By going after multimedia applications like HDTV, video games, etc. they're targeting a market historically willing to pay a premium to get the best performance.
Is this really true? I think you're smoking crack... HDTV and game consoles are consumer electronics, which is a market almost entirely driven by price. I don't know about HDTV, but I can't think of any way the popular consoles could be construed as including "premium" parts...
GMR in hard disk read heads isn't really spintronics, though. It depends on the interaction between magnetic fields and electrons of different spins, but you don't measure the spin of electrons coming out of a GMR sensor, you measure the current flow just like with a traditional disk read head.
could you please translate all those acronyms into more plain English just so electronics illiterates such as me could actually read your comment?
So tell me, moderators (because the question just sounds like a troll to me), does knowing what those acronyms stand for actually help you understand the post?:)
Really the important point is that it took them 30 years from the initial development of the new IC manufacturing technology (CMOS) to really replace the old crappy technology (PMOS/NMOS). I'm just out of school, and by analogy none of this fancy new technology will be commonplace until I'm almost retired... of course the IC business is much bigger now and there's more pressure to bring new technologies into production quickly.
The areal density increases about 1560-fold (assuming 640Mb/CD), but the linear density increases only by the square root of this.
You're assuming that the number of tracks increases at the same rate the linear density increases. That might be a reasonable assumption for DVDs vs. CDs, which are made denser by scaling everything down (which you can do because the light is a smaller wavelength and can resolve smaller features on the disc), but not so much for this format.
This format uses a red laser for tracking, so the tracks can't really be packed closer together than on a DVD; and anyway the big boost in data density comes from the holographic technology, which packs more data per linear unit, but does not pack it in a narrower space.
You got screwed. You should have been able to get refunds for the NSF fees and letters of apology to every person who tried to deposit your cheques, saying it was their fault and not yours. Unfortunately getting this out of a bank can take a bit of cajoling, but it's usually possible. They screwed up and tarnished your reputation -- aren't they big enough to take responsibility for their mistakes?
Cisco retains the right to assert its patents against any product or portion thereof that is not necessary for compliance with RFC XXXX
Nice. This means that nobody can implement this in GPL'd software (wherever software patents apply), because the GPL requires that anybody be able to modify and redistribute the software without encumbrance, regardless of what they're doing with it. So, not in Linux.
AMD is more and more concentrated on taking the server room from Intel. Once they've done this, they'll trickle home just the same way as Intel processors did in ages ago.
What are you talking about? Intel started out in el-cheapo desktop systems, and gradually trickled into the server market, taking market share from mainframes and high-end servers.
You're missing the point: "Commercial support can be an extremely important thing". The comment was a response to someone poo-pooing commercial support for Plone, so I think it's pretty clear from context that it's making the case for commercial support generally -- for open source software as well as for closed source software.
20 times as much power as the old C64 in every measurement
Well, not exactly. The C64 CPU was a total wuss, but it had dedicated peripheral hardware that was designed for games -- in particular the video hardware included a sprite and tile engine that did animation, collision detection, and scrolling, and it also had a multi-voice sound synthesizer. Palms have none of these things, they just have a linear framebuffer and a single tone generator, and duplicating even some of these animation and sound capabilities is very taxing for most Palms. There are scrolling action games for Palm OS, but not many.
The newer, expensive Palms (with PalmOS 5) use ARM chips, which probably do have the requisite horsepower to emulate those graphics and sound functions, and Clies have offboard DSPs for sound processing. Maybe we'll see more scrolling games in the future for Palms..
You've stumbled onto the state of gaming "journalism." The hard questions don't get asked and every upcoming game is exciting and interesting.
and anything deeper than "it's late, but it'll be out by christmas, here's a quick excuse" is really only interesting if you are or want to be a game developer, or you're a part-owner of valve. it's hard for this kind of article to be hard-hitting and deep, simply because of the subject matter and audience. it's not exactly watergate.
if you are seriously interested in why games are late and other game development topics, you should subscribe to Game Developer magazine. they publish post-mortems of big games every month.
if you do re-assign the rights to the FSF, you are no longer the owner and may not reissue the program under any other license
Hmmm, is that really true? I don't think so, but feel free to find some evidence to support the statement.
it would be true, except that the FSF grants you back basically all the rights you had before you assigned your copyright to them. you can still use the software and sublicense it however you want to anyone else, as long as you inform the FSF you're doing it (presumably so they don't try to drag the sublicensee through court for a GPL violation); the only thing you can't do is revoke the FSF's rights. as far as I know.
alright, I'm a long way from being an EE, but I remember my high-school physics pretty well.
a magnetic field will form in concentric rings around any conductor that's carrying current, and the strength of that magnetic field will be proportional to the amount of current flowing through the conductor, and will drop off sharply with distance.
now, currents are only induced by a changing magnetic field, and the current will be proportional to the change in magnetic field strength. so if the power lines were carrying DC, and you were riding parallel to the lines, no current would be induced in you because the strength of the magentic field where you were riding would stay about the same -- but the power lines are probably carrying AC, so even if you stay still, the magnetic field will be reversing polarity every 1/60th of a second, and a current will be induced in you.
however, there are a few mitigating factors.
one is that you're probably several meters away from the lines, and magnetic fields don't couple well through air. transformers often have an iron core to guide the magnetic field between the two coils, because otherwise not much power gets transferred between them.
another factor is that the magnetic field around a single conductor is not very strong, unless it's carrying really a lot of current (like wire-melting amounts). again, transformers and electromagnets use coils with hundreds or thousands of turns to get a decent magnetic field going.
the last factor is that the lines are carrying high voltages. the comment in your submission seems to indicate that you think this will create a stronger magnetic field, but actually the opposite is true. for our purposes, electric power has two aspects, voltage and current; the voltage determines the strength of the electric field, and the current determines the strength of the magnetic field. for the same amount of power, if you increase the voltage, you reduce the current. the power company doesn't want current being induced in random objects, because that power gets lost as heat and costs them money (actually, they probably care more about the fact that large currents create heat in the wire itself -- but it works out to the same meaning for you). they run long lines at very high voltages as part of a strategy to minimize this power loss and save money.
I've tried the "ya know, this really needs changing, and here's a few reasons why..." approach. The response I've gotten was "No. You're an idiot. Your idea is stupid. We'll never do that. Go away."
having dealt with a cranky developer or two, I know what you're talking about. however, being a cranky developer myself, and knowing that a lot of these people are doing work for you for free -- I have to agree with the subject that "beggars can't be choosers".
here's a parallel:
I work in a touristy corner of downtown. when I go out for lunch, at least 3 beggars ask me for change each way. occasionally one of them follows me around, and gets angry that I won't give him anything for the irritating song he's playing on his busted guitar.
I don't have enough money to give them each a dollar. I don't even have the time to explain why I'm not giving them money or why I don't want to listen to their song. it would take me an hour and a half to do it, and my lunch break is only 30 minutes.
similarly, you're one user among millions. the kernel hackers in charge of the main tree don't even have time to explain to every user why their ideas are wrong. it's not worth the time to consider every single user's ideas seriously; occasionally a gem might be lost, but many many wasted hours are saved.
so what can you do? well, even if you can't hack C yourself, you can sponsor someone who can. you're not powerless, you're just used to existing on handouts. if you think user responsiveness is important enough to be worth money, why not start an organization that sponsors kernel developers in exchange for them spending a little more time answering questions and seriously considering your ideas?
bear in mind, no commercial software vendor lets users talk directly to its developers for free. you have to pay through the nose for any guarantee of service at all, and you've already paid for their product! when was the last time a Microsoft technical support person told you, "you know, your ideas about how we should modify user permissions handling are really good. We'll definitely be including those changes in Blackcomb!"? even as a developer of Palm OS-based software I have to pay PalmSource $500 just for the privilege of learning how to work around bugs in their software, much less have input into how their software is designed.
Linux has nothing to do with Gnu. If you wanted to, you could make an OS using the Linux kernel & the BSD tools.
...except that you'd still need gnu libc to run those tools and gcc to compile them. you can easily find replacements for the gnu textutils or for bash, but libc is kinda essential.
on the other hand it's easy to see why people drop the gnu part of gnu/linux, and it has nothing to do with credit or attribution; they're just picking out the most identifying feature of an os that otherwise has no name.
gcc and gnu libc are critical (and incidentally account for about as much source code as the linux kernel last I checked), but all libc's and cc's aspire to work according to the same standard; so saying that your os was compiled on gcc or uses gnu libc isn't helpful to someone trying to grok its major characteristics.
Since these are DIGITAL subpixels, for 24 bit color, you'd need 256 sub-pixels just for the 'red' to mimic the 256 intensity levels per gun
indeed. on the other hand, most colour printing processes only have 1-bit colour depth; but at 720 DPI dithering can look very good -- and apparently they can pack these elements as densely as 1000 DPI, so a dithered display might work well. also, they might have the option of using pulse width modulation (blinking the display elements on and off very rapidly to get halftones).
the long and short is that although the digital nature of the display elements is a limitation, it probably isn't a killer given a little time and ingenuity.
what I'd like to know is, how do they get those lovely paper white images?
if the pixels switch between "colour" and "black", a straight red-green-blue display should never be able to achieve full-intensity paper white if the display is purely reflective.
(in fact, I thought the only way to get full gamut in a reflective display was to use 3 transmissive layers and do subtractive mixing -- can someone correct me if I'm wrong?)
since they're advertising paper white displays, I'm really curious what other tricks they've got that aren't described in the promotional material.
for what it's worth, a hi-res, highly reflective, thin, low power black-and-white display would be a dream come true for my PDA, even if there was no way to extend it to full colour.
I love how the totally bogus description of an attack against one-time-pads gets moderated way up (there is no possible attack against a one-time-pad, besides knowing what the pad is), while the author's message pointing out that he made a mistake languishes..
in C++ you can, but it's a bad idea because if someone overrides that function things may not work as expected:
#include <iostream>
struct A {A() {foo();} virtual void foo() {std::cout << "in A::foo" << std::endl;}};
struct B: public A {virtual void foo() {std::cout << "in B::foo" << std::endl;}};
executing the constructor B() produces the output "in A::foo", despite the fact that B overrides A's foo. this happens for the perfectly logical reason that since B hasn't been constructed yet, you can't call any of its methods, but it can be quite confusing. a parallel argument applies to destructors.
Meanwhile if an American died it'd be so outrageous and all over the frontpage news...
er, in the 4 page globe-and-mail friendly fire insert, it indicated that so far there have been 36 american friendly-fire deaths. I didn't see those reported in the canadian news..
Slashdot Mirror
A more... evocative term for this is Software EXchange.
Moderators! This is a *funny* retort to a *stupid* grandparent... "Interesting" my ass...
GMR in hard disk read heads isn't really spintronics, though. It depends on the interaction between magnetic fields and electrons of different spins, but you don't measure the spin of electrons coming out of a GMR sensor, you measure the current flow just like with a traditional disk read head.
So tell me, moderators (because the question just sounds like a troll to me), does knowing what those acronyms stand for actually help you understand the post? :)
Really the important point is that it took them 30 years from the initial development of the new IC manufacturing technology (CMOS) to really replace the old crappy technology (PMOS/NMOS). I'm just out of school, and by analogy none of this fancy new technology will be commonplace until I'm almost retired... of course the IC business is much bigger now and there's more pressure to bring new technologies into production quickly.
If you want to spend $50 a month, 5 accounts should be enough downloads to experiment quite a bit...
You're assuming that the number of tracks increases at the same rate the linear density increases. That might be a reasonable assumption for DVDs vs. CDs, which are made denser by scaling everything down (which you can do because the light is a smaller wavelength and can resolve smaller features on the disc), but not so much for this format.
This format uses a red laser for tracking, so the tracks can't really be packed closer together than on a DVD; and anyway the big boost in data density comes from the holographic technology, which packs more data per linear unit, but does not pack it in a narrower space.
You got screwed. You should have been able to get refunds for the NSF fees and letters of apology to every person who tried to deposit your cheques, saying it was their fault and not yours. Unfortunately getting this out of a bank can take a bit of cajoling, but it's usually possible. They screwed up and tarnished your reputation -- aren't they big enough to take responsibility for their mistakes?
What are you talking about? Intel started out in el-cheapo desktop systems, and gradually trickled into the server market, taking market share from mainframes and high-end servers.
You're missing the point: "Commercial support can be an extremely important thing". The comment was a response to someone poo-pooing commercial support for Plone, so I think it's pretty clear from context that it's making the case for commercial support generally -- for open source software as well as for closed source software.
Well, not exactly. The C64 CPU was a total wuss, but it had dedicated peripheral hardware that was designed for games -- in particular the video hardware included a sprite and tile engine that did animation, collision detection, and scrolling, and it also had a multi-voice sound synthesizer. Palms have none of these things, they just have a linear framebuffer and a single tone generator, and duplicating even some of these animation and sound capabilities is very taxing for most Palms. There are scrolling action games for Palm OS, but not many.
The newer, expensive Palms (with PalmOS 5) use ARM chips, which probably do have the requisite horsepower to emulate those graphics and sound functions, and Clies have offboard DSPs for sound processing. Maybe we'll see more scrolling games in the future for Palms..
and anything deeper than "it's late, but it'll be out by christmas, here's a quick excuse" is really only interesting if you are or want to be a game developer, or you're a part-owner of valve. it's hard for this kind of article to be hard-hitting and deep, simply because of the subject matter and audience. it's not exactly watergate.
if you are seriously interested in why games are late and other game development topics, you should subscribe to Game Developer magazine. they publish post-mortems of big games every month.
if you do re-assign the rights to the FSF, you are no longer the owner and may not reissue the program under any other license
Hmmm, is that really true? I don't think so, but feel free to find some evidence to support the statement.
it would be true, except that the FSF grants you back basically all the rights you had before you assigned your copyright to them. you can still use the software and sublicense it however you want to anyone else, as long as you inform the FSF you're doing it (presumably so they don't try to drag the sublicensee through court for a GPL violation); the only thing you can't do is revoke the FSF's rights. as far as I know.
alright, I'm a long way from being an EE, but I remember my high-school physics pretty well.
a magnetic field will form in concentric rings around any conductor that's carrying current, and the strength of that magnetic field will be proportional to the amount of current flowing through the conductor, and will drop off sharply with distance.
now, currents are only induced by a changing magnetic field, and the current will be proportional to the change in magnetic field strength. so if the power lines were carrying DC, and you were riding parallel to the lines, no current would be induced in you because the strength of the magentic field where you were riding would stay about the same -- but the power lines are probably carrying AC, so even if you stay still, the magnetic field will be reversing polarity every 1/60th of a second, and a current will be induced in you.
however, there are a few mitigating factors.
one is that you're probably several meters away from the lines, and magnetic fields don't couple well through air. transformers often have an iron core to guide the magnetic field between the two coils, because otherwise not much power gets transferred between them.
another factor is that the magnetic field around a single conductor is not very strong, unless it's carrying really a lot of current (like wire-melting amounts). again, transformers and electromagnets use coils with hundreds or thousands of turns to get a decent magnetic field going.
the last factor is that the lines are carrying high voltages. the comment in your submission seems to indicate that you think this will create a stronger magnetic field, but actually the opposite is true. for our purposes, electric power has two aspects, voltage and current; the voltage determines the strength of the electric field, and the current determines the strength of the magnetic field. for the same amount of power, if you increase the voltage, you reduce the current. the power company doesn't want current being induced in random objects, because that power gets lost as heat and costs them money (actually, they probably care more about the fact that large currents create heat in the wire itself -- but it works out to the same meaning for you). they run long lines at very high voltages as part of a strategy to minimize this power loss and save money.
I've tried the "ya know, this really needs changing, and here's a few reasons why..." approach. The response I've gotten was "No. You're an idiot. Your idea is stupid. We'll never do that. Go away."
having dealt with a cranky developer or two, I know what you're talking about. however, being a cranky developer myself, and knowing that a lot of these people are doing work for you for free -- I have to agree with the subject that "beggars can't be choosers".
here's a parallel:
I work in a touristy corner of downtown. when I go out for lunch, at least 3 beggars ask me for change each way. occasionally one of them follows me around, and gets angry that I won't give him anything for the irritating song he's playing on his busted guitar.
I don't have enough money to give them each a dollar. I don't even have the time to explain why I'm not giving them money or why I don't want to listen to their song. it would take me an hour and a half to do it, and my lunch break is only 30 minutes.
similarly, you're one user among millions. the kernel hackers in charge of the main tree don't even have time to explain to every user why their ideas are wrong. it's not worth the time to consider every single user's ideas seriously; occasionally a gem might be lost, but many many wasted hours are saved.
so what can you do? well, even if you can't hack C yourself, you can sponsor someone who can. you're not powerless, you're just used to existing on handouts. if you think user responsiveness is important enough to be worth money, why not start an organization that sponsors kernel developers in exchange for them spending a little more time answering questions and seriously considering your ideas?
bear in mind, no commercial software vendor lets users talk directly to its developers for free. you have to pay through the nose for any guarantee of service at all, and you've already paid for their product! when was the last time a Microsoft technical support person told you, "you know, your ideas about how we should modify user permissions handling are really good. We'll definitely be including those changes in Blackcomb!"? even as a developer of Palm OS-based software I have to pay PalmSource $500 just for the privilege of learning how to work around bugs in their software, much less have input into how their software is designed.
ok, rant over.
Linux has nothing to do with Gnu. If you wanted to, you could make an OS using the Linux kernel & the BSD tools.
...except that you'd still need gnu libc to run those tools and gcc to compile them. you can easily find replacements for the gnu textutils or for bash, but libc is kinda essential.
on the other hand it's easy to see why people drop the gnu part of gnu/linux, and it has nothing to do with credit or attribution; they're just picking out the most identifying feature of an os that otherwise has no name.
gcc and gnu libc are critical (and incidentally account for about as much source code as the linux kernel last I checked), but all libc's and cc's aspire to work according to the same standard; so saying that your os was compiled on gcc or uses gnu libc isn't helpful to someone trying to grok its major characteristics.
Since these are DIGITAL subpixels, for 24 bit color, you'd need 256 sub-pixels just for the 'red' to mimic the 256 intensity levels per gun
indeed. on the other hand, most colour printing processes only have 1-bit colour depth; but at 720 DPI dithering can look very good -- and apparently they can pack these elements as densely as 1000 DPI, so a dithered display might work well. also, they might have the option of using pulse width modulation (blinking the display elements on and off very rapidly to get halftones).
the long and short is that although the digital nature of the display elements is a limitation, it probably isn't a killer given a little time and ingenuity.
what I'd like to know is, how do they get those lovely paper white images?
if the pixels switch between "colour" and "black", a straight red-green-blue display should never be able to achieve full-intensity paper white if the display is purely reflective.
(in fact, I thought the only way to get full gamut in a reflective display was to use 3 transmissive layers and do subtractive mixing -- can someone correct me if I'm wrong?)
since they're advertising paper white displays, I'm really curious what other tricks they've got that aren't described in the promotional material.
for what it's worth, a hi-res, highly reflective, thin, low power black-and-white display would be a dream come true for my PDA, even if there was no way to extend it to full colour.
I love how the totally bogus description of an attack against one-time-pads gets moderated way up (there is no possible attack against a one-time-pad, besides knowing what the pad is), while the author's message pointing out that he made a mistake languishes..
Why cant you call virtuals from a constructor?
in C++ you can, but it's a bad idea because if someone overrides that function things may not work as expected:
#include <iostream>
struct A {A() {foo();} virtual void foo() {std::cout << "in A::foo" << std::endl;}};
struct B: public A {virtual void foo() {std::cout << "in B::foo" << std::endl;}};
executing the constructor B() produces the output "in A::foo", despite the fact that B overrides A's foo. this happens for the perfectly logical reason that since B hasn't been constructed yet, you can't call any of its methods, but it can be quite confusing. a parallel argument applies to destructors.