I don't get it. Never before coming to Slashdot have I seen the two mixed up so frequently. I never would've placed it on the level of "they're", "their", "there", which even I accidentally mix up sometimes in quick writing because they sound the same in my head. Loose and lose don't even sound the same.
It can't be a British difference, because then somebody would've told us by now and defended it.
So will it be a computer running a "traditional" OS with the whole Windows, Icons, Mouse Pointer interface or will Apple make it a big iPhone like device that hides a lot of that stuff under a (relatively) minimalist user interface?
A friend and I were arguing over this the other day. We agreed we really wanted something about the size of a Kindle but with a capacitive touchscreen, an LCD display, and a full suite of Internet apps.
The argument was whether this would happen earliest by scaling up cell phones (iPhone, etc) or by scaling down netbooks. This makes a difference in terms of what audience is being targeted, and what style of interface/control is acceptable, even if the base hardware is exactly identical.
I'm starting to side with him that it's going to be phones scaled up.
Everything is a file, and all files have the three permissions: Users, Groups, and Other.
Yes. This. I wish DBUS et al would realize that having more levels of abstraction just makes something harder to secure and use with standard Unix tools.
As a side-note, there *is* a flaw with this security model: it's very difficult if not impossible to sandbox an app so that it has some write access for the user running it, but not access to trash his entire home directory. SELinux and AppArmor supposedly fix this. Perhaps they will get more popular if Linux malware starts to pop up.
Oh, I see. "ABI differences in Linux" is a code word for "I am a Microsoft marketdroid".
Possibly, I haven't been on Windows for my primary workstation for 5 years so if that's what the phrase typically means I apologize for misusing it.
What I was referring to was primarily the 32-bit/64-bit split. On Windows all libraries are still 32-bit, whereas on Linux some people have 32-bit libraries and some have 64-bit libraries, so you have to build for both. Admittedly not that big a headache.
Also I've had some bad luck with binary apps since I've been on Gentoo, the Amazon MP3 downloader for example. I assume the problems were related to libaries that change subtly between minor versions in ways a source-compiled app wouldn't notice (Boost, for example, moved a function from a macro in a header to compiled code in the library).
These are the kind of reasons I don't like native code when I don't have the source available, and why I like to target managed-code when I can for my own projects. Again I apologize if I've been misusing terms.
No, the Supreme Court did it at the Federal Level, in 1978 in the Marquette decision, which is what I posted and you may want to look it up.
If you really want you can blame the Banking Act of 1864 for pre-empting state banking regulations, or say that the Banking Act was out of date and Congress should've updated it. The fact that Congress under Reagan didn't do anything about it is by no means unique to Reagan, neither did Carter, either Bush, or Clinton. Actually George H.W. Bush tried to enact a national usury law but dropped it when the stock market dropped in reaction.
But go ahead and blame Reagan for a Supreme Court decision made 4 years before he took office.
I think it's beautiful in concept, but implementation would be REALLY hard.
As a developer, I've tried using a managed-code runtime to stay truly platform agnostic (I.E. not have to produce builds for every OS under the sun). I tried to use Mono and Tao to write a current project, but little things like lack of Vorbis support made me give up and move to native code using Qt, which means I now need to compile things, deal with potential segfaults from ABI differences in Linux, the whole headache.
It would be nice if there was a higher-level game library written *natively* in.Net (not just wrappers to DLLs like Irrlicht currently offers). Maybe something like this exists in Java and someone can point me to it.
Well, one way to do this is to mount the users home / groups with the noexec flag.
No... GP is right. You can stop them from running native code if that's your goal, and maybe that's good protection against running dangerous system calls or god knows what.
BUT, they still have access to turing-complete languages, hell even full VMs via Java and Mono. If the goal was to stop them from running arbitrary code that interacts with the user, IE to stop them from playing games, you've lost. I wrote TiCalc programs to waste time in high school, I sure as hell can do the same in shell if I'm interested enough. Maybe your average user can't do it, but then they can bring their own laptop in and waste time on that, or sneak off to the bathroom and read a book.
These are battles you just can't win. Noexec should be used to guard against a potentially buggy kernel, nothing more.
My guess would be MCPs would be organizations without very much Linux/FOSS skills. If the Windows way is all you know, you live and die by the whims of Microsoft.
In the Land of UNIX Where Everything Works you can send GLX over the network for 3D graphics where ever the card lives, whether it's a VM host or a cluster headnode. That's probably more useful than emulating the 25 year old VGA BIOS and umpteen stupid extensions.
That's a neat idea, I had forgotten OpenGL worked like that. However, I don't really see a use case. You're going to virtualize an X11 app and have it connect to the X11 server on the host? Surely this is something you only want to do for one app at a time, in which case why the VM?
I think it behaves as though it was attached with the other quark it was produced with.
Not really, we're talking about *really* short timespans, much faster than strong interactions (order m_t-m_b is about a thousand times greater than order Lambda).
It behaves pretty much like a free top quark, that is, it's gone before it matters and only shows up as a mass resonance in a particular observable, or an excess of events over other Standard Model background after particular cuts in observables.
That's not entirely fair. We *do* keep a number of accelerators, like the neutron source in Oakridge, or the deep-inelastic-scattering electron accelerator at Jefferson Lab. Each machine has it's own niche, much like telescopes.
Fermilab, however, serves the exact same purpose as the LHC: high-energy hadron-hadron collisions in a synchrotron. The LHC will take data at 100 times the rate and 10 times the energy. The only reason to keep the Tevatron around (and there *are* other things at Fermilab) is to keep all that institutional knowledge employed for when you're ready to build the next-big-thing.
Once the LHC shows that it's running reliably, the Tevatron really is obsolete.
or find some screwed up roundabout way to push out to every machine via scripts...
That's a very Windows-centric POV. A script like "for i in workstations ; do scp newconfig admin@${i}:/etc/newconfig ; done" is *not* a roundabout way to do it. It is THE way to do it.
Explain to me how that doesn't tell you exactly what you're doing instead of relying on some "man behind the curtains" to distribute and enforce your Windows policies.
For those who were too young to know, or forgot, banks also had a limit to how much interest they could charge on a credit card, but Reagan decided it wasn't fair and made them free to charge whatever they wanted.
That's factually incorrect. South Dakota repealed their usury laws to lure the big banks away from NY. Marquette National Bank vs. First of Omaha Corp in 1978 made it so that a bank operating in SD could charge these rates to anyone in the country. That was well before Reagan.
You're right, companies seem hesitant to lower prices when they encounter lowered costs, even if it would gain them more customers and more total profit. That seems to be a bit of business foolishness that most companies suffer with. Luckily inflation ends up making this point irrelevant (prices fail to rise as fast as they could've).
Verizon, however, is not dumb. They know Sprint is eating their lunch on data plan pricing. They know they don't have exclusives like AT+T, and (comparative) bargain rates on voice and text like T-Mobile. They've got some work to do to stop customers from outflowing once their contracts are up, and pricing would be a good place to start.
Basic microeconomics says that production cost doesn't dictate price: what dictates price is what consumers are willing to pay.
I thought basic microeconomics said price depends on both. Equilibrium price is where the supply curve (amount willing to be produced for a unit price) intersects demand curve (amount willing to be purchased for a unit price).
So production costs matter. If suddenly every producer of an MMORPG had to pay a $5/mo/user tax, price *would* go up by some amount up to (and probably very close to) those $5, depending on what demand looked like at those prices.
Troll and Flamebait do not mean "we disagree with your argument". What exactly was it about the GP that was said in a trollish or flamebaiting manner other than mirroring, with only the briefest emphasis, a phrase that was part of the (highly modded!) comment he replied to?
Parent wasn't all that insightful either. GP's argument would hold for physical goods just as much as copyable goods. Copyright isn't involved.
We really need to stop abusing the moderation system as a way to suppress dissent.
That depends on how it's implemented. You could have it set up so a President could veto a bill and send back a version to Congress that they must vote on without further changes. That might pass constitutional muster, since Congress still votes on the laws, and while not perfect, could be effective at getting rid of some amounts of pork.
Features like VRML and javascript, CSS, a dizzying array of choices that seemed like it could go on forever.
That's not a good thing. You forget that many of the "innovations" of these browsers in the bad-old-days were to give themselves something the others didn't support. When we're talking about standards for the interchange of data, you want them to move slow so everyone can keep up.
Now I'll admit, the ubiquity of a browser with a lack of standards compliance and ui features like tabs are painful results of MS's monopoly, but let's be careful when we talk about "innovation" anywhere near standards.
Slashdot Mirror
If we could achieve with nuclear fusion what we have achieved with DHS
What? A parasitic reaction that just consumes and consumes and consumes, is more of a hindrance than a help, and wastes tons of money in the process?
Only to be given up on when another disaster claims the lives of thousands?
Wouldn't it work this way?:
In Soviet Russia:
1. Profit
2. ???
3. You
I don't get it. Never before coming to Slashdot have I seen the two mixed up so frequently. I never would've placed it on the level of "they're", "their", "there", which even I accidentally mix up sometimes in quick writing because they sound the same in my head. Loose and lose don't even sound the same.
It can't be a British difference, because then somebody would've told us by now and defended it.
I'm stymied.
If you can't see the causal relationship between video games and stabbing hookers, then you've never played Hello Kitty Island Adventure.
No, but I want to now!
So will it be a computer running a "traditional" OS with the whole Windows, Icons, Mouse Pointer interface or will Apple make it a big iPhone like device that hides a lot of that stuff under a (relatively) minimalist user interface?
A friend and I were arguing over this the other day. We agreed we really wanted something about the size of a Kindle but with a capacitive touchscreen, an LCD display, and a full suite of Internet apps.
The argument was whether this would happen earliest by scaling up cell phones (iPhone, etc) or by scaling down netbooks. This makes a difference in terms of what audience is being targeted, and what style of interface/control is acceptable, even if the base hardware is exactly identical.
I'm starting to side with him that it's going to be phones scaled up.
Everything is a file, and all files have the three permissions: Users, Groups, and Other.
Yes. This. I wish DBUS et al would realize that having more levels of abstraction just makes something harder to secure and use with standard Unix tools.
As a side-note, there *is* a flaw with this security model: it's very difficult if not impossible to sandbox an app so that it has some write access for the user running it, but not access to trash his entire home directory. SELinux and AppArmor supposedly fix this. Perhaps they will get more popular if Linux malware starts to pop up.
Oh, I see. "ABI differences in Linux" is a code word for "I am a Microsoft marketdroid".
Possibly, I haven't been on Windows for my primary workstation for 5 years so if that's what the phrase typically means I apologize for misusing it.
What I was referring to was primarily the 32-bit/64-bit split. On Windows all libraries are still 32-bit, whereas on Linux some people have 32-bit libraries and some have 64-bit libraries, so you have to build for both. Admittedly not that big a headache.
Also I've had some bad luck with binary apps since I've been on Gentoo, the Amazon MP3 downloader for example. I assume the problems were related to libaries that change subtly between minor versions in ways a source-compiled app wouldn't notice (Boost, for example, moved a function from a macro in a header to compiled code in the library).
These are the kind of reasons I don't like native code when I don't have the source available, and why I like to target managed-code when I can for my own projects. Again I apologize if I've been misusing terms.
No, the Supreme Court did it at the Federal Level, in 1978 in the Marquette decision, which is what I posted and you may want to look it up.
If you really want you can blame the Banking Act of 1864 for pre-empting state banking regulations, or say that the Banking Act was out of date and Congress should've updated it. The fact that Congress under Reagan didn't do anything about it is by no means unique to Reagan, neither did Carter, either Bush, or Clinton. Actually George H.W. Bush tried to enact a national usury law but dropped it when the stock market dropped in reaction.
But go ahead and blame Reagan for a Supreme Court decision made 4 years before he took office.
Does that mean they're half way done?
OHHHHHHHH LIving on a prayer!
Sorry. Had to. It's in my contract. =)
I think it's beautiful in concept, but implementation would be REALLY hard.
.Net (not just wrappers to DLLs like Irrlicht currently offers). Maybe something like this exists in Java and someone can point me to it.
As a developer, I've tried using a managed-code runtime to stay truly platform agnostic (I.E. not have to produce builds for every OS under the sun). I tried to use Mono and Tao to write a current project, but little things like lack of Vorbis support made me give up and move to native code using Qt, which means I now need to compile things, deal with potential segfaults from ABI differences in Linux, the whole headache.
It would be nice if there was a higher-level game library written *natively* in
Well, one way to do this is to mount the users home / groups with the noexec flag.
No... GP is right. You can stop them from running native code if that's your goal, and maybe that's good protection against running dangerous system calls or god knows what.
BUT, they still have access to turing-complete languages, hell even full VMs via Java and Mono. If the goal was to stop them from running arbitrary code that interacts with the user, IE to stop them from playing games, you've lost. I wrote TiCalc programs to waste time in high school, I sure as hell can do the same in shell if I'm interested enough. Maybe your average user can't do it, but then they can bring their own laptop in and waste time on that, or sneak off to the bathroom and read a book.
These are battles you just can't win. Noexec should be used to guard against a potentially buggy kernel, nothing more.
My guess would be MCPs would be organizations without very much Linux/FOSS skills. If the Windows way is all you know, you live and die by the whims of Microsoft.
In the Land of UNIX Where Everything Works you can send GLX over the network for 3D graphics where ever the card lives, whether it's a VM host or a cluster headnode. That's probably more useful than emulating the 25 year old VGA BIOS and umpteen stupid extensions.
That's a neat idea, I had forgotten OpenGL worked like that. However, I don't really see a use case. You're going to virtualize an X11 app and have it connect to the X11 server on the host? Surely this is something you only want to do for one app at a time, in which case why the VM?
It's 2009, Rick-Rolls are ancient history. THIS is what I predict will replace it:
http://www.youtube.com/watch?v=oHg5SJYRHA0
I think it behaves as though it was attached with the other quark it was produced with.
Not really, we're talking about *really* short timespans, much faster than strong interactions (order m_t-m_b is about a thousand times greater than order Lambda).
It behaves pretty much like a free top quark, that is, it's gone before it matters and only shows up as a mass resonance in a particular observable, or an excess of events over other Standard Model background after particular cuts in observables.
That's not entirely fair. We *do* keep a number of accelerators, like the neutron source in Oakridge, or the deep-inelastic-scattering electron accelerator at Jefferson Lab. Each machine has it's own niche, much like telescopes. Fermilab, however, serves the exact same purpose as the LHC: high-energy hadron-hadron collisions in a synchrotron. The LHC will take data at 100 times the rate and 10 times the energy. The only reason to keep the Tevatron around (and there *are* other things at Fermilab) is to keep all that institutional knowledge employed for when you're ready to build the next-big-thing.
Once the LHC shows that it's running reliably, the Tevatron really is obsolete.
or find some screwed up roundabout way to push out to every machine via scripts...
That's a very Windows-centric POV. A script like "for i in workstations ; do scp newconfig admin@${i}:/etc/newconfig ; done" is *not* a roundabout way to do it. It is THE way to do it.
Explain to me how that doesn't tell you exactly what you're doing instead of relying on some "man behind the curtains" to distribute and enforce your Windows policies.
For those who were too young to know, or forgot, banks also had a limit to how much interest they could charge on a credit card, but Reagan decided it wasn't fair and made them free to charge whatever they wanted.
That's factually incorrect. South Dakota repealed their usury laws to lure the big banks away from NY. Marquette National Bank vs. First of Omaha Corp in 1978 made it so that a bank operating in SD could charge these rates to anyone in the country. That was well before Reagan.
You're right, companies seem hesitant to lower prices when they encounter lowered costs, even if it would gain them more customers and more total profit. That seems to be a bit of business foolishness that most companies suffer with. Luckily inflation ends up making this point irrelevant (prices fail to rise as fast as they could've).
Verizon, however, is not dumb. They know Sprint is eating their lunch on data plan pricing. They know they don't have exclusives like AT+T, and (comparative) bargain rates on voice and text like T-Mobile. They've got some work to do to stop customers from outflowing once their contracts are up, and pricing would be a good place to start.
Basic microeconomics says that production cost doesn't dictate price: what dictates price is what consumers are willing to pay.
I thought basic microeconomics said price depends on both. Equilibrium price is where the supply curve (amount willing to be produced for a unit price) intersects demand curve (amount willing to be purchased for a unit price).
So production costs matter. If suddenly every producer of an MMORPG had to pay a $5/mo/user tax, price *would* go up by some amount up to (and probably very close to) those $5, depending on what demand looked like at those prices.
As of right now, GP is 0 and parent is +5.
Troll and Flamebait do not mean "we disagree with your argument". What exactly was it about the GP that was said in a trollish or flamebaiting manner other than mirroring, with only the briefest emphasis, a phrase that was part of the (highly modded!) comment he replied to?
Parent wasn't all that insightful either. GP's argument would hold for physical goods just as much as copyable goods. Copyright isn't involved.
We really need to stop abusing the moderation system as a way to suppress dissent.
That depends on how it's implemented. You could have it set up so a President could veto a bill and send back a version to Congress that they must vote on without further changes. That might pass constitutional muster, since Congress still votes on the laws, and while not perfect, could be effective at getting rid of some amounts of pork.
Features like VRML and javascript, CSS, a dizzying array of choices that seemed like it could go on forever.
That's not a good thing. You forget that many of the "innovations" of these browsers in the bad-old-days were to give themselves something the others didn't support. When we're talking about standards for the interchange of data, you want them to move slow so everyone can keep up.
Now I'll admit, the ubiquity of a browser with a lack of standards compliance and ui features like tabs are painful results of MS's monopoly, but let's be careful when we talk about "innovation" anywhere near standards.