But isn't that the ideal way to run government? Like a business? Instead of stealing our hard-earned money?
There are arguments to be made in favor of running certain aspects of government like a business(if our national parks got to set ticket prices according to actual demand, they'd have a lot less trouble with understaffing and overcrowding...); but there are some issues to watch out for:
1. If you are a business, you have 'customers' and you are beholden(sooner or later) to provide 'customer service'. Be very careful that your 'customers' are the same people that it is your mission to serve and that 'customer service' is the same product that it is your mission to provide. This is a particular problem with government departments that have regulatory functions. In terms of day-to-day interaction, shared professional backgrounds and skill sets, etc. the 'customer' is usually the party who needs to be kept in line; but the mission of the department is the protection of the public(who should be the 'customer'; but who the regulators rarely interact with). In the case of the USPTO, the de-facto 'customer' ends up being the patent applicant, not the vague, voiceless, largely inchoate mass of 'people who don't want inefficiencies introduced by bad patents'. It's natural enough, and likely to progress even faster if the entity is overwhelmed by its caseload, or if there is a revolving door between USPTO examiners and corporate patent attorneys(which, even in the absence of corruption of any kind, the fact that similar skills are required by both jobs tends to mean will happen to some degree).
(To end on a positive note) The institution of 'Agricultural extension programs', typically associated(in the US) with the research programmes and faculty experts at local Land-Grant Colleges and Universities that operates reasonably successfully as a sort of 'like a business; but in a broad sense' program. Their objective is the improvement of agricultural standards and outcomes in their area, through consultation and expertise on local conditions, pests, etc. along with research made possible in part through access to the data gathered by working with the agricultural population at large, and often offering certain soil testing, analytical, and pest identification services at accessible prices. These aren't "like a business" in the sense that they are run for-profit, and they do have a basic research, R&D, and educational mission; but they are operated as an essentially pragmatic, productivity and profit improving, adjunct to private agriculture in their region.
Those guys look similar to Meraki, before they starting branching out into wired stuff as well. Having been there when Meraki made the Big Switch and firmware-updated everybody out of their devices and into a 'cloud managed only' model, are there any sellers you know of that don't require some slightly creepy hookup to the mothership but do make things a trifle easier than "Obtain a FooCorp Model X router(version 3 only, other versions contain totally different chipset and run VXworks) and flash the one true OpenWRT_bleeding_edge_experimental with the kernel patches from xhackerx's git repository"?
Of the ones I've seen, mesh gear is either "Dead simple, The Cloud handles it for you, until we decide to change the terms of service." or moderately heavy firmware tweaker stuff that depends on router hardware sourced from mass market vendors who change chipsets about three times as fast as they change model numbers.
Do you know anybody who handles meshing out of the box/with actual stability in hardware you can buy, (I'm not averse to paying more than Best Buy's sale rack); but with local control?
At least in the US, it isn't entirely clear that the mobile vendors want to decimate the old school ISP model except where it fits their existing business interests.
If you are a wireline ISP, and already have sunk costs(quite literally) buried in the dirt, it only makes sense to 'decimate' the old school model if you can make more money by selling cut-price wireless access, rather than selling your already-amortized wireline services comparatively cheaply and bleeding the business users and/or compulsive smartphone crowd for their data use.
It has been argued that Verizon is actually doing this to their DSL and copper POTS phone services, in order to rid themselves of their gradually decaying copper infrastructure and its unionized linemen. They are still competing in areas where they already have fiber laid; but their DSL pricing and plans have actually gotten sharply worse, with apparent indifference to the fact that this is handing those customers over to cable, where it has a presence. The cable guys, for their part, tend not to have a wireless service that they can work with directly; but they very aggressively bundle cable 'content' packages with internet service' and have been pretty clear about their desire to restrict video-over-IP that competes with cable content exclusively to cable subscribers as much as possible(note, for instance, all the cable company ads that feature somebody streaming on their smartphone/tablet, on the go(ie. not on the internet service they are selling; but thanks to the 'content' service they are selling) rather than the "Cable is a fuckload faster than DSL, switch now you peons!" message from a few years back..
There have been a few attempts("Clearwire" comes to mind); but that road seems littered with skeletons and desperation.
If the bottom plan (500MB) is five minutes (10 seconds a day), the top plan is only 8GB, which, by my calculations, is only 16 times.
So... 160 seconds a day on average? That's the maximum plan?
What's the point of even having speeds that fast? (besides marketing)
My suspicion would be that the sales pitch(aside from a simple 4 IS BIGGER THAN 3 AND THEREFORE BETTER CONSUME!!!) would be very similar to to one given to Ma and Pa AOLer in the earlier days of wireline broadband: instant gratification.
If I'm a light user, I may only get 10mb of email(hey FWD:FWD:FWD:FWD:FWD:Inspirational!! has some cute puppy pictures with saccharine quotes attached, that adds up...); but I also only check once every day or three. If I'm on 56k, I'll be twiddling my fingers for ages while my computer does its POP3 thing, and won't even be able to check my Yahoo horoscope in under 10 minutes while my email is loading. If I upgrade to DSL or cable, my inbox will be pulled down in moments, and I'll be able to do anything I'd ever care to do at the same time.
In the phone case, since staying on all the time and quietly pulling things down in the background murders your battery, very high burst speeds might be able to offer faster time-to-gratification if somebody picks up their phone, knocks it out of standby, and expects their twitfeed or mailbox to update for use, or some youtube clip to load.
A heavily capped plan is useless to truly heavy data users; but high peak speeds do substantially improve user experience in terms of time between user request and completion of retrieval and rendering of whatever it is they asked for.
What in the world do you intend to do with an iPad that could possibly use 15GB a month? I've owned a 3G iPad, literally since the day it first retailed, and I don't think I've used that much in total.
It's also worth noting that the existence of potentially useful services that would consume data tends(with a certain amount of 'lived-before-my-time' leapfrogging and inertial lagging) to depend on available bandwidth(and vice versa, to some degree).
Nobody is going to bother rolling out a service like netflix streaming or that-video-game-streaming-outfit-that-just-died-horribly if there aren't enough people with broadband(or who can upgrade to broadband just by switching plans) to be potential customers. There also isn't much incentive to get out the backhoes and put new wires in the ground if there are very few people willing to pay a premium for more bandwidth, because there isn't anything to spend that bandwidth on.
To some extent, protocols that spawned on academic or corporate campuses and then migrated out provide some built-in, market-insensitive, impetus: if all your evolving is done on screaming fast LANs you have a great deal of freedom to build cool stuff without worrying about who has dialup and who has fiber-to-premises. The merry world of piracy also does some good work in this area, since the costs of entry are free-as-in-stolen(unlike something like netflix); and the content offerings(again, because of the ease with which new content partners are involuntarily brought in to the ecosystem) scale neatly from shitty 96kb/s mp3 rips and ultra-crunchy video snippets, up to high res textbook scans and bit-for-bit BD-ROM rips, providing a convenient ladder of incentives for lower bandwidth people to move up to better services.
For those customers who aren't corporate/geek enough to make heavy use of LAN-dwelling protocols and capabilities, and who aren't motivated to hoist the jolly roger and set sail for the bay, there is very much a reciprocal relationship between available bandwidth and available services. Not much incentive to improve bandwidth if nobody can buy services to use it on, and not much incentive to start services if nobody has the bandwidth needed to become a customer...
As a competitive force, that falls into the "Oh, you math geeks are so adorable. Why don't you go off and play in the ISM band where the big kids' table doesn't have to look at you..." category.
So, guys... how's that whole "Let the market decide" argument working out for you? Capitalism works great for non-critical, non-infrastructure goods and services... but when it gets its hands on something everybody needs, it's gonna take you to the cleaners. Every single time.
It's not so much the criticality(or lack thereof) that gets you or saves you; but the barriers to entry and costs of duplicate entrants.
Plenty of user-critical-without-which-modern-civilization-would-come-to-a-screeching-halt goods get produced just fine under market conditions, so long as the barriers to entry aren't too high and competition between suppliers of identical or reasonably substitutable goods is fairly robust.
Infrastructure, of course, is sort of a classic case of high barriers to entry coupled with high costs of duplicate entrants. You've got a finite amount of desirable spectrum to work with, and every market entrant needs to put a net-connected box of RF gear on a pole in every chunk of land they want to provide service for. The spectrum and buildout hassles tend to leave you with an oligopoly, at best, among people who actually provide wireless bandwidth, and possibly some MVOs who provide specialized customer support/billing/advertising offerings to sell bandwidth to subscriber groups that the primary telcos don't want to deal with.
In addition to having strong 'natural monopoly' characteristics, Infrastructure goods and services bite you twice: Not only are they natural monopolies/oligopolies, and thus tend to fall into ugly pricing, they are infrastructure, something over which other goods and services and interactions take place, so they are extremely well placed to allow holders of market power in the infrastructure market to exert it in any of the markets (or socially valuable nonmarket activities like 'communication') that occur on top of that infrastructure. Wacky fun!
" 'It's evolving literally to be a single platform,' he said, referring to how Windows 8 and Windows Phone 8 share a kernel, file system, graphics support, and other elements. At least in theory, that will allow developers to port apps from the desktop/tablet OS to the smartphone OS with relatively little work."
Hasn't Gates been chasing the dream of one Windows to rule them all for something like two decades now? The line of 'Handheld PC' and 'Pocket PC' devices didn't share as much low level architecture, because the hardware wouldn't permit it at the time; but did everything they could to drag a desktop UI onto a teeny touchscreen, and 'tablet' meant getting Windows for Pen Computing 1.0 with your Win3.1 back when meteorites were still mopping up the last of the dinosaurs....
The other issue with the "give it to a 3 year old for 10 minutes" test, in addition to the 'unlearning' piece, is that it only measures the slope of the first tiny bit of the learning curve. What people who actually use computers(or who have to do actual work with them) really care about is whether the interface makes them more powerful and efficient in exchange for a modicum of skill.
When did the notion start to circulate that anything remotely related to wood is some kind of incendiary deathtrap? Is it when people stopped having to start fires with nothing more than minimal tools and careful arrangement of sticks?
Christ, you've got something designed to cut through plastics with a laser, plastics which are basically just waiting for some added heat to turn into sticky, flaming, hydrocarbon death, and nobody says a thing. Suddenly, terrifying wood,. notorious for perfunctory smoldering in response to heat, bursts onto the scene and everybody is freaking out about ignition. Kids these days.
Somehow, people have been practicing pyrography for millenia without bursting into flames.
How would you suggest they solve the funding dilemma?
The details are a nighmarish, ever-shifting, morass of acronyms and "talk to your rep"; but Microsoft already has a plan where customers can pay annually for "Software Assurance".
Obviously, Microsoft wants to be bug-hunting as few codebases as possible, and presenting as unified a "platform" to 3rd party application vendors as possible, so it is to be expected that the cost/seat of continued support would rise over time as the number of seats in the field dropped; but they already have a mechanism for charging customers on an annual volume basis, which could easily enough be extended to offering paid security support after the support period provided by a boxed license is over...
Not necessarily something that they want to do; but they could.
We'll see if they start playing any games with WHQL/driver signing. If they don't, it's the hardware vendors' problem. If it turns out that "Windows XP driver Exists" is an unofficial ticket to substantial additional delays in getting the WHQL signature for the Win7/Win8 driver... Well, that'll be a different story.
What kind of arm-twisting, exactly, is MS planning against Dell, HP, etc. to get them to stop shipping boring corporate boxes that don't support XP?
Yeah, sure, the odds of having XP run properly without a bit of scrounging on some random machine from Best Buy(this goes double if it's a laptop, triple if it's some wacky touch/hybrid/thing), aren't getting any better; but if your business is shipping pallet-loads of identical machines to assorted volume customers, you damn well better support the OSes they want supported. If you don't, the largely interchangeable shipper of near-identical machines will.
Even if MS plays serious hardball, and just starts refusing to WHQL sign XP drivers, XP doesn't force driver signing very hard, so IT shouldn't have much trouble with that. Now, I'd be totally unsurprised to learn that XP toasts the battery life of newer laptops with super-fancy power saving features, or requires that you turn on the 'legacy bios emulation' switch in whatever UEFI pit the system ships with; but I'd be shocked to see the end of the ability to buy XP boxes(through corporate and volume license channels, not necessarily at retail) before 2020...
Wasn't there an entire section in Snow Crash about 'Gargoyles' who had embraced wearable computing in a big way, and how everybody hated them and thought that they were freaks?
I think that having a local MTA is still pretty normal on Linux(debian certainly gave me the rudiments of an exim4 install the last time I reinstalled it); but it seems that, even there, the MTA is largely cut out of the loop. It still shuttles system emails around, mostly warnings and such sent to root; but if you install Thunderbird or Kmail or whatnot, those packages completely ignore exim unless you know what you are doing and make the appropriate changes to both the exim configuration and the client config. The interface defaults on the mail client packages are very much aimed at getting you connected directly to your IMAP/POP3 account of choice, and don't even mention the local MTA.
The AIX on RS/6000 system I used to telnet into to check my mail in pine certainly took its local MTA more seriously; but I think that it was run by traditionalists...
In any event, it will be interesting to see how this school district manages.
What I want to know is where the http://pixelqi.com/ guys are hiding... They had a workable device, shipping in nontrivial volume with the OLPC XO-1, and then seemingly dropped off the map.
All the refresh rate of an LCD panel(because it is one); but, in transreflective mode, looks more like e-ink than any LCD I've ever seen and has the option to do color if you crank the backlight....
We know(because all but the nastiest LCD tablets running Android or iOS can and do do it) that contemporary low-power ARM chipsets are up to the challenge of crunching PDFs; but e-ink displays are mostly too small to display 8.5x11 or A4 pages, too slow for panning/zooming/etc, and PDF reflow is crap. If they would just start existing, the Pixel Qi screens would fairly efficiently solve this problem, at lower cost and lower power than standard LCD panels; but nobody seems to have heard a peep from them.
Thankfully, it is axiomatic that Stallman Is An Extremist, so we needn't listen to his(often strident, as often correct) warnings!
The awesome thing about the emerging DRM economy is that it combines the economic relations of feudalism with the efficient, data-driven surveillance that East Germany was too low-tech to achieve...
Interesting. I have one of the little RTL2832/E4000 dongles plugged in right now, and I'm familiar by reputation with the USRP stuff; but my knowledge of the midrange, and any of the gear that has evolved from the hobbyist radio side, or that is designed for crunching RF down to something that a soundcard works with, is pretty much zero.
Just use the magic word "H1B" and all that changes... The disappearance of other people's jobs is just the inevitable march of progress, and will probably make them better off in the long run anyway. My job, though, now that's a different story...
Slashdot Mirror
But isn't that the ideal way to run government? Like a business? Instead of stealing our hard-earned money?
There are arguments to be made in favor of running certain aspects of government like a business(if our national parks got to set ticket prices according to actual demand, they'd have a lot less trouble with understaffing and overcrowding...); but there are some issues to watch out for:
1. If you are a business, you have 'customers' and you are beholden(sooner or later) to provide 'customer service'. Be very careful that your 'customers' are the same people that it is your mission to serve and that 'customer service' is the same product that it is your mission to provide. This is a particular problem with government departments that have regulatory functions. In terms of day-to-day interaction, shared professional backgrounds and skill sets, etc. the 'customer' is usually the party who needs to be kept in line; but the mission of the department is the protection of the public(who should be the 'customer'; but who the regulators rarely interact with). In the case of the USPTO, the de-facto 'customer' ends up being the patent applicant, not the vague, voiceless, largely inchoate mass of 'people who don't want inefficiencies introduced by bad patents'. It's natural enough, and likely to progress even faster if the entity is overwhelmed by its caseload, or if there is a revolving door between USPTO examiners and corporate patent attorneys(which, even in the absence of corruption of any kind, the fact that similar skills are required by both jobs tends to mean will happen to some degree).
(To end on a positive note) The institution of 'Agricultural extension programs', typically associated(in the US) with the research programmes and faculty experts at local Land-Grant Colleges and Universities that operates reasonably successfully as a sort of 'like a business; but in a broad sense' program. Their objective is the improvement of agricultural standards and outcomes in their area, through consultation and expertise on local conditions, pests, etc. along with research made possible in part through access to the data gathered by working with the agricultural population at large, and often offering certain soil testing, analytical, and pest identification services at accessible prices. These aren't "like a business" in the sense that they are run for-profit, and they do have a basic research, R&D, and educational mission; but they are operated as an essentially pragmatic, productivity and profit improving, adjunct to private agriculture in their region.
Don't worry, 'rubber banding' on all mainframes with touchscreen interfaces is still safe...
Those guys look similar to Meraki, before they starting branching out into wired stuff as well. Having been there when Meraki made the Big Switch and firmware-updated everybody out of their devices and into a 'cloud managed only' model, are there any sellers you know of that don't require some slightly creepy hookup to the mothership but do make things a trifle easier than "Obtain a FooCorp Model X router(version 3 only, other versions contain totally different chipset and run VXworks) and flash the one true OpenWRT_bleeding_edge_experimental with the kernel patches from xhackerx's git repository"?
Of the ones I've seen, mesh gear is either "Dead simple, The Cloud handles it for you, until we decide to change the terms of service." or moderately heavy firmware tweaker stuff that depends on router hardware sourced from mass market vendors who change chipsets about three times as fast as they change model numbers.
Do you know anybody who handles meshing out of the box/with actual stability in hardware you can buy, (I'm not averse to paying more than Best Buy's sale rack); but with local control?
At least in the US, it isn't entirely clear that the mobile vendors want to decimate the old school ISP model except where it fits their existing business interests.
If you are a wireline ISP, and already have sunk costs(quite literally) buried in the dirt, it only makes sense to 'decimate' the old school model if you can make more money by selling cut-price wireless access, rather than selling your already-amortized wireline services comparatively cheaply and bleeding the business users and/or compulsive smartphone crowd for their data use.
It has been argued that Verizon is actually doing this to their DSL and copper POTS phone services, in order to rid themselves of their gradually decaying copper infrastructure and its unionized linemen. They are still competing in areas where they already have fiber laid; but their DSL pricing and plans have actually gotten sharply worse, with apparent indifference to the fact that this is handing those customers over to cable, where it has a presence. The cable guys, for their part, tend not to have a wireless service that they can work with directly; but they very aggressively bundle cable 'content' packages with internet service' and have been pretty clear about their desire to restrict video-over-IP that competes with cable content exclusively to cable subscribers as much as possible(note, for instance, all the cable company ads that feature somebody streaming on their smartphone/tablet, on the go(ie. not on the internet service they are selling; but thanks to the 'content' service they are selling) rather than the "Cable is a fuckload faster than DSL, switch now you peons!" message from a few years back..
There have been a few attempts("Clearwire" comes to mind); but that road seems littered with skeletons and desperation.
If the bottom plan (500MB) is five minutes (10 seconds a day), the top plan is only 8GB, which, by my calculations, is only 16 times.
So... 160 seconds a day on average? That's the maximum plan?
What's the point of even having speeds that fast? (besides marketing)
My suspicion would be that the sales pitch(aside from a simple 4 IS BIGGER THAN 3 AND THEREFORE BETTER CONSUME!!!) would be very similar to to one given to Ma and Pa AOLer in the earlier days of wireline broadband: instant gratification.
If I'm a light user, I may only get 10mb of email(hey FWD:FWD:FWD:FWD:FWD:Inspirational!! has some cute puppy pictures with saccharine quotes attached, that adds up...); but I also only check once every day or three. If I'm on 56k, I'll be twiddling my fingers for ages while my computer does its POP3 thing, and won't even be able to check my Yahoo horoscope in under 10 minutes while my email is loading. If I upgrade to DSL or cable, my inbox will be pulled down in moments, and I'll be able to do anything I'd ever care to do at the same time.
In the phone case, since staying on all the time and quietly pulling things down in the background murders your battery, very high burst speeds might be able to offer faster time-to-gratification if somebody picks up their phone, knocks it out of standby, and expects their twitfeed or mailbox to update for use, or some youtube clip to load.
A heavily capped plan is useless to truly heavy data users; but high peak speeds do substantially improve user experience in terms of time between user request and completion of retrieval and rendering of whatever it is they asked for.
What in the world do you intend to do with an iPad that could possibly use 15GB a month? I've owned a 3G iPad, literally since the day it first retailed, and I don't think I've used that much in total.
It's also worth noting that the existence of potentially useful services that would consume data tends(with a certain amount of 'lived-before-my-time' leapfrogging and inertial lagging) to depend on available bandwidth(and vice versa, to some degree).
Nobody is going to bother rolling out a service like netflix streaming or that-video-game-streaming-outfit-that-just-died-horribly if there aren't enough people with broadband(or who can upgrade to broadband just by switching plans) to be potential customers. There also isn't much incentive to get out the backhoes and put new wires in the ground if there are very few people willing to pay a premium for more bandwidth, because there isn't anything to spend that bandwidth on.
To some extent, protocols that spawned on academic or corporate campuses and then migrated out provide some built-in, market-insensitive, impetus: if all your evolving is done on screaming fast LANs you have a great deal of freedom to build cool stuff without worrying about who has dialup and who has fiber-to-premises. The merry world of piracy also does some good work in this area, since the costs of entry are free-as-in-stolen(unlike something like netflix); and the content offerings(again, because of the ease with which new content partners are involuntarily brought in to the ecosystem) scale neatly from shitty 96kb/s mp3 rips and ultra-crunchy video snippets, up to high res textbook scans and bit-for-bit BD-ROM rips, providing a convenient ladder of incentives for lower bandwidth people to move up to better services.
For those customers who aren't corporate/geek enough to make heavy use of LAN-dwelling protocols and capabilities, and who aren't motivated to hoist the jolly roger and set sail for the bay, there is very much a reciprocal relationship between available bandwidth and available services. Not much incentive to improve bandwidth if nobody can buy services to use it on, and not much incentive to start services if nobody has the bandwidth needed to become a customer...
As a competitive force, that falls into the "Oh, you math geeks are so adorable. Why don't you go off and play in the ISM band where the big kids' table doesn't have to look at you..." category.
So, guys... how's that whole "Let the market decide" argument working out for you? Capitalism works great for non-critical, non-infrastructure goods and services... but when it gets its hands on something everybody needs, it's gonna take you to the cleaners. Every single time.
It's not so much the criticality(or lack thereof) that gets you or saves you; but the barriers to entry and costs of duplicate entrants.
Plenty of user-critical-without-which-modern-civilization-would-come-to-a-screeching-halt goods get produced just fine under market conditions, so long as the barriers to entry aren't too high and competition between suppliers of identical or reasonably substitutable goods is fairly robust.
Infrastructure, of course, is sort of a classic case of high barriers to entry coupled with high costs of duplicate entrants. You've got a finite amount of desirable spectrum to work with, and every market entrant needs to put a net-connected box of RF gear on a pole in every chunk of land they want to provide service for. The spectrum and buildout hassles tend to leave you with an oligopoly, at best, among people who actually provide wireless bandwidth, and possibly some MVOs who provide specialized customer support/billing/advertising offerings to sell bandwidth to subscriber groups that the primary telcos don't want to deal with.
In addition to having strong 'natural monopoly' characteristics, Infrastructure goods and services bite you twice: Not only are they natural monopolies/oligopolies, and thus tend to fall into ugly pricing, they are infrastructure, something over which other goods and services and interactions take place, so they are extremely well placed to allow holders of market power in the infrastructure market to exert it in any of the markets (or socially valuable nonmarket activities like 'communication') that occur on top of that infrastructure. Wacky fun!
" 'It's evolving literally to be a single platform,' he said, referring to how Windows 8 and Windows Phone 8 share a kernel, file system, graphics support, and other elements. At least in theory, that will allow developers to port apps from the desktop/tablet OS to the smartphone OS with relatively little work."
Hasn't Gates been chasing the dream of one Windows to rule them all for something like two decades now? The line of 'Handheld PC' and 'Pocket PC' devices didn't share as much low level architecture, because the hardware wouldn't permit it at the time; but did everything they could to drag a desktop UI onto a teeny touchscreen, and 'tablet' meant getting Windows for Pen Computing 1.0 with your Win3.1 back when meteorites were still mopping up the last of the dinosaurs....
The other issue with the "give it to a 3 year old for 10 minutes" test, in addition to the 'unlearning' piece, is that it only measures the slope of the first tiny bit of the learning curve. What people who actually use computers(or who have to do actual work with them) really care about is whether the interface makes them more powerful and efficient in exchange for a modicum of skill.
In the specific case of natural languages, most of us are...
Does someone with your sig ever need to scroll far to find a Snow Crash reference?
The power supply for a 40 watt gas laser will give you a bit of a tickle, as well, if you fuck up... Fire is the least of your problems here.
When did the notion start to circulate that anything remotely related to wood is some kind of incendiary deathtrap? Is it when people stopped having to start fires with nothing more than minimal tools and careful arrangement of sticks?
Christ, you've got something designed to cut through plastics with a laser, plastics which are basically just waiting for some added heat to turn into sticky, flaming, hydrocarbon death, and nobody says a thing. Suddenly, terrifying wood,. notorious for perfunctory smoldering in response to heat, bursts onto the scene and everybody is freaking out about ignition. Kids these days.
Somehow, people have been practicing pyrography for millenia without bursting into flames.
How would you suggest they solve the funding dilemma?
The details are a nighmarish, ever-shifting, morass of acronyms and "talk to your rep"; but Microsoft already has a plan where customers can pay annually for "Software Assurance".
Obviously, Microsoft wants to be bug-hunting as few codebases as possible, and presenting as unified a "platform" to 3rd party application vendors as possible, so it is to be expected that the cost/seat of continued support would rise over time as the number of seats in the field dropped; but they already have a mechanism for charging customers on an annual volume basis, which could easily enough be extended to offering paid security support after the support period provided by a boxed license is over...
Not necessarily something that they want to do; but they could.
We'll see if they start playing any games with WHQL/driver signing. If they don't, it's the hardware vendors' problem. If it turns out that "Windows XP driver Exists" is an unofficial ticket to substantial additional delays in getting the WHQL signature for the Win7/Win8 driver... Well, that'll be a different story.
What kind of arm-twisting, exactly, is MS planning against Dell, HP, etc. to get them to stop shipping boring corporate boxes that don't support XP?
Yeah, sure, the odds of having XP run properly without a bit of scrounging on some random machine from Best Buy(this goes double if it's a laptop, triple if it's some wacky touch/hybrid/thing), aren't getting any better; but if your business is shipping pallet-loads of identical machines to assorted volume customers, you damn well better support the OSes they want supported. If you don't, the largely interchangeable shipper of near-identical machines will.
Even if MS plays serious hardball, and just starts refusing to WHQL sign XP drivers, XP doesn't force driver signing very hard, so IT shouldn't have much trouble with that. Now, I'd be totally unsurprised to learn that XP toasts the battery life of newer laptops with super-fancy power saving features, or requires that you turn on the 'legacy bios emulation' switch in whatever UEFI pit the system ships with; but I'd be shocked to see the end of the ability to buy XP boxes(through corporate and volume license channels, not necessarily at retail) before 2020...
Pictures
Hideous Contraption 1
Head Crab 1
Heavy Crap 1
Horrifying Cranium 1
Headborn Casheater 1?
Wasn't there an entire section in Snow Crash about 'Gargoyles' who had embraced wearable computing in a big way, and how everybody hated them and thought that they were freaks?
I think that having a local MTA is still pretty normal on Linux(debian certainly gave me the rudiments of an exim4 install the last time I reinstalled it); but it seems that, even there, the MTA is largely cut out of the loop. It still shuttles system emails around, mostly warnings and such sent to root; but if you install Thunderbird or Kmail or whatnot, those packages completely ignore exim unless you know what you are doing and make the appropriate changes to both the exim configuration and the client config. The interface defaults on the mail client packages are very much aimed at getting you connected directly to your IMAP/POP3 account of choice, and don't even mention the local MTA.
The AIX on RS/6000 system I used to telnet into to check my mail in pine certainly took its local MTA more seriously; but I think that it was run by traditionalists...
In any event, it will be interesting to see how this school district manages.
What I want to know is where the http://pixelqi.com/ guys are hiding... They had a workable device, shipping in nontrivial volume with the OLPC XO-1, and then seemingly dropped off the map.
All the refresh rate of an LCD panel(because it is one); but, in transreflective mode, looks more like e-ink than any LCD I've ever seen and has the option to do color if you crank the backlight....
We know(because all but the nastiest LCD tablets running Android or iOS can and do do it) that contemporary low-power ARM chipsets are up to the challenge of crunching PDFs; but e-ink displays are mostly too small to display 8.5x11 or A4 pages, too slow for panning/zooming/etc, and PDF reflow is crap. If they would just start existing, the Pixel Qi screens would fairly efficiently solve this problem, at lower cost and lower power than standard LCD panels; but nobody seems to have heard a peep from them.
Thankfully, it is axiomatic that Stallman Is An Extremist, so we needn't listen to his(often strident, as often correct) warnings!
The awesome thing about the emerging DRM economy is that it combines the economic relations of feudalism with the efficient, data-driven surveillance that East Germany was too low-tech to achieve...
Interesting. I have one of the little RTL2832/E4000 dongles plugged in right now, and I'm familiar by reputation with the USRP stuff; but my knowledge of the midrange, and any of the gear that has evolved from the hobbyist radio side, or that is designed for crunching RF down to something that a soundcard works with, is pretty much zero.
How is an English-only voice processor going to help you comprehend the strange utterances of southerners?
Just use the magic word "H1B" and all that changes... The disappearance of other people's jobs is just the inevitable march of progress, and will probably make them better off in the long run anyway. My job, though, now that's a different story...
So, just as mean and feral; but without the excuse of youthful stupidity...