It also seems like a sensible move on DeepComputing's part. Framework parts aren't free; but compared to what it would cost to bang out some low-volume laptop bits to suit your dev board they are pretty attractively priced and widely accessible for minimal additional effort.

Not really a replacement for the sort of dev board that breaks out a zillion headers and DIP switches and things; but compared to just your basic rectangular bare board it presumably cost a pretty modest amount to make the board in a shape that allows people to slap it into a laptop format quickly and easily.

You are ignoring the tying; which is absolutely corrosive to the "FREE MARKET":

Apple creates a phone that people like. Ok, so far so good. Innovation and polish rewarded, free market in action, hooray.

Apple sets up a cloud storage offering that is deeply undistinguished and not particularly well priced. Ok, nobody requires you to make actually-good products, market will sort it out, whatever.

Apple gives their own, and only their own, there is no standard a 3rd party can meet and no configuration a device owner can do to change this, cloud service uniquely privileged access on their phones in order to make it the only option for iOS cloud backups. Now we have a problem: Apple gets to use their actually well liked product to shield their frankly mediocre one from competitive pressures. Any you are still banging on about how this is an exercise of the glorious free market?

It's honestly a couple of notches more audacious than the MS bundling of teams to get rid of slack and zoom despite teams' abject mediocrity(which also attracted regulatory scrutiny): sure they made it 'free' with Office licensing to bury the competition; but they didn't tweak Windows to break Zoom's webcam access so that Teams could be the better video conferencing product; while Apple has essentially made a number of iOS design choices aimed at making iCloud the only viable cloud storage option.

Are you just unaware of how tying works; or are you working with a definition of 'product' that expands to include every last 'ecosystem' thing a vendor chooses to argue is an integral part of a 'product' for market purposes?

It seems pretty much as-expected that subsurface would be where the soil biology remains when you sterilize everything on the surface; but it's somewhat more surprising that there are apparently such limited alternative methods.

Sure, if you want results within hours small burrowers seem like an unbeatable candidate; but it's much more surprising that even several decades later there would still be a layer of basically free real estate that apparently no mycelial networks, miscellaneous free-living microbes, etc. are willing or able to colonize at a slower but steady pace.

Aside from being novelty-addled herd animals; I think that there's a much stronger cultural affinity for the technology that is all about the fact that you can sometimes get surprisingly plausible outputs from nescience so profound that it would be anthropomorphizing to call it ignorance; than for the technology founded on the hope that if you systematically plug away at knowing enough you might eventually be rewarded by competent outputs.

This should be a real comfort to all users of java applications that have run out of bugs exploitable by adversaries without sophisticated physics expertise and nation state resources available.

Anyone heard of such a java application?

If you have the 'prompt' for the video and either lack the interest or the ability to actually add anything; why would you destroy something that at least has the virtue of being succinct by blowing it into a meandering video?

Text-to-speech, at least, has use cases for limited screens, drivers, the visually impaired, and the illiterate; but dragging stock video slurry into it?

Why do all those examples of something "truly revolutionary" sound so crushingly banal? I'm not disputing the new means potentially being an improvement over the older methods(in particular machine translation seems notably less bad that it used to be); but when the examples are 'searches you might have done two decades or more ago' and 'check email' it's a little hard to detect the revolution.

It can definitely sometimes still be an issue with devices that either have some odd quirks or where the vendor is really serious about it not being a supported configuration unless it's Thunderbolt(tm)(r) rather than that other kind of PCIe tunneling using a USB-C AUX mode; but that's definitely a niche case. Especially when you get outside of devices that are basically sold to mac users but you can try plugging one into a PC if you must.

It also doesn't hurt that the general quality of USB-C docks and dock-adjacent peripherals(eg. monitors with DP MST daisy chaining and laptop power) seems to have improved at least a bit. Through some combination of gen 1 never being as polished as it needs to be and business laptop vendors all being Intel focused it used to be that actually reliable docking normally meant either their proprietary-but-reasonably-solid or their thunderbolt dock; more recently I've seen a lot less trouble from DP alt-mode based docking options; and a lot more cases where those are the 'default' option for even Intel laptops; while explicitly thunderbolt is positioned as more of a mobile workstation thing.

I'm not sure if it genuinely just costs that much more to implement, or if Intel was too greedy and kept too tight a grip on thunderbolt peripheral controllers for too long; but, despite mostly not screwing around with weird market segmentation games around thunderbolt as a platform feature(you have to really be slumming it to find intel laptops that don't have it) it seems to have remained really niche.

The situation was less of a blowout on the laptop side; but perhaps more of a dramatic shift in that AMD had been relegated to the trash tier on the laptop side for ages; used to be that of course if you wanted something that wasn't the 2-inch thick consumer tier monster from best buy that meant something intel; then hey, look at that, T-series thinkpads come in AMD now; then when the T490 and 495 were replaced they were just the "T14" with intel and AMD variants. Intel still had thunderbolt, if you cared; and AMT is significantly more mature than AMD's DASH implementation if you use that; and Intel clearly had more experience with being able to supply the CPU, wired and wireless NICs and bluetooth; while AMD was doing some futzing in the earlier ryzen parts to slap the rest of the expected components together; but AMD going from 'haha LOL' to "at least comparable; in some cases superior; and OEMs are willing to actually use them" was a reversal of what had basically been an unbroken period of victory since the Banias and Dothan pentium Ms back in 2003/2004.

The comparatively lower demands for raw CPU punch; and high rates of physical and battery attrition(plus work-from-home provisioning) did mean that(at least among PC buyers) the fairly tepid advances didn't really put intel into any positions that a discount couldn't get them out of: none of the 13900k and 14900ks getting curb stomped even when running at more than twice the wattage of one of AMD's high-cache parts; or hardcore Intel pals like Dell basically having no choice but to ship Epyc for datacenter customers: but it's quite a tumble to go from close to two decades of AMD being almost irrelevant in laptops to them being somewhere between comparable and better unless you specifically need thunderbolt or AMT.

It wouldn't be entirely surprising if the coffee got cancelled as part of the termination of a broader catering arrangement that was made back when spirits and margins where considerably higher; that's the sort of thing where the vendor rep is just going to keep smiling warmly and upselling you as long as you feel like a high roller; but unless there were outright running boutique coffee shops with dedicated staff and coffee with under-ten-minutes TTL guarantees it would be vastly more surprising if the $100 million is actually the coffee/tea item broken out rather than an entire back-when-xeons-were-mandatory-and-high-margin deal going away.

Quite possibly some fat dumb and happy agreements that did deserve to get revisited; but seems really stupid to have dragged it out publicly as a cancellation then a reinstatement rather than just a change of suppliers that nobody internally is going to expect to be an upgrade given the circumstances. There's no way we'd be talking about this if it were just a "man, the free coffee isn't as good as it was back when we made money" rather than "Intel cuts coffee/Intel repents of stupid idea"; which is a headline.

Coffee seems like a really weird one to have ever cut. Unless you are just getting screwed by the vendor or fairly badly mis-sizing for your requirements unexciting-but-not-fucking-terrible coffee is pretty cheap; tends to be modestly performance enhancing(or at least restorative to baseline, for the addicts out there); and benefits a great deal from being fresh(or at least not obviously warmed-over) which makes it more likely that having it in-house will discourage people from popping out to grab some during the day.

Especially for a company whose employees presumably skew mostly on the expensive-ish side that just seems like a false economy to mess with too much. Doesn't mean that you need to hit "artisanal beans prepared to your exact preferences by cool onsite barista" tier, or should even try(that's actually the other virtue of making it free: there's not going to be any one 'right' quality level: if the coffee is lousy charging people for it is going to give the sense that you are trying to pull one over on them, even if you are genuinely doing zero-margin cost recovery. If the coffee is amazing even genuine cost-recovery will make the junior people feel like the perk is aimed at the high earners and they are an afterthought, while if you subsidize it it's strictly speaking a good deal but then having to pull your wallet out because Facilities is only willing to cover the first $5 of a $6 cup of coffee feels like some sort of weird attempt to make you think about costs because you aren't a normal human who just drinks coffee until satisfied but must have your appetites reigned in by price signals: if the sticker price is free people will tolerate poor to fair and appreciate good to excellent; and (usually) recognize that if their tastes are highly atypical they'll just have to sort that out with a different vendor).

It's also going to make things that people already don't like, like loading screens, feel extra abrasive. Both because you are paying for them and because now doing something like just pausing the game while you go off to do something else for a little while will still have the clock ticking.

It's pretty typical to pause a game and leave it if you are going off to do something else for a little while; rather than fully closing it down and reloading a save(if it even allows saves anywhere) because you need to do something else for 15 minutes. That sort of behavior starts to get a little decadent if you are just letting it burn all night because you can't be bothered; but for short breaks it's convenient and quite common.

I'd assume that, with some combination of effort on their part and developer buy-in, Nvidia would probably be reasonably well placed to support markedly faster save state/restore state operations for specific games to make that matter less; but getting that to work in the general case is likely pretty tricky; so even if supported it would only be on some titles.

I think what's frustrating in Mozilla's case is the fact that what they are advocating for overlaps with what they are supposed to be interested in implementing; and there are some questions about how effectively they've been implementing.

There's nothing intrinsically privacy-preserving about browsers that aren't chrome skins; but technical standard-setting and market adoption both tend to require actually shipping an alternative to what you are against, rather than just dragging out a rearguard action against an opponent with unlimited opportunity to change the defaults again tomorrow.

If Mozilla's positions were bad then their software work would be somewhere between irrelevant and negative value; we've got Microsoft if we want chrome working against us on someone else's behalf; but the good they are able to do is substantially tied to their ability to point to a browser that does what they advocate browsers should do; rather than just hoping to somehow control the direction of a browser that isn't technically proprietary but, as a project, is substantially just a Google operation whose repo doesn't strictly require internal corporate credentials to interact with.

Aside from architecture holy wars; this is the main reason why server CPUs are still offered in very low core counts(along with some specialty parts that are all about frequency and single thread performance; or low-ish core count parts that are aligned with specific common licensing schemes):

Your big database or mostly-cached web server, say, is still going to need a whole bunch of RAM and some high speed networking, possibly storage, so you are basically buying a big fat memory controller and lots of PCIe lanes; at which point the percentage of the silicon that is 8-16 86 cores vs. something else starts to look a lot less interesting compared to that 12 channel DDR5 controller and 128 lanes of PCIe; which would need to be there regardless of architecture.

Obviously the existence of actually-viable server ARM puts some limits on Intel and AMD's ability to play pricing games(at least against large customers); but, if anything, it's the computationally lightweight stuff where I/O is going to be a huge chunk of the package(literally and figuratively).

There's also the fact that a nontrivial amount of what is being provided has already been spent: there has been some new production, mostly of ammunition and smaller systems; and some draw-down of stockpiled hardware that's faster than purely letting it age out and destroying it at end of life, which implies some production in the future; but (both because it is what is available and because the process for approving transfer of existing DoD property is easier than the one for doing appropriations for new stuff, bidding the job out, having it manufactured, and then transferring it) the bulk of the military equipment provided is stuff that was already HIMARs or howitzers or Bradleys or whatever. There's also the specific case of the various cluster munitions, which would otherwise have been destroyed because the US hasn't wanted the PR hassle of using cluster munitions since Desert Storm.

As noted, just because it's in stock doesn't mean its free(especially if it's something you expect to need to replace; rather than something that was close to aging out or no longer approved for use and you were expecting to pay someone to scrap it in the relatively near future); but it's nonsense to just take the unit acquisition cost of a 10 year old M142 and pretending that that's a cash transfer that could have been used to build science machines or fix potholes.