(with items known to have higher secondary market values displayed in the same way that casinos display high cash value outcomes) or the like?

Ultimately definitions can be,(and always are) matters of what we want them to be; but none of what you describe seems to suggest that it is 'not hard' to have 'very easy and clear boundaries and definitions'. It's not impossible to have boundaries and definitions of some sort; but the whole affair looks ripe for ambiguity and boundary pushing; especially when the incentives overwhelmingly favor someone who can get gambling effects without suffering gambling regulations.

It also seems like a mistake to tag it as a 'content consumption device' when anyone buying one of these is specifically choosing it vs. the Ipad Air of the same price that is actually more powerful across the board; but without a keyboard. It's not a high end video editing and finite element analysis workstation; but, when it is going up directly against ipads more or less because it has keyboard and a conventional pointing device it looks like the opposite of a 'content consumption device'(at least in the hopes of people handing them to students; not necessarily in the youtube being watched).

Absolutely, N150s are very much the cheap seats. My point with them was more that PC OEMs mostly don't seem to do build quality first/performance second designs, at least not as a category of cheap laptop(that's much more a thing you find with the 'gamer'/'glossy consumer' vs. 'corporate IT' divide above $1000).

Intel and AMD are in the somewhat embarrassing position of a 4w phone SoC actually punching pretty hard compared to their cheapie offerings(though in both cases you'll have more options on RAM, at least at time of purchase); but PC OEMs seem to mostly offer specs and build quality roughly correlated with one another (with the exception of the low end of 'gamer'; which will have a comparatively powerful discrete GPU and suck in multiple ways; and the grimmest config of a given 'enterprise' line which will invariably have a dim TN 1920x1080 that looks like it was stolen from 5 years ago; but will be as well built as the most expensive config of that line). There mostly isn't a "Give me an N150; but with a chassis and screen that looks like it cost $1000" option.

This 'neo' thing looks to be of interest, vs. the old low end M1 config they were keeping around at Walmart; in that it's basically Apple making some fairly sharp spec sacrifices(albeit with the advantage of their CPU and GPU designs apparently hitting pretty hard) while delivering a decent chunk of the build quality.

Calling out the 8GB of RAM is fair; but it seems hard to ignore how much of this devices is really a shot across the bow of the part that Asus and friends do control: the general build quality, screen, etc.

Asus can't do much about the fact that Apple can just throw a phone CPU into a computer and call it good enough with a reasonably straight face; but it is on them that the PC bargain bin has a lot of plasticy trash; and much of it isn't even the "clunky; but at least trivially repairable" trash like it once was. You do see the occasional combination of a real cheapie CPU (normally an N150 or one of the quasi-embedded Intels that they don't really want anyone to think about) with some degree of build quality from the slightly weirder chinese outfits like chuwi(when they aren't fudging the model numbers in firmware); but low-end PCs, to the degree they aren't just plain bad, seem to systematically overpay for CPU and neglect other elements.

There isn't a 'basic definition of gambling'; there are some slightly vague ones you get in dictionaries and people's opinions and specific but often easily wordsmithed ones you get in various statutes.

For the purposes of a lawsuit it's obviously the statutory definition that ends up mattering; but those typically offend common sense in one way or another, at least after the more cynical types build businesses carefully structured to demonstrate the holes(hey! did you know that you are a degenerate bookie if you both set up bets about outcomes in sports and act as a market-maker for anyone willing to take them; but a totally legitimate(tm) financial institution if you stick to either running the market or acting as market-maker; but not both at the same time!).

I suspect that Valve set up their lootbox system as it is(very much without any first-party redemption for value of anything you get inside the boxes) on the advice of counsel; so there's probably a carefully written legal opinion to the effect that it is on the right side of the law; but it pretty clearly looks like gambling; and when Valve runs one of the major shops that handle resale for the items supposedly not of value the idea that the prizes aren't prizes is...very specific...at best.

Especially when 'tokens' are a consumable and sessions are typically auditable.

There likely have been cases, back when PCs and workstations were expensive and endpoint auditing and management were typically fairly light, when computers were used to sweeten the deal at least for certain sorts of lower level nerds; but because it was fairly easy to just turn a blind eye to recreational use off hours and relatively difficult to have random endpoint hardware doing something productive 24/7.

With 'cloud' resources expensive enough that someone would actually bring them up as part of a compensation package, though, that's just not the case. Those are metered(or 'unlimited' but subject to various throttling strategies that will interfere with using them for work if you hit them too hard); at which point there's not really any 'extra' value you can provide by paying in kind rather than in cash; unless your company has an exceptionally stupid bulk contract with one of the 'AI' outfits that beats the normal API price or something.

It sounds like the actual problem here is that Apple isn't being sufficiently constrained; but of course the adtech scum are looking for something else entirely.

I'd be very, very, careful. More or less the whole point of 'attestation' schemes is that they allow, as one would expect, attestation of what a device is running; chained to some cryptographic secret that you'll need to conduct a hands-on hardware attack to extract if it is implemented competently.

I'm not denying that this proposal is technically open source; it's just attestation, like tivoization, is one of those places where essentially none of the control is based on obfuscated proprietary blobs, so 'open' gets you fuck all unless you own the platform keys.

Given the mostly terrible(both in security vs. external threats and treatment of the user) state of OEM android builds; it seems like handset OEM controlled attestation schemes are an awful idea; whether they are the 'play integrity' ones that are google controlled with licensors of the play APIs getting their platform keys blessed or 'open' with this consortium's members getting their platform keys blessed.

The ability of the bootloader to ensure that you are booting what you think you are booting is one thing, and actually useful if you want to be careful; but 'play integrity' is not bad because of google; but because the theory that you can obtain security by verifying that a phone is as the OEM wanted it is simultaneously hilarious bullshit and wholly user hostile.

The company that spent a hilariously long time fighting with users over the 'bixby button' has now decided that customization will be a thing. I'm sure I can't even imagine how they are going to screw it up; but it should be exciting.

They are the first step toward a slippery slope toward a ban on anonymity.

I can see why 'AI' CEOs are fantasizing about the ultimate greater fool swooping in to save them; but the reasons they are hyped about the idea seem like basically the same ones for not nationalizing them.

Normally fed R&D programs are aimed at projects that don't have particularly good economic incentives; and state run organizations tend to be clustered in areas where the economics check out but the market incentives are troublesome(like infrastructure and public health) or where there are serious structural issues with letting the private sector handle it(like military and police). Even there it's common, though not universal, for a lot of the actual work to be farmed out to private sector suppliers with the directly state run part handling the project management and doling out requirements to bidders rather than running a vertically integrated operation.

In the case of 'AI', despite it being an apparently terrible idea based on the money nobody is making, private capital has been running toward even the most blatantly unprofitable capex hells like it's the most exciting thing in human history. Apparently not a manhattan project or apollo program instance where the only reason anyone is going to bother is because the feds are paying; so the need to put massive capital behind something with not market case apparently doesn't apply.

And if the issue is not getting exactly the product they want or labor market disruption; it's again not clear why nationalization would be in the cards. It's not like the broligarchy is all that unwilling to collaborate with the state anyway(libertarianism is for capital gains taxes, not mass surveillance programs...) and their principles are even more flexible when they are losing money and desperate for customers. If the concern is massive labor market disruption it's also not clear why that would lead to nationalization: it's vastly easier and softer touch to just tax them enough to run programs to keep the people rendered obsolete from starving or rioting than it is to rebuild the economy around The First People's Patriotic AI Factory and hope that goes well.

That was why I specified "the wars we've seen so far" and specifically called out yet worse future aspirations as the likely point of the concern.

I have absolutely zero reason to suspect that the goals are anything less than "worse than I can imagine"; especially when such a tantrum is being thrown about the importance of unfettered access to features that the DoD doesn't currently use; I'm just not seeing anything in the wars currently available for inspection that suggests the sort of significant break that would either have openai tech flunkies concerned or the DoD crying about features they simultaneously claim not to be using and claim to have an urgent and existential need for. I have every reason to assume that the entire plan is to make things worse; and that DoD leadership sees the fact that the existing wars are relatively conventional as a problem that needs to be fixed in a way that people who aren't them will be horrified by; I just haven't seen that in the ones currently being undertaken.

In a great many cases the very concept of 'visonary' is something of a comforting lie; above and beyond leadership approaches.

It can be a dangerous lie; given that it justifies systematically rewarding the photogenic 'idea guy' while doing everything you can(and somethings that will catch up with you, not necessarily even in the long term) to treat the unsexy people who just do stuff as expendable cost centers who should be outsourced or exterminated whenever possible; but it is also a comforting one if you don't like the idea that, in most places at most times, there isn't going to be some flash of insight that is going to change everything and you are just going to have to sit down and do some actual incremental work on something locally useful.

It's not as though everything is above a rethink just because someone is busily doing it; inertia is a thing and projects meandering down blind alleys does happen; but even when you are doing something 'visionary' it's going to end up being a lot of unsexy execution; and even when you cull some myopic legacy process or doomed project it's normally so that you can tell people who did just have their heads down executing on that to go execute on something else instead.

Aside from the cases where that claim is just wrong(he does tend to use words that sound simple and concrete; but often aren't, like the rather protean concept of 'woke', which apparently will be kept out of science and government AI, whatever it actually means, or 'won', which we have apparently done despite the war being ongoing with no announced objectives and at least three distinct and in parts contradictory justifications); there's a fairly important distinction to draw between people who speak in generalities to avoid being pinned down on knowing or doing anything; and people who are capable of speaking in abstractions in order to allow generalization.

Watching the president spitballing what sort of acceleration mechanism an aircraft carrier is going to use to shove planes around based on his vague feelings about magnets and excavators isn't a sign of good, common-sense, leadership; it's the equivalent of a CEO who has never heard of the OSI seven layer model trying to micromanage fiber out of a random network closet because by god RS-485 was the shit back in his day and he has no idea that different PHYs have different applications and more confidence in his own ability to do it in his head than he does in the ability of an engineering team to be assigned the task of designing an elevator suitable for naval work. (There's also a certain irony in the fact that 'magnets' are apparently just the worst thing in the ocean until it's time for the Trump-class battleship to get a railgun; at which point that has been completely forgotten because reasons; and we receive no lecture at all about how all guns since forever have been operated by chemical propellants.)

It's usually a good idea to be very leery of people who refuse to be concrete about anything; and for more specific implementations you want to actively avoid the people who are either taking refuge in generalities because they are clueless or emulating the speech of those further up the ladder they wish to climb: you need someone to specifically do a particular thing themselves or delegate a very precise set of requirements; but don't fall for the impression that someone who genuinely seems unwilling or incapable of thinking at any level of abstraction is either better or even suitable for many purposes: that's what a congenital incapacity for anything but tacking between wildly disengaged emotion driven demands for extremely vague good results and extreme micromanagement and bikeshedding of whatever details they think they actually do understand looks and acts like. As we've basically seen: essentially nothing between a wildly over specific procurement-by-executive-order decision on how to raise aircraft to the flight deck based on what's common and cost effective in earthmoving equipment and exceptionally vague outbursts regarding wanting to 'win' or 'make a good deal' in inchoate terms that admit only emotional metrics.

In some respects, it's the idea that all abstraction must be bullshit that is more dangerous that any of the specific bullshit produced by peddlers of vague platitudes. The ones who are good at that tend to end up overpaid and overpromoted; but those who think they are immunizing themselves against bullshit by surrendering all of abstraction are massively crippling their ability to carve up problems in ways that allow you to fight against bullshit by following the process of decomposing a large objective into specific pieces in a systematic way.

I'm...curious...about the justification for treating authentic corporate buzzwords as a meaningfully distinct sample from the synthetic ones as part of the study design.

The hypothesis that corporate buzzwords generated by actual 'leaders' are better than the synthetic ones that are merely syntactically correct seems plausible enough to serve as a basis for further inquiry; but far, far, from being the sort of thing you can treat as a given for the purpose of testing something else.