There is one successful method. It can be deduced from onomatopeic writings (when people tried to write down how something sounds). Of course exact dialect is almost unrecoverable.
Btw. one familiar backronym example is RIP, originally "requiem in pace", now people assume that it's "rest in peace".
So the real cause of sexual education programs is to scare people into not having sex at all because of viruses that cause cancer? I could imagine religious sects to think that those are a sign from God.
Unfurtunately, genital herpes, HIV, CMV, EBR, hepatitis and myriad of nasty bacterial infections still exist to scare teenagers.
Gaming problems are probably related to still experimental and less optimised 3D drivers for Vista's new graphic card driver model. 3D drivers for XP by ATI/Nvidia are usually full of nasty hacks, purpose of which is to speed up major gaming benchmarks (3Dmark, Quake, Doom, UT etc.). I guess many of those tricks aren't, well, portable (easily) to a new driver model. Vista can use XP drivers drivers though, but I guess this could work less than optimal.
As DWM shuts down whenever app locks front buffer (and direct3D games do that), I doubt that compositing is a reason why games perform slower. Another possibility is that scheduler classes (described in first part article) might affect CPU performance of games tweaked to run well with XP scheduling (e.g. I experienced hiccups with PES6).
Regarding article(s), I must say I'm not that impressed with amount of changes that got in their kernel since XP (except maybe by new driver framework which is a different story). In many areas mentioned in articles, Linux kernel is leaps and bounds ahead (scheduler,MM,IO-areas which saw huge refinements during 2.6 cycle and are constantly being improved). In others it is on-par (or almost here, like with PM or init mechanism). Exception seems to be Ready* stuff, but this is not really "a next big" thing anyway and seems as something that few decent kernel hackers could implement in a month or two. Also Vista now seems to have better graphic driver model (though DRI is closing the gap) and better userspace driver model (except they don't have FUSE equivalent).
True. But it's still easier to make a good management tool than a hypervisor or implement usual non-VT/Pacifica x86 virtualisation with good performance (the latter arcanely hard). So Vmware, which really excells at software engineering will still have to keep their virtualisation ahead of competition to be industry leader. Yes, they are now pressured by both Windows and Linux, but their package is ahead of both.
It's hard to guess how ahead Microsoft is with this, but i somehow doubt that it's as advanced as virtualisation that's (becoming) available on Linux (KVM, paravirt ops, etc.). I doubt that they are using full kernel as hypervisor (like KVM - this is IMO best long-term approach, though it takes time to pefect). In addition Linux also has good support for jail-like separation (Virtuozzo etc) which is even of bettr performance. Microsoft could be basing their stuff on Xen, along with acquired Connectix.
Maybe Microsoft is planning to take advance of control they have over graphic drivers - to implement 3D acceleration over paravirtualised interface (I'm sure they DON'T want this to run on any other host OS except Microsoft). As usual, they don't care about Windows running paravirtualised on any other OS, and probably doing steps to prevent this from happening anytime.
Also it's hard to believe that Microsoft cares about Vmware that much, more about Red Hat which will soon release RHEL5 with Xen support and Fedora with both Xen and KVM support.
For two processor cores having 1Mb cache each, it isn't that much (Prescott was single core). Wonder why they still stick to P4 performance scheme, as that processor family is a history. Are they planning to improve on existing K8 design? Else they won't be able to catch-up with Intel in performance, especially as Intel has now maybe solved current leakage problem (and AMD usually lags behind Intel in production process). Currently they only have advantage in merging with ATI GPU tech as they're planning to produce combo GPU/CPU - it isn't clear whether it will be for low-end ultra-cheap machines or will it compete with mainstream gaming machines.
Unfortunately real democracy exists only on this level, especially in the USA, but in some way also in other countries. On "macroscopic" level there are only democrats and republicans. True oligarchy, USA is aristocratic country. (lesser parties can't even get ONE candidate into a parliament without having (relative) majority of votes in big pieces of US land, isn't that bad enough?)
Hackers seem to be able to get this info from Nv cards anyway (renouveau tool, Radeons already figured out), so giving documentation to OSS developer under NDA won't cost Nvidia anything: It will be figured out sooner or later and open source driver code produced - it's just a matter of time.
Yes, it's hard to write graphic card drivers, but you are underestimating community skill (just look at amount and quality of driver work already present in kernel). As I said, whole community working on commong driver framework core features can significantly reduce total amount of work required, and also make getting working drivers for future hardware much easier. (btw. GLSL compiler will be in next Mesa)
Not true. It may be hard to get something on reverse-engineered boards, but Intel example shows that good performance and advanced drivers are possible if company decides to pay skilled programmers to work on it (and give them proper documentation). OTOH, dirty performance hacks, often used by manufacturers at the price of compatibility and stability of drivers, aren't acceptable in open source, so performance might be somewhat lower (Windows gamers are lucky that they can't see what kind of driver garbage is installed on their machines).
Good side of open source is certainly that there is common framework which can be used across graphic hardware: this significantly reduces duplication and is beneficial to all users, not just those using particular hw. (and btw. Microsoft if trying to force manufacturers into something like that in Vista, their system is similar in concept to DRI). Unfortunately manufacturers don't like this: especially those which invest a lot in Linux drivers, like Nvidia. They want to differentiate, not give their work to other companies for free (and btw, there is also issue of various patented stuff, like S3TC).
Hos about "fake" reviews by gaming sites or hardware sites (many of them having a deal with publishers/manufacturers), or TV shows which present new cars? It's all fraud marketing with lots of praising and hiding of facts.
NDA is OK if it contains contain only reasonable clauses like guarantee that no information regarding hardware will be revealed except the source code (withough huge comments). Though it is obvious that source code will reveal stuff. I believe companies are just afraid to simply release docs to anyone, as it is very easy to copy and produce identical clone. A driver probably won't access all, but rather a subset of interfaces. Instead of waiting for community to reverse engineer windows drivers, it is better for them to cooperate in this fashion, as there really isn't any real cost. Another nice option here is to cooperate with a company developer on the driver code.
It isn't a big issue for a FOSS (krenel) developer to sign a NDA with hardware manufacturers if he works on software only. For example, one Xorg driver developer signed NDA with ATI to develop 2D driver for X1300 and newer cards. (Unfortunately he still isn't allowed to release source).
As for sample devices, I don't think it is a problem to ship few of them under NDA, even hardware review sites get those.
U.S. didn't even stop pursuing the cold war, as they continued the arms race solely, to get a quick advantage over other emerging powers as China, ASAP. And now, when someone else tests an effective weapon which can significantly (and cheaply) cripple U.S. surveillance power, suddenly they cry for "disarmament" - but obviously of all but themselves, as they keep pumping $500 bln/yearly into the war machinery and R&D (not eve counting Iraq expenses!). How lame, this doesn't have anything to do with defence, it's an aggressive armament with purpose to wage wars like one in Iraq.
I hope their shiny world dominance model will collapse sooner or later.
Good thing is that those stabilised releases are much more similar to development versions than when 2.4 and 2.5 were state of art, so most fixes can easily be comitted to latest upstream version (or backported from it).
Turning experimental version of such a big project into a stable one is a long and painful job. It's order of magintude easier for Linus and kernel team to wait for a patchset to stabilise out of main tree until it's refactored, tested and good enough to go in. I believe both users and developers are happier with this (and improvements get a fair review, aren't just bumped in).
Whatever, it's their contry and it's not busineess of G.W.Bush nor US neocons to decide what will happen with it. Yes, people there maybe aren't as civilised as westerners (not including US population!), but they still are nationalists fignting rightfully for their soil.
Of course U.S. can't just leave this place with vast oil reserves, as it will once again turn into anti-US playground and probably later divide between factions, with bloody battles for oil field areas. As UN sanctions are already lifted, they will be free to do business with french, russian and other non-US oil companies, maybe even use Euro, which would be a pending disaster for U.S. economy. In other words, enemies of the U.S will get huge power in their hands.
US did loose Iran, and can't affort to loose another such country again. It is perfectly understandable why Bush is going to try to drawn this resistance in blood.
Noone believes that the engagement will be successful, instead US will have to leave eventually (I believe a democrat president after this idiot Bush will retreat). Only a total massacre can put Iraquis in line (temporarily), but that is something people will remember and finally try to push USA out again, while any pro-US government will be considered as traitors (even now they are).
US has a history of bad attempts to force a "democracy" (better like puppet-democracy) upon someone which in some occasion ends up like Vietnam, Afganistan or Iraq. This time however much more is at stake than previous attempts, as big portion of oil production can slip out of US hands.
I should probably mention a recent scandal with local Croatian pop bend "Magazin" which was very popular across Yugoslavia in 80's and in all ex-YU states in 90's. Their first (female) singer described on her blog recently method of creating songs (she now lives in California):
First a demo tune was produced, and lyrics afterwards. All bend members were required to write down few of their most popular hits. Then a band leader (which is now regarded as a famous and very good composer in Croatia) would mix parts of those songs (or even play some short sequences backwards!) until something good gets out. The result in fact was really good, as the band always had catchy tunes with great melodies (instant hits).
It can be called some sort of plagiarism, but it is obviously hard to come up with something simple and unique at the same time, so probably almost anyone does it. In other words there aren't as many musical geniuses around as industry requires (but they are copied most often, either living or long dead ones).
I also believe this is a reason why open source servers aren't being as rapidly developed as fast as desktop currently is. It is boring and unrewarding. Something not hard but requires much work to copy the bloat like Exchange. And You won't use it as a hobbyst, corporations will (without even saying thanks for it). Unfortunately this same work is crucial for adoption of Linux because linux environments can't penetrate small businesses until a reasonably good and Office/Outlook/AD-comatible software stack is available and standardly available in most distros (that's right, just one or two such suites have power to survive and become mainstream). While Samba are doing good work at windows protocol descrambling and AD and SMB interoperability, it is only part of the equation of providing alternatives to MS stack.
I believe this problem be dramatically changed when documentation that MS had to surrender to EU authorities lands in hands of open source developers. This is expected to happen very soon.
It shouldn't be hard to implement that level of integration anymore. Samba recent work focused on interoperability for AD/MS-Kerberos and recent samba3 versions (i.e. pam&nss winbind plugin) allow Linux machines to seamlessly integrate with AD accounts.
But Samba4 really will really come with their own AD server implementation (based on modified Heimdal and OpenLDAP). If FOSS Exchange clones and e-mail clients include interoperability with help of Samba research (depending of course on samba license compatibility) it will be a huge anti-lock-in hit against MS and their collaborative suite.
Protected video path is intended to make ripping harder with HD-DVD and Blu-Ray disc playback (and other new DRM'd content where encryption isn't *yet* broken).
DVD CSS scheme was cracked long time ago, so you can rip it with just ability to read files from a DVD disc and bypass all that region and encryption-codes crap. Exception can be "protection" schemes which intentionally corrupt ISO structure of the disc so drive(r) can "choke" on garbage data, but userspace filter-driver like AnyDVD or free specialised applications handle this well and certainly work on Vista (and will continue to work unless in an extremely unlikely case Microsoft decides to break almost any existing DVD player app as well as many other burning programs, various install DVD's or CD's, game copy protections etc., and blocks Vista users from reading plain files from a DVD - in which case we will just use Linux or windows XP to do the ripping).
First, only the 64-bit version can't load unsigned *kernel* drivers (it allows "userspace drivers" which don'tneed to be signed). Second, you don't need a driver to rip DVD, because everything can be done in userspace (with tools like DVDdecrypter).
I don't see why you couldn't rip your DVD under Vista. Vista "protections" target new, DRM'd content (that not many people will buy and won't be affected).
Mod the parent up, please. First informative post on the theme.
Yes, there is something here. I remember learning in my physics class about states that slowly transfer energy through tunneling to the other side (a double potential well with a barries - it was addressed on a class when WKB wave function approximation was the topic, IIRC). This just seems like thorough analysis of equivalent effect and how real world "perturbations" would affect it.
One big problem was probably that he wasn't American, but a Serb from Croatia (then part of Austro-Hungaria state). And Americans lke Edison always had better business talent. Tesla was ripped off by companies that today run U.S. (or even world) energy production.
Duh, less then electronvolt is enough to start breaking molecule bonds in tissue of organisms. This starts just above visible spectrum. But this's > 1000 THz if I remember correctly.
It is however possible that harm can be done even with much lower energies (=frequencies) but at higher intensity, but I'd say unlikely and very hard to proove if it happens.
Slashdot Mirror
There is one successful method. It can be deduced from onomatopeic writings (when people tried to write down how something sounds). Of course exact dialect is almost unrecoverable.
Btw. one familiar backronym example is RIP, originally "requiem in pace", now people assume that it's "rest in peace".
Almost any DOS game was kind of operating system. With exception of file system access, most of hardware was directly programmed.
So the real cause of sexual education programs is to scare people into not having sex at all because of viruses that cause cancer? I could imagine religious sects to think that those are a sign from God.
Unfurtunately, genital herpes, HIV, CMV, EBR, hepatitis and myriad of nasty bacterial infections still exist to scare teenagers.
Gaming problems are probably related to still experimental and less optimised 3D drivers for Vista's new graphic card driver model. 3D drivers for XP by ATI/Nvidia are usually full of nasty hacks, purpose of which is to speed up major gaming benchmarks (3Dmark, Quake, Doom, UT etc.). I guess many of those tricks aren't, well, portable (easily) to a new driver model. Vista can use XP drivers drivers though, but I guess this could work less than optimal.
As DWM shuts down whenever app locks front buffer (and direct3D games do that), I doubt that compositing is a reason why games perform slower. Another possibility is that scheduler classes (described in first part article) might affect CPU performance of games tweaked to run well with XP scheduling (e.g. I experienced hiccups with PES6).
Regarding article(s), I must say I'm not that impressed with amount of changes that got in their kernel since XP (except maybe by new driver framework which is a different story). In many areas mentioned in articles, Linux kernel is leaps and bounds ahead (scheduler,MM,IO-areas which saw huge refinements during 2.6 cycle and are constantly being improved). In others it is on-par (or almost here, like with PM or init mechanism). Exception seems to be Ready* stuff, but this is not really "a next big" thing anyway and seems as something that few decent kernel hackers could implement in a month or two. Also Vista now seems to have better graphic driver model (though DRI is closing the gap) and better userspace driver model (except they don't have FUSE equivalent).
True. But it's still easier to make a good management tool than a hypervisor or implement usual non-VT/Pacifica x86 virtualisation with good performance (the latter arcanely hard). So Vmware, which really excells at software engineering will still have to keep their virtualisation ahead of competition to be industry leader. Yes, they are now pressured by both Windows and Linux, but their package is ahead of both.
It's hard to guess how ahead Microsoft is with this, but i somehow doubt that it's as advanced as virtualisation that's (becoming) available on Linux (KVM, paravirt ops, etc.). I doubt that they are using full kernel as hypervisor (like KVM - this is IMO best long-term approach, though it takes time to pefect). In addition Linux also has good support for jail-like separation (Virtuozzo etc) which is even of bettr performance. Microsoft could be basing their stuff on Xen, along with acquired Connectix.
Maybe Microsoft is planning to take advance of control they have over graphic drivers - to implement 3D acceleration over paravirtualised interface (I'm sure they DON'T want this to run on any other host OS except Microsoft). As usual, they don't care about Windows running paravirtualised on any other OS, and probably doing steps to prevent this from happening anytime.
Also it's hard to believe that Microsoft cares about Vmware that much, more about Red Hat which will soon release RHEL5 with Xen support and Fedora with both Xen and KVM support.
For two processor cores having 1Mb cache each, it isn't that much (Prescott was single core).
Wonder why they still stick to P4 performance scheme, as that processor family is a history. Are they planning to improve on existing K8 design? Else they won't be able to catch-up with Intel in performance, especially as Intel has now maybe solved current leakage problem (and AMD usually lags behind Intel in production process). Currently they only have advantage in merging with ATI GPU tech as they're planning to produce combo GPU/CPU - it isn't clear whether it will be for low-end ultra-cheap machines or will it compete with mainstream gaming machines.
That was "Back Office", a product by Microsoft (IIRC, they renamed it later because of that trojan malware).
Unfortunately real democracy exists only on this level, especially in the USA, but in some way also in other countries. On "macroscopic" level there are only democrats and republicans. True oligarchy, USA is aristocratic country. (lesser parties can't even get ONE candidate into a parliament without having (relative) majority of votes in big pieces of US land, isn't that bad enough?)
Hackers seem to be able to get this info from Nv cards anyway (renouveau tool, Radeons already figured out),
so giving documentation to OSS developer under NDA won't cost Nvidia anything:
It will be figured out sooner or later and open source driver code produced - it's just a matter of time.
Yes, it's hard to write graphic card drivers, but you are underestimating community skill (just look at amount and quality of driver work already present in kernel). As I said, whole community working on commong driver framework core features can significantly reduce total amount of work required, and also make getting working drivers for future hardware much easier.
(btw. GLSL compiler will be in next Mesa)
Not true.
It may be hard to get something on reverse-engineered boards, but Intel example shows that good performance and advanced drivers are possible if company decides to pay skilled programmers to work on it (and give them proper documentation). OTOH, dirty performance hacks, often used by manufacturers at the price of compatibility and stability of drivers, aren't acceptable in open source, so performance might be somewhat lower (Windows gamers are lucky that they can't see what kind of driver garbage is installed on their machines).
Good side of open source is certainly that there is common framework which can be used across graphic hardware: this significantly reduces duplication and is beneficial to all users, not just those using particular hw. (and btw. Microsoft if trying to force manufacturers into something like that in Vista, their system is similar in concept to DRI). Unfortunately manufacturers don't like this: especially those which invest a lot in Linux drivers, like Nvidia. They want to differentiate, not give their work to other companies for free (and btw, there is also issue of various patented stuff, like S3TC).
Hos about "fake" reviews by gaming sites or hardware sites (many of them having a deal with publishers/manufacturers), or TV shows which present new cars?
It's all fraud marketing with lots of praising and hiding of facts.
NDA is OK if it contains contain only reasonable clauses like guarantee that no information regarding hardware will be revealed except the source code (withough huge comments). Though it is obvious that source code will reveal stuff. I believe companies are just afraid to simply release docs to anyone, as it is very easy to copy and produce identical clone. A driver probably won't access all, but rather a subset of interfaces. Instead of waiting for community to reverse engineer windows drivers, it is better for them to cooperate in this fashion, as there really isn't any real cost. Another nice option here is to cooperate with a company developer on the driver code.
It isn't a big issue for a FOSS (krenel) developer to sign a NDA with hardware manufacturers if he works on software only. For example, one Xorg driver developer signed NDA with ATI to develop 2D driver for X1300 and newer cards. (Unfortunately he still isn't allowed to release source).
As for sample devices, I don't think it is a problem to ship few of them under NDA, even hardware review sites get those.
U.S. didn't even stop pursuing the cold war, as they continued the arms race solely, to get a quick advantage over other emerging powers as China, ASAP. And now, when someone else tests an effective weapon which can significantly (and cheaply) cripple U.S. surveillance power, suddenly they cry for "disarmament" - but obviously of all but themselves, as they keep pumping $500 bln/yearly into the war machinery and R&D (not eve counting Iraq expenses!). How lame, this doesn't have anything to do with defence, it's an aggressive armament with purpose to wage wars like one in Iraq.
I hope their shiny world dominance model will collapse sooner or later.
Good thing is that those stabilised releases are much more similar to development versions than when 2.4 and 2.5 were state of art, so most fixes can easily be comitted to latest upstream version (or backported from it).
Turning experimental version of such a big project into a stable one is a long and painful job. It's order of magintude easier for Linus and kernel team to wait for a patchset to stabilise out of main tree until it's refactored, tested and good enough to go in. I believe both users and developers are happier with this (and improvements get a fair review, aren't just bumped in).
Whatever, it's their contry and it's not busineess of G.W.Bush nor US neocons to decide what will happen with it. Yes, people there maybe aren't as civilised as westerners (not including US population!), but they still are nationalists fignting rightfully for their soil.
Of course U.S. can't just leave this place with vast oil reserves, as it will once again turn into anti-US playground and probably later divide between factions, with bloody battles for oil field areas. As UN sanctions are already lifted, they will be free to do business with french, russian and other non-US oil companies, maybe even use Euro, which would be a pending disaster for U.S. economy. In other words, enemies of the U.S will get huge power in their hands.
US did loose Iran, and can't affort to loose another such country again. It is perfectly understandable why Bush is going to try to drawn this resistance in blood.
Noone believes that the engagement will be successful, instead US will have to leave eventually (I believe a democrat president after this idiot Bush will retreat). Only a total massacre can put Iraquis in line (temporarily), but that is something people will remember and finally try to push USA out again, while any pro-US government will be considered as traitors (even now they are).
US has a history of bad attempts to force a "democracy" (better like puppet-democracy) upon someone which in some occasion ends up like Vietnam, Afganistan or Iraq. This time however much more is at stake than previous attempts, as big portion of oil production can slip out of US hands.
I should probably mention a recent scandal with local Croatian pop bend "Magazin" which was very popular across Yugoslavia in 80's and in all ex-YU states in 90's. Their first (female) singer described on her blog recently method of creating songs (she now lives in California):
First a demo tune was produced, and lyrics afterwards. All bend members were required to write down few of their most popular hits. Then a band leader (which is now regarded as a famous and very good composer in Croatia) would mix parts of those songs (or even play some short sequences backwards!) until something good gets out. The result in fact was really good, as the band always had catchy tunes with great melodies (instant hits).
It can be called some sort of plagiarism, but it is obviously hard to come up with something simple and unique at the same time, so probably almost anyone does it. In other words there aren't as many musical geniuses around as industry requires (but they are copied most often, either living or long dead ones).
I also believe this is a reason why open source servers aren't being as rapidly developed as fast as desktop currently is. It is boring and unrewarding. Something not hard but requires much work to copy the bloat like Exchange. And You won't use it as a hobbyst, corporations will (without even saying thanks for it). Unfortunately this same work is crucial for adoption of Linux because linux environments can't penetrate small businesses until a reasonably good and Office/Outlook/AD-comatible software stack is available and standardly available in most distros (that's right, just one or two such suites have power to survive and become mainstream). While Samba are doing good work at windows protocol descrambling and AD and SMB interoperability, it is only part of the equation of providing alternatives to MS stack.
I believe this problem be dramatically changed when documentation that MS had to surrender to EU authorities lands in hands of open source developers. This is expected to happen very soon.
It shouldn't be hard to implement that level of integration anymore. Samba recent work focused on interoperability for AD/MS-Kerberos and recent samba3 versions (i.e. pam&nss winbind plugin) allow Linux machines to seamlessly integrate with AD accounts.
But Samba4 really will really come with their own AD server implementation (based on modified Heimdal and OpenLDAP). If FOSS Exchange clones and e-mail clients include interoperability with help of Samba research (depending of course on samba license compatibility) it will be a huge anti-lock-in hit against MS and their collaborative suite.
Protected video path is intended to make ripping harder with HD-DVD and Blu-Ray disc playback (and other new DRM'd content where encryption isn't *yet* broken).
DVD CSS scheme was cracked long time ago, so you can rip it with just ability to read files from a DVD disc and bypass all that region and encryption-codes crap. Exception can be "protection" schemes which intentionally corrupt ISO structure of the disc so drive(r) can "choke" on garbage data, but userspace filter-driver like AnyDVD or free specialised applications handle this well and certainly work on Vista (and will continue to work unless in an extremely unlikely case Microsoft decides to break almost any existing DVD player app as well as many other burning programs, various install DVD's or CD's, game copy protections etc., and blocks Vista users from reading plain files from a DVD - in which case we will just use Linux or windows XP to do the ripping).
First, only the 64-bit version can't load unsigned *kernel* drivers (it allows "userspace drivers" which don'tneed to be signed). Second, you don't need a driver to rip DVD, because everything can be done in userspace (with tools like DVDdecrypter).
I don't see why you couldn't rip your DVD under Vista.
Vista "protections" target new, DRM'd content (that not many people will buy and won't be affected).
Mod the parent up, please. First informative post on the theme.
Yes, there is something here. I remember learning in my physics class about states that slowly transfer energy through tunneling to the other side (a double potential well with a barries - it was addressed on a class when WKB wave function approximation was the topic, IIRC). This just seems like thorough analysis of equivalent effect and how real world "perturbations" would affect it.
One big problem was probably that he wasn't American, but a Serb from Croatia (then part of Austro-Hungaria state). And Americans lke Edison always had better business talent. Tesla was ripped off by companies that today run U.S. (or even world) energy production.
Duh, less then electronvolt is enough to start breaking molecule bonds in tissue of organisms. This starts just above visible spectrum. But this's > 1000 THz if I remember correctly.
It is however possible that harm can be done even with much lower energies (=frequencies) but at higher intensity, but I'd say unlikely and very hard to proove if it happens.