I think flamebait is probably more appropriate, but nobody uses that mod much. Your posts are poorly thought out, logically inconsistent, and antagonistic. They also appear to be what you honestly personally believe, which says some very sad things about you. Nonetheless, they detract from, rather than adding to, the discussion at hand; consequently, they get modded down.
You're not a libertarian. Those have some vestige of rationality and at least a vague grasp on certain aspects of the real world. You're an anarchist who finds it convenient to use the label "libertarian" because it sounds much better and some people can't tell them apart.
Pro-tip: the part that gave you away was the following line: "The state is not a scientific institution. It is the antithesis of science -- it is merely the organized used of force to dominate a population. The government in all its forms is merely a manifestation of the threat or actual use of violence". I just read the rest of your posts before responding.
The state is an institution of leadership, instituted by any and every non-trivial-sized human community ever. The difference between a tribe led by family elders and the government of a superpower is mostly one of scale and environment, not of purpose, method of coming into existence, or means of sustaining itself.
One of the roles of a leader is arbitrator. Another is defender. These are the ones you seem to think are all that matters, and I bet you reject the arbitrator and would like to think you could get by without the defender (hint: just because you could get by without some of the things they do does not mean you would make it without them at all). Other roles of a leader: guidance (which includes the promotion of science), provider (public services to avoid tragedy of the commons, public funding to ensure that things like science actually happen, etc.), caretaker (seeing to the well-being of your people - welfare being a part of that - and ensuring they can get what they need), and ambassador (interfacing with other states, other leaders). Every single one of those things requires wealth (although they either produce more wealth than they consume - yes, even welfare, go read some real economic and sociological studies - or they are necessary for the protection of the value of peoples' wealth). Therefore, the state must fund itself and - again, to avoid the tragedy of the commons - this funding is mandatory.
Note that I make no attempt to claim that every state actually embodies these leadership characteristics. For example, corruption and accumulation of personal wealth is contrary to the roles of caretaker and provider. There is probably no non-trivial-sized (for modern values of non-trivial; more than a few dozen people is unlikely, more than a few thousand impossible) example of an ideal state in existence anywhere today, or in modern history. The USA certainly isn't, and never was; there were deep flaws in the original government, many (but not all) of which have been fixed but replaced with new ones.
NASA is not the state doing the right thing, overall, by any means. However, it is a part of what doing the right thing would be. It is giving a many-times-over growth in wealth on an investment that individuals typically do not choose to make, and also returns other benefits leading to improved military (defender) and higher standard of living (caretaker).
There's nothing missing about the "preview" of Office RT. Office 2012 wasn't finished when RT went RTM, so the OS ships with a pre-release copy (which works fine, incidentally).
However, within about a month of the device's retail release, Office RT was also released, and the full upgrade is an automatic install over Windows Update. The reason you don't see anything about the "preview" of Office RT is because, if for some reason the device you buy is still running the initial RT image, that preview will be replaced by a working version as soon Windows Update runs.
Fine then, I'll echo the GP AC's comments, with an additional caveat:
s/"Surface RT"/"jailbroken Surface RT"/g
The ability to run PuTTY, and gvim and 7-Zip and python and perl and such natively. The ability to take any.NET 4.x app, like the recent versions of Fiddler or IKVM (for Java support), and run it. The ability to (slowly, but it works) emulate x86 so I can run some non-demanding closed-source software that I like.
*THAT* makes the Surface RT an awesome device, when combined with everything else. Without that, it's merely a nice device running a badly (but fixably!) flawed operating system.
For what it's worth, the Touch Cover is actually stiff enough to use in your lap, never mind the Type Cover (the one with actual keys). The tablet will also happily support itself with its kickstand on my knees. With that said, the angle that the screen is at (determined by the kickstand) is intended for use on a desk and is not optimal on your lap.
I personally think that the second point was their biggest fuck-up. It's dead easy to recompile most desktop software for RT. Microsoft even provides the tools, although the public toolchain needs a little convincing before it will let you use them that way (changing a single line in an XML configuration file). Even the lack of certain features, such as OpenGL, can be worked around; gldirect works for wrapping OpenGL calls as DirectX functions, and is itself portable to RT.
However, you can't *run* those programs, per Microsoft's say-so. Never mind that a "jailbreak" appeared shortly after release. Never mind that it makes the OS tremendously more useful. Microsoft still tries to lock it down, rather than embracing the market for an ARM device. Here's what I would have done: 1) First and foremost: no lockdown. Limit it to Microsoft-signed binaries by *default* if they feel it's needed but allow users to unlock that restriction. Hide it somewhere, like a bcdedit option, if they really want to. Similarly, don't lock the bootloader; use SecureBoot, but allow it to be disabled and/or allow other keys to be installed, permitting the use of non-MS operating systems. We already paid for the hardware and software it came with, right? 2) Push the use of.NET (which doesn't need recompiling to switch from x86 to ARM). Possibly even include legacy.NET versions (RT only comes with.NET 4.x), just to get the maximum app compatibility. Strongly encourage independent software vendors for Windows to publish.NET apps compiled for "AnyCPU" to get maximum compatibility. This should have been done prior to RT's launch, but they could still give it a go now. 3) For devs who don't use.NET, or for the owners of legacy codebases, convince them to recompile their software for ARM. Make it really easy; no restrictions in Visual Studio, all the library files included with the SDK (so we don't have to cut them out of system DLLs), as many pre-compiled and built-in libraries as they can convince themselves to supply. Again, this should have been done long ago. 4) Create and push architecture-independent installers. It should be possible to use.MSI files as "fat installers" where the arch-specific version of the program appropriate to your hardware gets installed. Tiny.NET-based installer programs could pull the correct native binaries over the Internet and install them. Wherever possible, don't make the user choose what architecture to install; just let them go on installing programs the way Windows users always have. 5) Market desktop apps (for RT; they already do this for x86 and x64) through the Windows Store. Make it really easy for users to find the apps they want, and at the same time get people used to using the Store. If they decided to try the lockdown-by-default route, include dire warnings about how desktop apps are inherently less secure than WinRT (Modern / Metro) ones because WinRT apps run in a sandbox, but let people enable the installing of such apps through the store. 6) Get BlueStacks, or somebody like them, to provide an Android compatibility layer for RT. It doesn't have to be amazing, but make it *good enough* that Android apps can be effectively considered to be available (preferably by default, or with minimal effort) on RT. Don't put them front and center, but *do* include it in your feature bullet points ("can run software from the Windows Store, plus Android applications") and "how many apps" lists! 7) Encourage hardware vendors to build and certify RT drivers, then distribute them via Windows Update (as is done o
That's a pretty stupid argument. I wouldn't put an SD card in a desktop for storage, but I would certainly use one for transferring files to a mobile device... except, oh wait, he mobile device doesn't have a card slot. And I would gladly put a second HDD in a desktop for storage, which is equivalent (in terms of expected storage used on the device) to putting a microSD into a phone for storage.
128GB sounds like a lot for a phone now - my current phone only has 80GB, and that's counting the uSD card - but in a couple years it won't be so impressive. On the other hand, in a couple years, 128GB microSDXC cards will be affordable (not cheap, but in line with expected price/GB); just one of those would bring the phone to a cool 256GB. I already have *more* than 256GB of data (recorded TV alone, never mind other things like music and apps and synching all my email instead of just the recent bit) which I'd like to put on my phone rather than trying to stream it over the cellular link or even over WiFi (which is often actually slower than a good LTE connection).
The TMo plan you describe is actually post-pay, it's just contract-free and month-to-month. Pre-pay means you pay up front for a certain amount (of talk/text/data, each of which usually spends your pre-paid amount at some rate, such as $0.10/SMS), and if you run out you have to either top up or go without.
Prepaid is the cheapest option if you really don't use your phone much, and are way below the thresholds of even the cheapest monthly plans on most months. For the rest of us, no-contract monthly plans (which TMo specializes in) are the way to go.
By the way, TMo now also offers a $50/month contract-free plan that gives unlimited talk, unlimited text, 500MB of unthrottled data and/or tethering, and unlimited untethered (but *possibly* throttled) data past then. Extra unthrottled or tethering data can be added at $10/2GB/month.
Those are all post-paid plans though; you get a bill in the mail on your billing date, and then your limits reset.
If you spend over $300 on either the CPU or the GPU, you're wasting money for gaming. There's just no point, when I can already run WQHD at max specs and still hit 60FPS.
Yes, my desktop will be obsolete before yours, but that doesn't matter; yours won't last enough longer than mine (it's a toss up whether a system like that will suffer part failure before going obsolete, too) to make up the cost. I'll just upgrade the components. A $300 GPU that runs all the games today just fine will be paired with a (by then) $200GPU when games catch up to it a few years down the line, or possibly replaced with a new $300 one to save on power usage.
I'm unclear if you're an idiot, an elitist, or a troll (not that those are mutually exclusive) but if what you say is true then you're throwing money away. Lots of it.
No, that's about right. I only ever got throttled once or twice on my old "unlimited" plan, and I consistently use about 1-3GB (hooray streaming music).
Something people here haven't been mentioning: if you pay an extra $20/month, you get "true" unlimited data - no throttling at all.
Also, your high-speed data cap also counts as a tethered data cap. So normally you get up to 500MB tethering/month included. If you spend the extra $10 per 2GB extra high-speed data, you also get an extra 2GB of tethered data (per $10). The $20 for unlimited high-speed is a flat-rate deal that doesn't apply to the tethering cap, though.
It's still an excellent deal, and they are not, generally, complete dicks about things.
Agreed. I'm seriously confused as to how the US has managed to go to hell so rapidly and completely on this front. Did Sep 11 really scare the shit out of people *that* badly? I mean, yeah, it was awful. Yeah, it was a huge shock. Yeah, people demanded ACTION in response.
But that was almost 12 years ago. People who were six years old at the time, and quite possibly don't even remember it, are now voting age. Even with a significant portion of the government (including that two-faced bastard of a president) scaremongering to the best of their ability, you'd think enough people in this country would have gotten sick of this bullshit to put an end to it by now, and worked to get the country back to something which vaguely approximates the principles which it claims to uphold...
On a snarkier note, by and large Russians aren't such awful people. I don't really think they deserve our government either (not that their own is anything to be proud of). How about we offer them a nice month-long cruise along the coast of Somalia (after de-funding the anti-piracy patrols there) instead?
This is a solid explanation of how the phone could be safely plugged in, but very dangerous once it's brought to your head. A cable with a resistance of 0.1 Ohms carrying a current of 1 Amp will experience a voltage drop of 0.1V, which means it will dissipate 0.1 Watts (I * dV) - a meager amount that the wire can maintain indefinitely (and indeed, is designed to). This happens regardless of whether the wire is carrying 5V or 225V; all that matters is the voltage drop *along* the wire (and the current on the wire, which is partially a function of that voltage drop).
Now, if you shorted the wire itself (rather than shorting the mains to one side of the wire), then (under normal circumstances) your wire is now dropping the full 5V that is the difference between its positive and ground terminals. To get a voltage drop of 5V with only 0.1 Ohms of resistance, you need 50 Amps (dV = I * R -> dV/R = I). 50A current times 5V potential drop is 250 Watts of dissipated heat, which would destroy your wire if the USB port could supply anywhere near that much current. The numbers for 220V are even more frightening (P = dV^2/R, so that would be 484000W). But, that won't happen unless the wire is shorted to itself. If the wire (at 220V) is instead connected to ground through a human being (with a resistance between that person's head and ground of, say, 49.9 Ohms) then you've got (220)^2 / 50, which is a mere 0.2% of the number given above. That's still about a kilowatt, enough to kill you very quickly.
As for the wire, it's now carrying 4.4A (220V/50 Ohm), but it still only has 0.1 Ohm of resistance and is therefore dissipating a mere 1.936W across a voltage drop of.44V. That's many times its normal dissipation, but still it might not even be enough to melt the plastic coating.
First? Remember, no current is flowing until the circuit is completed. The amount of current required to flash-ignite the charging wire is quite high compared to the amount necessary to cook human nervous tissue. Even if the wire burned through its insulation almost immediately, it could already be too late.
Also, remember that USB cables actually have moderately high current tolerances. Carrying any meaningful amount of power at 5V requires a relatively high amperage rating. 1A at 220V is quite a lot of power when applied to the human body, and that's merely what these charging cables are expected to carry in normal operation. In practice, they can probably manage several times that without anything worse than becoming a bit warm will occur. At 2.3A, that's over half a kilowatt (minus minor resistive losses in the wire, which will heat it but probably not melt/ignite it) being delivered to the head...
You don't even need a password change. You just store in the database what password verifier scheme was used (and a pair of MD5s with a salt of unknown size is a damn weak one) and then when the user logs in, you derive the password verifier using the scheme stored in the DB for their account. If it matches, then you log them in (of course) but you *also* computer a new password verifier using the new, better scheme - say, PBKDF2 with 50000 iterations - and then store that new verifier, and the new scheme you used to derive it, to the user's account entry in the DB. It's completely transparent to the user. As a bonus, this makes it easy to adopt even stronger schemes in the future.
You know, for somebody who *sounds* like you're anti-DRM, that sure isn't reflected in your actions. If you want MS (and the other big players; they all watch each other) to reduce DRM in the future, reward them for *not* locking down the console like they originally planned to do. They listened to customer feedback. They responded. They made the right choice... and for that, you propose to punish them? Wow, that's a great plan right there. Which message do you think will be taken away from that?
1) Well, looks like nobody actually cared whether there was DRM or not in the first place. No need to listen to those whiney kids in the future! 2) Wow, sounds like people on the Internet are just petulant little children who will object *whatever* we do! May as well increase DRM in the future, at least the publishing industry likes that...
I'll give you a hint: neither one makes the attitude you're currently expressing look good, but neither one gets you what you want, either.
Oh, and out of curiosity, what other platforms do you develop for? Because there's only one that in the last few years has retroactively removed advertised features from its console in the name of more DRM, and then sued the people who worked around it. It's not Microsoft, either.
LOL! Remove or spoof the Referrer header (they only check the domain, don't bother changing the path) and it works. That's some awfully idiotic work right there.
It also makes it more appealing to actually copy the images and host them locally, rather than linking them directly. Yeah, that costs you a little, but with your link I at least knew where it came from... with a copied image that has the douchebag company's name cropped out, they get nothing.
Illegal, of course, unless the image was not actually under their copyright... but better than giving them money.
To be fair, part of the reason for the "sold only in our stores" thing was to ensure that the customers knew what they were buying. I have a Surface RT that I bought for research into the OS, and the guy at the Microsoft store made very sure I knew it wouldn't run legacy Windows programs.
He was wrong, after quite a bit of hacking (jailbreak + emulation layer = old games and undemanding desktop software work just fine), but that's beside the point.
This is actually already being worked on. The current version of the "jailbreak" hack for RT allows third-party kernel-mode drivers. The suggestion to write a driver which then dumps NT and bootstraps the Linux kernel has been discussed at length. It's not a small project, by any means, but it should be possible. Ideally, we would create a second bootloader entry (we do have limited control over the bootloader; it just always must execute a Microsoft-signed kernel) which would load RT, then jailbreak and load the driver, which would immediately load Linux. Alternatively, we could pretty easily have a "switch to Linux" shortcut on the desktop which (assuming jailbreak) would load that driver.
However, the whole process will take a fair bit of work. Also, MS is trying to prevent such "jailbreaks" in RT 8.1. Currently, downgrading back to 8.0 is still possible, at which point the jailbreak works again, though.
Sorry to self-reply, but in case it isn't obvious from the previous post: Microsoft could "fix" RT with a single, simple update. Reboot the tablet and the restrictions are gone.
An official x86 compat layer would be a fair bit of work, of course, but it's not really necessary to do that; the simple ability to run.NET apps (and maybe they get a few of their more important partners to flip the Platform option in Visual Studio to "ARM" and hit Build again; often it really is that simple) would make RT a lot more appealing.
There's no "port" involved. Or rather, they already did that. Then they added literally one configuration change to lock out non-MS-signed desktop apps. One change. It's a single flag in the kernel. On x86 and x64 builds of Windows NT, it's not set. On ARM builds of Windows NT (RT and WP8), it is.
Clear that flag (which is what the current "jailbreak" hack for RT does), and you can run any desktop software that will compile for ARM, or any.NET program, or any other language that can be run through one of the others (for example, Java is possible through IKVM, a.NET program implementing a JVM).
Now, as for domain joining, that's actually a simpler problem. All versions of Windows NT have had multiple SKUs (editions) ranging from the do-anything highest-end Server builds to the very crippled Starter builds. It's all the same codebase, just a configuration change. RT falls somewhere between Win8[Home] and Win8Pro SKUs in terms of business-y features; it can use BitLocker encryption (usually not available on Home) but cannot join domains (usually available on anything *except* Home).
Working around that particular restriction is also possible, though it is not easy unless you also remove the signature enforcement ("jailbreak") at which point it becomes nearly trivial.
Oh, and there's already a (very early and still incomplete) x86 emulation layer (actually, dynamic recompilation) for "jailbroken" RT devices. It's slow, as one would expect, but it can run old games and desktop software just fine. It also is the work of a single homebrew developer working from public documentation and reverse engineering for the Windows interoperability (calls to system libraries are thunked to ARM code, which is both faster than using x86 libraries and requires less install space). Microsoft could do a better job easily by putting a few of their people who previously worked in that space (for example, the "Virtual PC for Mac" software worked the same way, some of them are probably still around) on the job.
It's funny how many people, even here and now, still conveniently forget what iPads cost when bashing competing products. Your statement "They were dramatically more expensive than most competing tablets" is a bald-faced lie. Surface RT has always been competitive with iPads and usually a bit cheaper than the newest model.
To be fair, sideloading (for "Metro") is present and free, though it's a little better hidden than on Android. The earlier versions of that "Jailbreak" you refer to required it, in fact.
Powershell command: Show-WindowsDeveloperLicenseRegistration (yes, it supports Tab-completion.)
Xbox, for all that it took a ton of money to get it where it is, is now competing pretty well (we'll see how the next generation goes...). Azure seems to be doing fine even if it's small potatoes next to AWS. Same for SQL Server, a bit better than Azure even. Windows Server (unlike Client) is nowhere near a monopoly in its market, but is also doing well. In the pre-iPhone days, Windows Mobile was one of the more popular smartphone platforms without ever being nearly a monopoly.
To be fair, there's a "jailbreak" that enables all of that..NET apps do run, unmodified and un-recompiled. Open source Win32 apps are being / have been ported. There are a few specifically-for-RT native apps as well, including an x86 dynamic recompilation layer that allows running (some) x86 apps directly.
This is all the work of a few volunteer devs on the XDA Developers forum, working for free on their own time. These efforts were made without MS support, just by public documentation and reverse engineering. Nonetheless, it makes the tablets (all RT devices) much, much more valuable to the (many) people who aren't willing to stick to "Modern" apps. MS could have done far, far better themselves in terms of a slick experience (starting with making the "jailbreak" be an option somewhere, instead of an Admin [which users have by default] to kernel [which is supposed to be inviolable] exploit that is scripted to run at bootup).
Slashdot Mirror
I think flamebait is probably more appropriate, but nobody uses that mod much. Your posts are poorly thought out, logically inconsistent, and antagonistic. They also appear to be what you honestly personally believe, which says some very sad things about you. Nonetheless, they detract from, rather than adding to, the discussion at hand; consequently, they get modded down.
You're not a libertarian. Those have some vestige of rationality and at least a vague grasp on certain aspects of the real world. You're an anarchist who finds it convenient to use the label "libertarian" because it sounds much better and some people can't tell them apart.
Pro-tip: the part that gave you away was the following line: "The state is not a scientific institution. It is the antithesis of science -- it is merely the organized used of force to dominate a population. The government in all its forms is merely a manifestation of the threat or actual use of violence". I just read the rest of your posts before responding.
The state is an institution of leadership, instituted by any and every non-trivial-sized human community ever. The difference between a tribe led by family elders and the government of a superpower is mostly one of scale and environment, not of purpose, method of coming into existence, or means of sustaining itself.
One of the roles of a leader is arbitrator. Another is defender. These are the ones you seem to think are all that matters, and I bet you reject the arbitrator and would like to think you could get by without the defender (hint: just because you could get by without some of the things they do does not mean you would make it without them at all). Other roles of a leader: guidance (which includes the promotion of science), provider (public services to avoid tragedy of the commons, public funding to ensure that things like science actually happen, etc.), caretaker (seeing to the well-being of your people - welfare being a part of that - and ensuring they can get what they need), and ambassador (interfacing with other states, other leaders). Every single one of those things requires wealth (although they either produce more wealth than they consume - yes, even welfare, go read some real economic and sociological studies - or they are necessary for the protection of the value of peoples' wealth). Therefore, the state must fund itself and - again, to avoid the tragedy of the commons - this funding is mandatory.
Note that I make no attempt to claim that every state actually embodies these leadership characteristics. For example, corruption and accumulation of personal wealth is contrary to the roles of caretaker and provider. There is probably no non-trivial-sized (for modern values of non-trivial; more than a few dozen people is unlikely, more than a few thousand impossible) example of an ideal state in existence anywhere today, or in modern history. The USA certainly isn't, and never was; there were deep flaws in the original government, many (but not all) of which have been fixed but replaced with new ones.
NASA is not the state doing the right thing, overall, by any means. However, it is a part of what doing the right thing would be. It is giving a many-times-over growth in wealth on an investment that individuals typically do not choose to make, and also returns other benefits leading to improved military (defender) and higher standard of living (caretaker).
There's nothing missing about the "preview" of Office RT. Office 2012 wasn't finished when RT went RTM, so the OS ships with a pre-release copy (which works fine, incidentally).
However, within about a month of the device's retail release, Office RT was also released, and the full upgrade is an automatic install over Windows Update. The reason you don't see anything about the "preview" of Office RT is because, if for some reason the device you buy is still running the initial RT image, that preview will be replaced by a working version as soon Windows Update runs.
Fine then, I'll echo the GP AC's comments, with an additional caveat:
s/"Surface RT"/"jailbroken Surface RT"/g
The ability to run PuTTY, and gvim and 7-Zip and python and perl and such natively. The ability to take any .NET 4.x app, like the recent versions of Fiddler or IKVM (for Java support), and run it. The ability to (slowly, but it works) emulate x86 so I can run some non-demanding closed-source software that I like.
*THAT* makes the Surface RT an awesome device, when combined with everything else. Without that, it's merely a nice device running a badly (but fixably!) flawed operating system.
For what it's worth, the Touch Cover is actually stiff enough to use in your lap, never mind the Type Cover (the one with actual keys). The tablet will also happily support itself with its kickstand on my knees. With that said, the angle that the screen is at (determined by the kickstand) is intended for use on a desk and is not optimal on your lap.
I personally think that the second point was their biggest fuck-up. It's dead easy to recompile most desktop software for RT. Microsoft even provides the tools, although the public toolchain needs a little convincing before it will let you use them that way (changing a single line in an XML configuration file). Even the lack of certain features, such as OpenGL, can be worked around; gldirect works for wrapping OpenGL calls as DirectX functions, and is itself portable to RT.
However, you can't *run* those programs, per Microsoft's say-so. Never mind that a "jailbreak" appeared shortly after release. Never mind that it makes the OS tremendously more useful. Microsoft still tries to lock it down, rather than embracing the market for an ARM device. Here's what I would have done: .NET (which doesn't need recompiling to switch from x86 to ARM). Possibly even include legacy .NET versions (RT only comes with .NET 4.x), just to get the maximum app compatibility. Strongly encourage independent software vendors for Windows to publish .NET apps compiled for "AnyCPU" to get maximum compatibility. This should have been done prior to RT's launch, but they could still give it a go now. .NET, or for the owners of legacy codebases, convince them to recompile their software for ARM. Make it really easy; no restrictions in Visual Studio, all the library files included with the SDK (so we don't have to cut them out of system DLLs), as many pre-compiled and built-in libraries as they can convince themselves to supply. Again, this should have been done long ago. .MSI files as "fat installers" where the arch-specific version of the program appropriate to your hardware gets installed. Tiny .NET-based installer programs could pull the correct native binaries over the Internet and install them. Wherever possible, don't make the user choose what architecture to install; just let them go on installing programs the way Windows users always have.
1) First and foremost: no lockdown. Limit it to Microsoft-signed binaries by *default* if they feel it's needed but allow users to unlock that restriction. Hide it somewhere, like a bcdedit option, if they really want to. Similarly, don't lock the bootloader; use SecureBoot, but allow it to be disabled and/or allow other keys to be installed, permitting the use of non-MS operating systems. We already paid for the hardware and software it came with, right?
2) Push the use of
3) For devs who don't use
4) Create and push architecture-independent installers. It should be possible to use
5) Market desktop apps (for RT; they already do this for x86 and x64) through the Windows Store. Make it really easy for users to find the apps they want, and at the same time get people used to using the Store. If they decided to try the lockdown-by-default route, include dire warnings about how desktop apps are inherently less secure than WinRT (Modern / Metro) ones because WinRT apps run in a sandbox, but let people enable the installing of such apps through the store.
6) Get BlueStacks, or somebody like them, to provide an Android compatibility layer for RT. It doesn't have to be amazing, but make it *good enough* that Android apps can be effectively considered to be available (preferably by default, or with minimal effort) on RT. Don't put them front and center, but *do* include it in your feature bullet points ("can run software from the Windows Store, plus Android applications") and "how many apps" lists!
7) Encourage hardware vendors to build and certify RT drivers, then distribute them via Windows Update (as is done o
That's a pretty stupid argument. I wouldn't put an SD card in a desktop for storage, but I would certainly use one for transferring files to a mobile device... except, oh wait, he mobile device doesn't have a card slot. And I would gladly put a second HDD in a desktop for storage, which is equivalent (in terms of expected storage used on the device) to putting a microSD into a phone for storage.
128GB sounds like a lot for a phone now - my current phone only has 80GB, and that's counting the uSD card - but in a couple years it won't be so impressive. On the other hand, in a couple years, 128GB microSDXC cards will be affordable (not cheap, but in line with expected price/GB); just one of those would bring the phone to a cool 256GB. I already have *more* than 256GB of data (recorded TV alone, never mind other things like music and apps and synching all my email instead of just the recent bit) which I'd like to put on my phone rather than trying to stream it over the cellular link or even over WiFi (which is often actually slower than a good LTE connection).
The TMo plan you describe is actually post-pay, it's just contract-free and month-to-month. Pre-pay means you pay up front for a certain amount (of talk/text/data, each of which usually spends your pre-paid amount at some rate, such as $0.10/SMS), and if you run out you have to either top up or go without.
Prepaid is the cheapest option if you really don't use your phone much, and are way below the thresholds of even the cheapest monthly plans on most months. For the rest of us, no-contract monthly plans (which TMo specializes in) are the way to go.
By the way, TMo now also offers a $50/month contract-free plan that gives unlimited talk, unlimited text, 500MB of unthrottled data and/or tethering, and unlimited untethered (but *possibly* throttled) data past then. Extra unthrottled or tethering data can be added at $10/2GB/month.
Those are all post-paid plans though; you get a bill in the mail on your billing date, and then your limits reset.
If you spend over $300 on either the CPU or the GPU, you're wasting money for gaming. There's just no point, when I can already run WQHD at max specs and still hit 60FPS.
Yes, my desktop will be obsolete before yours, but that doesn't matter; yours won't last enough longer than mine (it's a toss up whether a system like that will suffer part failure before going obsolete, too) to make up the cost. I'll just upgrade the components. A $300 GPU that runs all the games today just fine will be paired with a (by then) $200GPU when games catch up to it a few years down the line, or possibly replaced with a new $300 one to save on power usage.
I'm unclear if you're an idiot, an elitist, or a troll (not that those are mutually exclusive) but if what you say is true then you're throwing money away. Lots of it.
No, that's about right. I only ever got throttled once or twice on my old "unlimited" plan, and I consistently use about 1-3GB (hooray streaming music).
Something people here haven't been mentioning: if you pay an extra $20/month, you get "true" unlimited data - no throttling at all.
Also, your high-speed data cap also counts as a tethered data cap. So normally you get up to 500MB tethering/month included. If you spend the extra $10 per 2GB extra high-speed data, you also get an extra 2GB of tethered data (per $10). The $20 for unlimited high-speed is a flat-rate deal that doesn't apply to the tethering cap, though.
It's still an excellent deal, and they are not, generally, complete dicks about things.
Agreed. I'm seriously confused as to how the US has managed to go to hell so rapidly and completely on this front. Did Sep 11 really scare the shit out of people *that* badly? I mean, yeah, it was awful. Yeah, it was a huge shock. Yeah, people demanded ACTION in response.
But that was almost 12 years ago. People who were six years old at the time, and quite possibly don't even remember it, are now voting age. Even with a significant portion of the government (including that two-faced bastard of a president) scaremongering to the best of their ability, you'd think enough people in this country would have gotten sick of this bullshit to put an end to it by now, and worked to get the country back to something which vaguely approximates the principles which it claims to uphold...
On a snarkier note, by and large Russians aren't such awful people. I don't really think they deserve our government either (not that their own is anything to be proud of). How about we offer them a nice month-long cruise along the coast of Somalia (after de-funding the anti-piracy patrols there) instead?
This is a solid explanation of how the phone could be safely plugged in, but very dangerous once it's brought to your head. A cable with a resistance of 0.1 Ohms carrying a current of 1 Amp will experience a voltage drop of 0.1V, which means it will dissipate 0.1 Watts (I * dV) - a meager amount that the wire can maintain indefinitely (and indeed, is designed to). This happens regardless of whether the wire is carrying 5V or 225V; all that matters is the voltage drop *along* the wire (and the current on the wire, which is partially a function of that voltage drop).
Now, if you shorted the wire itself (rather than shorting the mains to one side of the wire), then (under normal circumstances) your wire is now dropping the full 5V that is the difference between its positive and ground terminals. To get a voltage drop of 5V with only 0.1 Ohms of resistance, you need 50 Amps (dV = I * R -> dV/R = I). 50A current times 5V potential drop is 250 Watts of dissipated heat, which would destroy your wire if the USB port could supply anywhere near that much current. The numbers for 220V are even more frightening (P = dV^2/R, so that would be 484000W). But, that won't happen unless the wire is shorted to itself. If the wire (at 220V) is instead connected to ground through a human being (with a resistance between that person's head and ground of, say, 49.9 Ohms) then you've got (220)^2 / 50, which is a mere 0.2% of the number given above. That's still about a kilowatt, enough to kill you very quickly.
As for the wire, it's now carrying 4.4A (220V/50 Ohm), but it still only has 0.1 Ohm of resistance and is therefore dissipating a mere 1.936W across a voltage drop of .44V. That's many times its normal dissipation, but still it might not even be enough to melt the plastic coating.
First? Remember, no current is flowing until the circuit is completed. The amount of current required to flash-ignite the charging wire is quite high compared to the amount necessary to cook human nervous tissue. Even if the wire burned through its insulation almost immediately, it could already be too late.
Also, remember that USB cables actually have moderately high current tolerances. Carrying any meaningful amount of power at 5V requires a relatively high amperage rating. 1A at 220V is quite a lot of power when applied to the human body, and that's merely what these charging cables are expected to carry in normal operation. In practice, they can probably manage several times that without anything worse than becoming a bit warm will occur. At 2.3A, that's over half a kilowatt (minus minor resistive losses in the wire, which will heat it but probably not melt/ignite it) being delivered to the head...
You don't even need a password change. You just store in the database what password verifier scheme was used (and a pair of MD5s with a salt of unknown size is a damn weak one) and then when the user logs in, you derive the password verifier using the scheme stored in the DB for their account. If it matches, then you log them in (of course) but you *also* computer a new password verifier using the new, better scheme - say, PBKDF2 with 50000 iterations - and then store that new verifier, and the new scheme you used to derive it, to the user's account entry in the DB. It's completely transparent to the user. As a bonus, this makes it easy to adopt even stronger schemes in the future.
You know, for somebody who *sounds* like you're anti-DRM, that sure isn't reflected in your actions. If you want MS (and the other big players; they all watch each other) to reduce DRM in the future, reward them for *not* locking down the console like they originally planned to do. They listened to customer feedback. They responded. They made the right choice... and for that, you propose to punish them? Wow, that's a great plan right there. Which message do you think will be taken away from that?
1) Well, looks like nobody actually cared whether there was DRM or not in the first place. No need to listen to those whiney kids in the future!
2) Wow, sounds like people on the Internet are just petulant little children who will object *whatever* we do! May as well increase DRM in the future, at least the publishing industry likes that...
I'll give you a hint: neither one makes the attitude you're currently expressing look good, but neither one gets you what you want, either.
Oh, and out of curiosity, what other platforms do you develop for? Because there's only one that in the last few years has retroactively removed advertised features from its console in the name of more DRM, and then sued the people who worked around it. It's not Microsoft, either.
Well, you can use scripts. Powershell, CMD, and I think even WSH (JS or VBS) work...
But yeah, it's stupid. You can sideload for free though, no need to install apps through the store.
LOL! Remove or spoof the Referrer header (they only check the domain, don't bother changing the path) and it works. That's some awfully idiotic work right there.
It also makes it more appealing to actually copy the images and host them locally, rather than linking them directly. Yeah, that costs you a little, but with your link I at least knew where it came from... with a copied image that has the douchebag company's name cropped out, they get nothing.
Illegal, of course, unless the image was not actually under their copyright... but better than giving them money.
To be fair, part of the reason for the "sold only in our stores" thing was to ensure that the customers knew what they were buying. I have a Surface RT that I bought for research into the OS, and the guy at the Microsoft store made very sure I knew it wouldn't run legacy Windows programs.
He was wrong, after quite a bit of hacking (jailbreak + emulation layer = old games and undemanding desktop software work just fine), but that's beside the point.
This is actually already being worked on. The current version of the "jailbreak" hack for RT allows third-party kernel-mode drivers. The suggestion to write a driver which then dumps NT and bootstraps the Linux kernel has been discussed at length. It's not a small project, by any means, but it should be possible. Ideally, we would create a second bootloader entry (we do have limited control over the bootloader; it just always must execute a Microsoft-signed kernel) which would load RT, then jailbreak and load the driver, which would immediately load Linux. Alternatively, we could pretty easily have a "switch to Linux" shortcut on the desktop which (assuming jailbreak) would load that driver.
However, the whole process will take a fair bit of work. Also, MS is trying to prevent such "jailbreaks" in RT 8.1. Currently, downgrading back to 8.0 is still possible, at which point the jailbreak works again, though.
Sorry to self-reply, but in case it isn't obvious from the previous post: Microsoft could "fix" RT with a single, simple update. Reboot the tablet and the restrictions are gone.
An official x86 compat layer would be a fair bit of work, of course, but it's not really necessary to do that; the simple ability to run .NET apps (and maybe they get a few of their more important partners to flip the Platform option in Visual Studio to "ARM" and hit Build again; often it really is that simple) would make RT a lot more appealing.
There's no "port" involved. Or rather, they already did that. Then they added literally one configuration change to lock out non-MS-signed desktop apps. One change. It's a single flag in the kernel. On x86 and x64 builds of Windows NT, it's not set. On ARM builds of Windows NT (RT and WP8), it is.
Clear that flag (which is what the current "jailbreak" hack for RT does), and you can run any desktop software that will compile for ARM, or any .NET program, or any other language that can be run through one of the others (for example, Java is possible through IKVM, a .NET program implementing a JVM).
Now, as for domain joining, that's actually a simpler problem. All versions of Windows NT have had multiple SKUs (editions) ranging from the do-anything highest-end Server builds to the very crippled Starter builds. It's all the same codebase, just a configuration change. RT falls somewhere between Win8[Home] and Win8Pro SKUs in terms of business-y features; it can use BitLocker encryption (usually not available on Home) but cannot join domains (usually available on anything *except* Home).
Working around that particular restriction is also possible, though it is not easy unless you also remove the signature enforcement ("jailbreak") at which point it becomes nearly trivial.
Oh, and there's already a (very early and still incomplete) x86 emulation layer (actually, dynamic recompilation) for "jailbroken" RT devices. It's slow, as one would expect, but it can run old games and desktop software just fine. It also is the work of a single homebrew developer working from public documentation and reverse engineering for the Windows interoperability (calls to system libraries are thunked to ARM code, which is both faster than using x86 libraries and requires less install space). Microsoft could do a better job easily by putting a few of their people who previously worked in that space (for example, the "Virtual PC for Mac" software worked the same way, some of them are probably still around) on the job.
Surface RT has 2GB of RAM. Any app that would run in x86 emulation won't have a clue what to do with that much RAM.
It's funny how many people, even here and now, still conveniently forget what iPads cost when bashing competing products. Your statement "They were dramatically more expensive than most competing tablets" is a bald-faced lie. Surface RT has always been competitive with iPads and usually a bit cheaper than the newest model.
To be fair, sideloading (for "Metro") is present and free, though it's a little better hidden than on Android. The earlier versions of that "Jailbreak" you refer to required it, in fact.
Powershell command: Show-WindowsDeveloperLicenseRegistration
(yes, it supports Tab-completion.)
Xbox, for all that it took a ton of money to get it where it is, is now competing pretty well (we'll see how the next generation goes...).
Azure seems to be doing fine even if it's small potatoes next to AWS.
Same for SQL Server, a bit better than Azure even.
Windows Server (unlike Client) is nowhere near a monopoly in its market, but is also doing well.
In the pre-iPhone days, Windows Mobile was one of the more popular smartphone platforms without ever being nearly a monopoly.
To be fair, there's a "jailbreak" that enables all of that. .NET apps do run, unmodified and un-recompiled. Open source Win32 apps are being / have been ported. There are a few specifically-for-RT native apps as well, including an x86 dynamic recompilation layer that allows running (some) x86 apps directly.
This is all the work of a few volunteer devs on the XDA Developers forum, working for free on their own time. These efforts were made without MS support, just by public documentation and reverse engineering. Nonetheless, it makes the tablets (all RT devices) much, much more valuable to the (many) people who aren't willing to stick to "Modern" apps. MS could have done far, far better themselves in terms of a slick experience (starting with making the "jailbreak" be an option somewhere, instead of an Admin [which users have by default] to kernel [which is supposed to be inviolable] exploit that is scripted to run at bootup).