Why diamond rather than gold? The value density is a little bit lower(assuming a decent quality stone); but metals tend to have a much, much, smaller gap between purchase price and sale price at any given time; can be divided if necessary without reducing their value, and are similarly imperishable, fairly hard to trace, and likely to have a buyer available almost anywhere.
As an 'investment', gold has a pretty tepid history; but at least you can actually buy and sell at reasonably close to the nominal market price, rather than eating a large automatic loss the moment you buy.
I suspect that it depends on how difficult the product of this process is to distinguish from naturally occurring diamond(assuming it can produce reasonably large crystals of suitable quality at competitive cost at all, not just industrial-grade abrasives or thin-films). There are already a variety of imitation materials available, some pretty terrible, some really quite adequate for most purposes. If these synthesized crystals can be reasonably easily distinguished(some sort of nondestructive test with reasonably low cost instruments by someone of moderate skill) then it will be easier to keep them out of the 'real' diamond supply chain; and to treat them as just one of the fakes for marketing purposes. If there is no such test available, then there will be some serious unhappiness.
Because diamonds have some extremely useful properties(superb abrasion resistance, extremely high thermal conductivity, good chemical stability, some potential as a semiconductor, etc.) trying to fight a process that can synthesize them in the necessary quantities and sizes would ultimately involve an industry that sells shiny trinkets going up against a whole bunch of larger and markedly more important industries that would have use for one or more of these properties. As long as the 'synthetics' can be distinguished from the 'real' diamonds; and stigmatized accordingly, there will be an incentive to stop trying to win the battle against all the potential industrial users; and focus instead on hyping the virtues of 'natural' diamonds, for everyone whose love isn't cold, sterile, and engineered in some laboratory.
If the only value of diamonds was being shiny, it'd be a lot easier to keep the lid on; but their properties are interesting enough that jewelry would be a footnote if bulk synthesis were available.
There's also the 'price' consideration: Unless you get really lucky and score somebody's cherished childhood collection at a yard sale from their parents who don't know that the lego secondary market is a well-oiled machine; enough lego(especially Technic/Mindstorms) to build something complex enough to be interesting to computer control gets pretty pricey. Cheaper than grabbing the McMaster-Carr catalog and ordering the moving parts from them; but still pretty expensive.
Perhaps there are people who actively prefer virtual building blocks, I don't know; but it is undoubtedly the case that you can't get much lego for whatever Minecraft costs, and a lego habit can get punishingly expensive.
I don't know how they do/did it; but Minecraft somehow made the jump from being a moderately novel; but not obviously interesting, "8-bit-styled" 3d multiplayer sandbox game(with some optional combat elements), sort of the 'Duplo' to Dwarf Fortress' 'Technic'; to having an enormous amount of mod support, which has mostly eclipsed such original 'game' as there actually was.(And not just 'mod support' in the 'Bethesda builds a world; and then the players fix it' sense; the 3rd party components include things like the popular servers, the popular mod interfaces, and similar lower level than usual stuff).
I'm not sure how it came to that; but that vast availability of 3rd party features seems to have become the major draw.
I realize that this probably makes me a horrible person, whose shriveled soul is incapable of joy, and who was obviously never a child; but exactly how much external infrastructure are you allowed to use before "In Minecraft!" is no longer true in any meaningful sense?
The various redstone logic arrangements, while obviously perverse and inefficient, do implement various computational widgets "in minecraft". As best I can tell from reading the project documentation, this exercise uses an external helper program to do all the heavy lifting(digesting web pages and incoming video into an array of textures to be applied to blocks, implementing the MMS for the 'selfie' feature), with the only involvement from Minecraft being a large rectangular array of blocks that the external helper program re-textures to produce something similar to a framebuffer(it's not quite the same, since each block displays a multi-pixel texture, rather than acting as a single pixel).
You certainly couldn't do this with just any game(at least not as easily; if you have access to the game's memory, you can probably scribble on its textures; but doing that without bringing the whole mess down in a screaming heap is easier said than done); but that arguably makes it less interesting: Minecraft's support for external modification of server state is relatively robust, so there really isn't much that you can't do, if you are willing to do most of the work in an external program and then use Minecraft as a needlessly perverse frontend. Am I just a joyless asshole? A generation too old for this 'minecraft' stuff? Insufficiently impressed by a Verizion marketing exercise?
Microsoft was knee-high to a grasshopper and might not have even yet been in the business of selling word processors to Mac users and a cheapo 'disk operating system' for 8086s back wehn IBM was already making licensing complicated.
They've certainly grown up since then, though.
I have no doubt that they'll make this change confusing, just because they always do, but the move from per-socket to per-core seems like it should come as basically no surprise from MS, or anyone else selling software whose scale is limited primarily by the power of the underlying system, rather than 'per user' or 'per seat': The number of cores, and amount of supported RAM, per socket has just skyrocketed lately, even for comparatively modest sums of money, while the sheer board complexity and need for fancy high-speed interconnect has kept socket counts relatively flat(the plummeting costs of computer equipment in general means that team supercomputer-with-custom-interconnect-fabric is still buying more sockets than ever; but among cost-sensitive customers I wouldn't be at all surprised if 8-socket systems are getting hammered, and 4 sockets dead in the water or even declining, as the number of cores you can cram into 1 or 2 socket systems increases).
On the minus side, this can't be good news for AMD: their per-core performance is lagging; but they have some parts that are kept somewhere in the running because they offer a lot of (fairly slow) cores, and support a lot of RAM, for a relatively low price(it's not terribly glamorous; but there are applications where you have a zillion lightweight http server tasks, or need a big huge memcached server for cheap and single threaded performance matters less than price). If MS is licensing per-core, without any modifiers for the power of the cores, that is going to put a great deal of emphasis on high per-core performance in any environment that tithes to Redmond. In their ideal world, they'd obviously just have a more competitive design; but AMD can't be happy that MS isn't 'weighting' cores for licensing purposes.
While I suspect that, in practice, most 'CME's will be better than flash(if only because they don't have flash's byzantine history and feature set; and people other than Adobe will be writing some of them; I'm not terribly optimistic about their chances of security in practice: Flash is, indeed, utterly broken; but because of that browsers(particularly Chrome, and Edge to at least some extent) spend a lot of time and energy actively distrusting it and trying to sandbox its inevitable failures.
Because the CDM is a DRM mechanism; it seems less likely that vendors will tolerate having their precious black-box kept on the same short leash(I'm not sure whether they'll view going full 'protected media path'/platform equivalent as worth the effort and breakage; but the more paranoid they are, the greater their desire will be to have the CDM outside the control, or even inspection, of anything that might be used to extract plaintext from it.
There's also the problem that, even if the CDM is perfect, there isn't a whole lot of reason to trust the benevolence and good intentions of the assorted media companies(and, inevitably, ads) that will be using EME on their videos, and demanding certain features from the content decryption module in order to accept it as valid for the purposes of interacting with their systems. In an ideal world, perhaps these parties would stick to "must keep encrypted content safe on its way to the screen" and nothing more; but given that the CDM can, technically, do anything a program running on the system could(possibly more, if a platform vendor includes a hardware component, TPM, dedicated video decryption and decode module, that sort of thing). It may be better at locking out random drive-by malice; but the opportunity to run high privileged code on your machine, behind a DRM-protected veil, seems far too juicy to resist misuse from 'legitimate' companies.
In a perverse sort of way, I suppose it counts as 'progress' that Li-ion batteries big enough to be dangerous are also cheap enough to find their way into dangerously shoddy crap that is just an electrical fire waiting to happen; but that doesn't change the fact that dangerously shoddy crap is just an electrical fire waiting to happen; and now it has moved into devices large enough to burn reasonably enthusiastically; which used to be the preserve of fairly expensive gear, with Ni-Cd batteries in the cheap seats.
On the other hand, unless you have a close-to-exclusive lock on some desirable routes, it can't be inexpensive to have your brand and 'hundreds dead in air crash' splattered across all major media for some days. Sure, the absolute odds are low; but if I'm looking at a list of more or less interchangeable flights between point A and point B; am I going to choose the one whose logo is subconsciously associated with huddles of grieving family members and NTSB officials thoughtfully scrutinizing piles of mangled debris?
Apparently the cracked joint just caused spurious errors; but the spurious errors caused the flight crew to pull a breaker in order to reset the system, leaving them without much of their computer assistance; which they apparently needed to get the job done.
As best I can tell from wading through the delightfully obscure document, aircraft parts should be ROHS-exempt(either Article 2, section 4, one or more of subsections c, f, or g).
Have they tightened things further, or do you suspect that somebody in the supply chain got tired of stocking ROHS and non-ROHS versions and Airbus, or the Air Asia maintenance people, didn't browbeat their supplies hard enough to get what they wanted?
A cracked solder joint that causes repeated spurious warnings, with the procedure being 'pull a circuit breaker and see if it goes away', certainly doesn't exclude poor training; and probably does suggest poor maintenance; but how much more 'true' does equipment failure need to be?
EME isn't really that much of an improvement over plugins, though it does its best to never call the 'CME' a 'plugin'. It's slightly better than just using the 'object' tag, since it standardizes a few details of interaction between javascript on the page and the 'CME'; but aside from a bunch of 'this section is non-normative' comments at the bottom to the effect of 'please don't write insecure CMEs, especially if they are going to run in a trusted context'; EME requires absolutely nothing of a CME aside from support for setting up a playback session.
If you feel like implementing a pure toy of a system, they have their goofy little cleartext-key javascript thing; but fully replacing the browser's media playback mechanism is accepted(and probably expected, since any DRM system that just hands the decrypted media back to the browser for play is going to be absurdly weak); and the standard neither requires any particular methods for interacting with the CME to be available, nor forbids the CME from having additional functions, communicating through channels unknown to the browser, or doing just about anything else it wants to.
The 'CME' does have the advantage of not starting with legacy baggage, which certainly isn't true of Flash; but architecturally browser plugins are much more tightly under the control of the browser than CMEs are.
I'm not really in the target audience; but I know that sage, at least, can optionally use sage as a frontend to mathematica. I don't know how the feature sets compare; but if you are using the sage web interface it is pretty painless to interact with a copy of mathematica installed on the sage server; instead of, or in addition to, the OSS tools that it works with.
It may just be added complexity, there may be something that mathematica handles particularly brilliantly that the parent poster has in mind; but it shouldn't be terribly unpleasant done over the network unless your problem is computationally expensive enough that it would crush the rPi whether done locally or remotely.
I'm not really sure how much staking-out of high ground is even possible until one gets closer to economic realization. Even if some treaty said that "Any touching of the asteroids is forbidden forever, with utter seriousness", one could safely enough do the R&D necessary to make grabbing them and chopping them up more practical; basically all the capabilities you'd need for asteroid mining can also be used for satellite launch, automation/robotics, improved astronomy and telescopes, and similar warm and fuzzy applications. The astronomy stuff would mostly fall under 'pure science', unless you can convince somebody that it will help detect ICBMs; but launch capabilities and improved robotics and remotely automated process research have a variety of plausible commercial applications even if the asteroids are off the table.
On the other side of the coin, highflown expressions of legal principle are usually given a great deal of latitude until they actually conflict with the interests of the nations that you need to sign and obey them. So long as the prospect is sufficiently science-fiction, anyone willing to spend a lot of time hounding UN delegates is more or less free to write whatever they want. Were somebody to step up to the table with a vaguely plausible plan, however, it's hard to imagine that they'd have much trouble finding a country large enough to be able to ignore the consequences and more than willing to do so in exchange for a cut of the take.
It seems to me that team lawyer wins more or less by default so long as the implementation isn't worked out(both because it won't actually be happening; and because there will be relatively little resistance to opining against it); but team mining will win more or less by default if they can actually make it cost effective; since laws national and international are bent, broken, or rewritten all the time for markedly less profitable(and much more ghastly) ventures.
Until that time, the posturing is symbolic(either banning the practice or laying claim to rocks you aren't already on course to intercept). If the law says nobody can do it; that will change once somebody concludes that they can turn a profit by doing it; and if I get myself crowned God Emperor of the Kupier belt now; I'll still have to get my tech up and running before somebody else does, or the claim will do me little good.
The answer appears to be no on the rPi. The BCM2835's USB port is OTG-capable, so it isn't master only; but I can find no mention of gadget device configuration actually being available. The CHIP is maybe: apparently OTG support was added in kernel 4.3; don't know how well it works, or whether the CHIP's mini-b port even has its data lines connected.
Whether or not they can actually keep the CHIP at $9, if the project does actually get Allwinner SoC support mainline and GPL compliant it would be a fairly big win. All kinds of low cost hardware is built on Allwinner parts; but the software situation is kind of dodgy, since Allwinner doesn't seem to care and the people making hardware cheap enough to use Allwinner SoCs really don't care.
With Rockchip and Mediatek in play, it's not as though they have a whole lot of room for 'mwahaha, loss leader and then suck the captive audience dry!' mustache-twirling, because the audience just isn't that captive and their parts aren't terribly special, though competent enough. It's hard to see any attempt on their part to make them more accessible as a bad thing, whether or not the intro pricing can be sustained.
While it is extremely impressive that one could do so(given that such power for such price was unheard of until very recently), it seems rather pointless to compare these two boards on the basis that you are planning on plugging in enough peripherals to use them as your next desktop. Aside from the plummeting cost and wide availability of ludicrously overqualified x86s; the most stripped down ARM SBC is obviously going to require costlier and uglier peripheral tangles than the less stripped down ones(the CHIP has no 'sibling' designs; but is similar to Allwinner-based boards from others; the rPi zero has otherwise-identical models that add the USB hub and ethernet for you, or slightly punchier options).
The only reason to go with either of these is because you have some more constrained purpose for which the low cost and small size make the difference. So, do you want HDMI support; at the cost of bringing your own NIC; or do you want the NIC; but HDMI as an optional extra? Depends on whether you are building a headless project or not.
The other matter is the software support; which is as yet an unknown. The rPi has some Broadcom blobs that are very, very, unlikely to go away; but benefits from a known, solid, supporting ecosystem. The Allwinner A8 is a bit of an unknown quantity: the project claims to be aggressively mainlining everything(which would be extremely attractive); but Allwinner's GPL compliance has been...haphazard...at best in the past; and the MALI-400 is ARM's toy, not theirs, so they have limited control over that. If the CHIP's aspirations bear out, then it will have the distinct advantage of working with mainline kernel and u-boot. If they don't, or do only in part, then the question becomes one of 'which slightly oddball BSP is better?'
I don't know how the expected lifetime service cost shakes down; but what the dealership cares about is the margins on the service and maintenance they perform; not the absolute cost.
I would suspect that battery swaps, while they involve a very expensive part, would be pretty unexciting for the dealer. Unless the manufacturer is extraordinarily tight-lipped, the price of the battery will become public knowledge; and the procedure for swapping it out(while it might require equipment that makes DIY impractical, depending on where the battery is located and what needs to be lifted) should be rigidly documented and leave little room for variation in how much labor you can bill for.
Somebody has to do the swap, and presumably they won't do it for free; but there is little room either for value-added expertise(as with problems that require diagnostic work) or just plain sleazy invoice padding(as with problems where the customer doesn't know the cost of the parts, or which parts are necessary, or what the expected labor time is); it's a rigidly scripted drop-in replacement of a single module.
If there were merely no value, it'd be an improvement. Not only does the dealership take its cut; the buyer has to interact with a car salesman, a definite negative-value-added experience.
If anything, it seems like deGrasse came closer to giving team Space!!! what they wanted to hear than I would have expected, in that he left open the implication that nation states might develop serious interest in colonizing nearby rocks and would then very likely find themselves in need of contractors for various purposes; and enable some more fully private side activities.
The ROI of getting things into earth orbit is well established; and it has a correspondingly robust market, with more outfits clamoring to enter it. Satellites are all sorts of useful and need more or less continual replacement, repair, and so on. Nobody doubts that.
The technical feasibility of snagging asteroids and chopping them up is still in the more speculative stages; but that also has an obvious possible ROI if the technical challenges can be overcome.
The case for the moon or mars, though, isn't just a matter of corporate shortsightedness, it's a matter of "Please, tell me about the ROI, within, say, the next 250 years...". Planetary colonization would undoubtedly be cool; and might be something that a nation state would get interested in as part of a prestige contest(like, say, the last time we were at all serious about the moon); but nobody ever seems to have any plans, aside from vague references to Helium 3, for what would make lunar or martian living more cost effective than some sort of aggressive colonization of underutilized desert regions or something similarly unsexy. The bounteous iron mines of mars? The endless plains of razor-sharp, static-clinging, vitrified silicates of the moon?
The point isn't whether you want to be connected. The point is that when you aren't connected it's harder to 'monetize' you; and that just isn't acceptable.
The only real reason(aside from a pathological hatred of having your ATSC/DVT-B tuner not be inside your display) is that 'big monitor' becomes increasingly hard to find if you want something fairly low resolution but physically large enough to suit a biggish living room, signage application, or the like.
You can get 'normal' monitors up to ~32 inches, with 1920x1080 being fairly cheap, even at that size, '2k' and '4k' rather more; but offerings thin considerably, and what is available gets very pricey, if you want anything bigger(the nominally-34-inch ultrawide screens are also fairly cheap; but are a poor fit for even 16:9 video, much less 4:3).
When sold as 'TVs' by contrast, you can get 1920x1080 in pretty much any size from 24-ish inches on the low end, to 65 on the high end; with '4k' up to about 80 inches in the reasonably priced section; with prices rising steeply thereafter.
It really depends on your intended viewing distance. A 32 inch panel dominates a desktop; but can look pretty anemic in a larger room; and if you can even find monitors much larger than that, they are likely to be staggeringly expensive specialty items; which is serious overkill when you plan to be sitting far enough away that the pixels will be harder to see anyway.
Assuming a suitably close viewing environment, definitely, TVs suck; but if you really do need or want a big image(and not a projector); it's pretty much a matter of picking the 'smart' TV that will whine least when you never ever connect it to the internet.
I suppose that that's the one blessing of the relatively high cost of US market cellular data: it isn't yet economic for TVs to literally phone home if they are denied free internet access. The hardware to do so is chillingly close to be plausibly cost effective; but the cost of exfiltrating any nontrivial amount of data, or serving ads, is presumably still too high.
Slashdot Mirror
Why diamond rather than gold? The value density is a little bit lower(assuming a decent quality stone); but metals tend to have a much, much, smaller gap between purchase price and sale price at any given time; can be divided if necessary without reducing their value, and are similarly imperishable, fairly hard to trace, and likely to have a buyer available almost anywhere.
As an 'investment', gold has a pretty tepid history; but at least you can actually buy and sell at reasonably close to the nominal market price, rather than eating a large automatic loss the moment you buy.
I suspect that it depends on how difficult the product of this process is to distinguish from naturally occurring diamond(assuming it can produce reasonably large crystals of suitable quality at competitive cost at all, not just industrial-grade abrasives or thin-films). There are already a variety of imitation materials available, some pretty terrible, some really quite adequate for most purposes. If these synthesized crystals can be reasonably easily distinguished(some sort of nondestructive test with reasonably low cost instruments by someone of moderate skill) then it will be easier to keep them out of the 'real' diamond supply chain; and to treat them as just one of the fakes for marketing purposes. If there is no such test available, then there will be some serious unhappiness.
Because diamonds have some extremely useful properties(superb abrasion resistance, extremely high thermal conductivity, good chemical stability, some potential as a semiconductor, etc.) trying to fight a process that can synthesize them in the necessary quantities and sizes would ultimately involve an industry that sells shiny trinkets going up against a whole bunch of larger and markedly more important industries that would have use for one or more of these properties. As long as the 'synthetics' can be distinguished from the 'real' diamonds; and stigmatized accordingly, there will be an incentive to stop trying to win the battle against all the potential industrial users; and focus instead on hyping the virtues of 'natural' diamonds, for everyone whose love isn't cold, sterile, and engineered in some laboratory.
If the only value of diamonds was being shiny, it'd be a lot easier to keep the lid on; but their properties are interesting enough that jewelry would be a footnote if bulk synthesis were available.
There's also the 'price' consideration: Unless you get really lucky and score somebody's cherished childhood collection at a yard sale from their parents who don't know that the lego secondary market is a well-oiled machine; enough lego(especially Technic/Mindstorms) to build something complex enough to be interesting to computer control gets pretty pricey. Cheaper than grabbing the McMaster-Carr catalog and ordering the moving parts from them; but still pretty expensive.
Perhaps there are people who actively prefer virtual building blocks, I don't know; but it is undoubtedly the case that you can't get much lego for whatever Minecraft costs, and a lego habit can get punishingly expensive.
I don't know how they do/did it; but Minecraft somehow made the jump from being a moderately novel; but not obviously interesting, "8-bit-styled" 3d multiplayer sandbox game(with some optional combat elements), sort of the 'Duplo' to Dwarf Fortress' 'Technic'; to having an enormous amount of mod support, which has mostly eclipsed such original 'game' as there actually was.(And not just 'mod support' in the 'Bethesda builds a world; and then the players fix it' sense; the 3rd party components include things like the popular servers, the popular mod interfaces, and similar lower level than usual stuff).
I'm not sure how it came to that; but that vast availability of 3rd party features seems to have become the major draw.
I realize that this probably makes me a horrible person, whose shriveled soul is incapable of joy, and who was obviously never a child; but exactly how much external infrastructure are you allowed to use before "In Minecraft!" is no longer true in any meaningful sense?
The various redstone logic arrangements, while obviously perverse and inefficient, do implement various computational widgets "in minecraft". As best I can tell from reading the project documentation, this exercise uses an external helper program to do all the heavy lifting(digesting web pages and incoming video into an array of textures to be applied to blocks, implementing the MMS for the 'selfie' feature), with the only involvement from Minecraft being a large rectangular array of blocks that the external helper program re-textures to produce something similar to a framebuffer(it's not quite the same, since each block displays a multi-pixel texture, rather than acting as a single pixel).
You certainly couldn't do this with just any game(at least not as easily; if you have access to the game's memory, you can probably scribble on its textures; but doing that without bringing the whole mess down in a screaming heap is easier said than done); but that arguably makes it less interesting: Minecraft's support for external modification of server state is relatively robust, so there really isn't much that you can't do, if you are willing to do most of the work in an external program and then use Minecraft as a needlessly perverse frontend. Am I just a joyless asshole? A generation too old for this 'minecraft' stuff? Insufficiently impressed by a Verizion marketing exercise?
Microsoft was knee-high to a grasshopper and might not have even yet been in the business of selling word processors to Mac users and a cheapo 'disk operating system' for 8086s back wehn IBM was already making licensing complicated.
They've certainly grown up since then, though.
I have no doubt that they'll make this change confusing, just because they always do, but the move from per-socket to per-core seems like it should come as basically no surprise from MS, or anyone else selling software whose scale is limited primarily by the power of the underlying system, rather than 'per user' or 'per seat': The number of cores, and amount of supported RAM, per socket has just skyrocketed lately, even for comparatively modest sums of money, while the sheer board complexity and need for fancy high-speed interconnect has kept socket counts relatively flat(the plummeting costs of computer equipment in general means that team supercomputer-with-custom-interconnect-fabric is still buying more sockets than ever; but among cost-sensitive customers I wouldn't be at all surprised if 8-socket systems are getting hammered, and 4 sockets dead in the water or even declining, as the number of cores you can cram into 1 or 2 socket systems increases).
On the minus side, this can't be good news for AMD: their per-core performance is lagging; but they have some parts that are kept somewhere in the running because they offer a lot of (fairly slow) cores, and support a lot of RAM, for a relatively low price(it's not terribly glamorous; but there are applications where you have a zillion lightweight http server tasks, or need a big huge memcached server for cheap and single threaded performance matters less than price). If MS is licensing per-core, without any modifiers for the power of the cores, that is going to put a great deal of emphasis on high per-core performance in any environment that tithes to Redmond. In their ideal world, they'd obviously just have a more competitive design; but AMD can't be happy that MS isn't 'weighting' cores for licensing purposes.
While I suspect that, in practice, most 'CME's will be better than flash(if only because they don't have flash's byzantine history and feature set; and people other than Adobe will be writing some of them; I'm not terribly optimistic about their chances of security in practice: Flash is, indeed, utterly broken; but because of that browsers(particularly Chrome, and Edge to at least some extent) spend a lot of time and energy actively distrusting it and trying to sandbox its inevitable failures.
Because the CDM is a DRM mechanism; it seems less likely that vendors will tolerate having their precious black-box kept on the same short leash(I'm not sure whether they'll view going full 'protected media path'/platform equivalent as worth the effort and breakage; but the more paranoid they are, the greater their desire will be to have the CDM outside the control, or even inspection, of anything that might be used to extract plaintext from it.
There's also the problem that, even if the CDM is perfect, there isn't a whole lot of reason to trust the benevolence and good intentions of the assorted media companies(and, inevitably, ads) that will be using EME on their videos, and demanding certain features from the content decryption module in order to accept it as valid for the purposes of interacting with their systems. In an ideal world, perhaps these parties would stick to "must keep encrypted content safe on its way to the screen" and nothing more; but given that the CDM can, technically, do anything a program running on the system could(possibly more, if a platform vendor includes a hardware component, TPM, dedicated video decryption and decode module, that sort of thing). It may be better at locking out random drive-by malice; but the opportunity to run high privileged code on your machine, behind a DRM-protected veil, seems far too juicy to resist misuse from 'legitimate' companies.
In a perverse sort of way, I suppose it counts as 'progress' that Li-ion batteries big enough to be dangerous are also cheap enough to find their way into dangerously shoddy crap that is just an electrical fire waiting to happen; but that doesn't change the fact that dangerously shoddy crap is just an electrical fire waiting to happen; and now it has moved into devices large enough to burn reasonably enthusiastically; which used to be the preserve of fairly expensive gear, with Ni-Cd batteries in the cheap seats.
On the other hand, unless you have a close-to-exclusive lock on some desirable routes, it can't be inexpensive to have your brand and 'hundreds dead in air crash' splattered across all major media for some days. Sure, the absolute odds are low; but if I'm looking at a list of more or less interchangeable flights between point A and point B; am I going to choose the one whose logo is subconsciously associated with huddles of grieving family members and NTSB officials thoughtfully scrutinizing piles of mangled debris?
If they had gone to the trouble of establishing themselves as important characters; that strategy might well have worked.
Apparently the cracked joint just caused spurious errors; but the spurious errors caused the flight crew to pull a breaker in order to reset the system, leaving them without much of their computer assistance; which they apparently needed to get the job done.
As best I can tell from wading through the delightfully obscure document, aircraft parts should be ROHS-exempt(either Article 2, section 4, one or more of subsections c, f, or g).
Have they tightened things further, or do you suspect that somebody in the supply chain got tired of stocking ROHS and non-ROHS versions and Airbus, or the Air Asia maintenance people, didn't browbeat their supplies hard enough to get what they wanted?
A cracked solder joint that causes repeated spurious warnings, with the procedure being 'pull a circuit breaker and see if it goes away', certainly doesn't exclude poor training; and probably does suggest poor maintenance; but how much more 'true' does equipment failure need to be?
EME isn't really that much of an improvement over plugins, though it does its best to never call the 'CME' a 'plugin'. It's slightly better than just using the 'object' tag, since it standardizes a few details of interaction between javascript on the page and the 'CME'; but aside from a bunch of 'this section is non-normative' comments at the bottom to the effect of 'please don't write insecure CMEs, especially if they are going to run in a trusted context'; EME requires absolutely nothing of a CME aside from support for setting up a playback session.
If you feel like implementing a pure toy of a system, they have their goofy little cleartext-key javascript thing; but fully replacing the browser's media playback mechanism is accepted(and probably expected, since any DRM system that just hands the decrypted media back to the browser for play is going to be absurdly weak); and the standard neither requires any particular methods for interacting with the CME to be available, nor forbids the CME from having additional functions, communicating through channels unknown to the browser, or doing just about anything else it wants to.
The 'CME' does have the advantage of not starting with legacy baggage, which certainly isn't true of Flash; but architecturally browser plugins are much more tightly under the control of the browser than CMEs are.
I'm not really in the target audience; but I know that sage, at least, can optionally use sage as a frontend to mathematica. I don't know how the feature sets compare; but if you are using the sage web interface it is pretty painless to interact with a copy of mathematica installed on the sage server; instead of, or in addition to, the OSS tools that it works with.
It may just be added complexity, there may be something that mathematica handles particularly brilliantly that the parent poster has in mind; but it shouldn't be terribly unpleasant done over the network unless your problem is computationally expensive enough that it would crush the rPi whether done locally or remotely.
I'm not really sure how much staking-out of high ground is even possible until one gets closer to economic realization. Even if some treaty said that "Any touching of the asteroids is forbidden forever, with utter seriousness", one could safely enough do the R&D necessary to make grabbing them and chopping them up more practical; basically all the capabilities you'd need for asteroid mining can also be used for satellite launch, automation/robotics, improved astronomy and telescopes, and similar warm and fuzzy applications. The astronomy stuff would mostly fall under 'pure science', unless you can convince somebody that it will help detect ICBMs; but launch capabilities and improved robotics and remotely automated process research have a variety of plausible commercial applications even if the asteroids are off the table.
On the other side of the coin, highflown expressions of legal principle are usually given a great deal of latitude until they actually conflict with the interests of the nations that you need to sign and obey them. So long as the prospect is sufficiently science-fiction, anyone willing to spend a lot of time hounding UN delegates is more or less free to write whatever they want. Were somebody to step up to the table with a vaguely plausible plan, however, it's hard to imagine that they'd have much trouble finding a country large enough to be able to ignore the consequences and more than willing to do so in exchange for a cut of the take.
It seems to me that team lawyer wins more or less by default so long as the implementation isn't worked out(both because it won't actually be happening; and because there will be relatively little resistance to opining against it); but team mining will win more or less by default if they can actually make it cost effective; since laws national and international are bent, broken, or rewritten all the time for markedly less profitable(and much more ghastly) ventures.
Until that time, the posturing is symbolic(either banning the practice or laying claim to rocks you aren't already on course to intercept). If the law says nobody can do it; that will change once somebody concludes that they can turn a profit by doing it; and if I get myself crowned God Emperor of the Kupier belt now; I'll still have to get my tech up and running before somebody else does, or the claim will do me little good.
The answer appears to be no on the rPi. The BCM2835's USB port is OTG-capable, so it isn't master only; but I can find no mention of gadget device configuration actually being available. The CHIP is maybe: apparently OTG support was added in kernel 4.3; don't know how well it works, or whether the CHIP's mini-b port even has its data lines connected.
Whether or not they can actually keep the CHIP at $9, if the project does actually get Allwinner SoC support mainline and GPL compliant it would be a fairly big win. All kinds of low cost hardware is built on Allwinner parts; but the software situation is kind of dodgy, since Allwinner doesn't seem to care and the people making hardware cheap enough to use Allwinner SoCs really don't care.
With Rockchip and Mediatek in play, it's not as though they have a whole lot of room for 'mwahaha, loss leader and then suck the captive audience dry!' mustache-twirling, because the audience just isn't that captive and their parts aren't terribly special, though competent enough. It's hard to see any attempt on their part to make them more accessible as a bad thing, whether or not the intro pricing can be sustained.
While it is extremely impressive that one could do so(given that such power for such price was unheard of until very recently), it seems rather pointless to compare these two boards on the basis that you are planning on plugging in enough peripherals to use them as your next desktop. Aside from the plummeting cost and wide availability of ludicrously overqualified x86s; the most stripped down ARM SBC is obviously going to require costlier and uglier peripheral tangles than the less stripped down ones(the CHIP has no 'sibling' designs; but is similar to Allwinner-based boards from others; the rPi zero has otherwise-identical models that add the USB hub and ethernet for you, or slightly punchier options).
The only reason to go with either of these is because you have some more constrained purpose for which the low cost and small size make the difference. So, do you want HDMI support; at the cost of bringing your own NIC; or do you want the NIC; but HDMI as an optional extra? Depends on whether you are building a headless project or not.
The other matter is the software support; which is as yet an unknown. The rPi has some Broadcom blobs that are very, very, unlikely to go away; but benefits from a known, solid, supporting ecosystem. The Allwinner A8 is a bit of an unknown quantity: the project claims to be aggressively mainlining everything(which would be extremely attractive); but Allwinner's GPL compliance has been...haphazard...at best in the past; and the MALI-400 is ARM's toy, not theirs, so they have limited control over that. If the CHIP's aspirations bear out, then it will have the distinct advantage of working with mainline kernel and u-boot. If they don't, or do only in part, then the question becomes one of 'which slightly oddball BSP is better?'
I don't know how the expected lifetime service cost shakes down; but what the dealership cares about is the margins on the service and maintenance they perform; not the absolute cost.
I would suspect that battery swaps, while they involve a very expensive part, would be pretty unexciting for the dealer. Unless the manufacturer is extraordinarily tight-lipped, the price of the battery will become public knowledge; and the procedure for swapping it out(while it might require equipment that makes DIY impractical, depending on where the battery is located and what needs to be lifted) should be rigidly documented and leave little room for variation in how much labor you can bill for.
Somebody has to do the swap, and presumably they won't do it for free; but there is little room either for value-added expertise(as with problems that require diagnostic work) or just plain sleazy invoice padding(as with problems where the customer doesn't know the cost of the parts, or which parts are necessary, or what the expected labor time is); it's a rigidly scripted drop-in replacement of a single module.
If there were merely no value, it'd be an improvement. Not only does the dealership take its cut; the buyer has to interact with a car salesman, a definite negative-value-added experience.
If anything, it seems like deGrasse came closer to giving team Space!!! what they wanted to hear than I would have expected, in that he left open the implication that nation states might develop serious interest in colonizing nearby rocks and would then very likely find themselves in need of contractors for various purposes; and enable some more fully private side activities.
The ROI of getting things into earth orbit is well established; and it has a correspondingly robust market, with more outfits clamoring to enter it. Satellites are all sorts of useful and need more or less continual replacement, repair, and so on. Nobody doubts that.
The technical feasibility of snagging asteroids and chopping them up is still in the more speculative stages; but that also has an obvious possible ROI if the technical challenges can be overcome.
The case for the moon or mars, though, isn't just a matter of corporate shortsightedness, it's a matter of "Please, tell me about the ROI, within, say, the next 250 years...". Planetary colonization would undoubtedly be cool; and might be something that a nation state would get interested in as part of a prestige contest(like, say, the last time we were at all serious about the moon); but nobody ever seems to have any plans, aside from vague references to Helium 3, for what would make lunar or martian living more cost effective than some sort of aggressive colonization of underutilized desert regions or something similarly unsexy. The bounteous iron mines of mars? The endless plains of razor-sharp, static-clinging, vitrified silicates of the moon?
The point isn't whether you want to be connected. The point is that when you aren't connected it's harder to 'monetize' you; and that just isn't acceptable.
The only real reason(aside from a pathological hatred of having your ATSC/DVT-B tuner not be inside your display) is that 'big monitor' becomes increasingly hard to find if you want something fairly low resolution but physically large enough to suit a biggish living room, signage application, or the like.
You can get 'normal' monitors up to ~32 inches, with 1920x1080 being fairly cheap, even at that size, '2k' and '4k' rather more; but offerings thin considerably, and what is available gets very pricey, if you want anything bigger(the nominally-34-inch ultrawide screens are also fairly cheap; but are a poor fit for even 16:9 video, much less 4:3).
When sold as 'TVs' by contrast, you can get 1920x1080 in pretty much any size from 24-ish inches on the low end, to 65 on the high end; with '4k' up to about 80 inches in the reasonably priced section; with prices rising steeply thereafter.
It really depends on your intended viewing distance. A 32 inch panel dominates a desktop; but can look pretty anemic in a larger room; and if you can even find monitors much larger than that, they are likely to be staggeringly expensive specialty items; which is serious overkill when you plan to be sitting far enough away that the pixels will be harder to see anyway.
Assuming a suitably close viewing environment, definitely, TVs suck; but if you really do need or want a big image(and not a projector); it's pretty much a matter of picking the 'smart' TV that will whine least when you never ever connect it to the internet.
I suppose that that's the one blessing of the relatively high cost of US market cellular data: it isn't yet economic for TVs to literally phone home if they are denied free internet access. The hardware to do so is chillingly close to be plausibly cost effective; but the cost of exfiltrating any nontrivial amount of data, or serving ads, is presumably still too high.