It can be done easily, people out there are doing it... if we cant find a way to truly fix it we might as well spread it around and learn to live with it.
What is your point? These processors do what they are designed to do... x86 arent what they are because of the ISA, the underlying architecture with any wide superscalar machine has bloated so much that a little extra ISA translation on top wont do jack diddly.
PPC is fine for embedded, but dont kid yourself it aint the be all and end all of low power... its only a little better than x86 compared to whats out there. Id never imagine using it for portable applications, theres just way too many far better alternatives.
x86 has shown one thing, having a poorly designed ISA is only a small bump in the road towards performance. Unless you are an assembly hacker you shouldnt care about the ISA, it just makes sense to concentrate the market onto one ISA... we would not have the present price performance ratio's without it.
Its a pity it had to be x86 but its a done deal now, for very highly parallel SMT machines small register sets might actually be an advantage in the future:)
You can keep your One World Fragmented Market, I rather have the One World One ISA one where effective competition is not only possible but present and working.
Thats a bit of a strange arguement... if you attacked the assertion that GPL is free software Id go along with it, but only the GPL insures source in all its incarnations will stay open. When I think about open source I think about being able to see the source, not use it whichever way I like including closing it off and selling it.
Why a buesiness would hire coders to produce any software under licenses which allow free redistribution is beyond me.
Oh its easy to see how you can make more money using BSD licensed software, but wether you can make more money producing it... I doubt it.
At least with GPL, if the software is your own entirely, there's an incentive for others to come to you for a traditional license. With BSD they just jack it off the net and run with it if they need it for a closed source product.
I think software patents are a bit of a misnomer... algorithm patents would better describe the concept, wether they are implemented in hardware or software is entirely besides the point (and courts will not recognise the difference IMO, because that would make it trivial to avoid any algorithm patent by implementing enough of it in low level configurable hardware).
Social security is not a pyramid scheme, if you really want to stretch the metaphor you could call it an inverted pyramid scheme. I dont think its a scam, Im just a socialist.
They are making high bandwith profiles of MPEG-4 to replace even MPEG-2 for archival use ya know.
Its not about how much data you discard, its about how visually important that data is. MPEG-4 discriminates a lot better in what to throw away as MPEG-1.
No it isnt... divx is the preferred format for ripping. OpenDivx is just a basic implementation of MPEG-4 v1. The rest of the world is moving on, m$ (of which divx was ripped) and 3ivx are at version 3 for instance.
OpenDivx is missing a lot of advanced compression features, and its going to get sued to hell for contributory patent infringement if they get succesfull.
Even with two files and without hiding data in the names you can split data between the decompressor and the file and the the ratio of sizes between the two is an extra bit of information not expressed in the sum of sizes, with a sufficiently large data file this will always allow you to solve the challenge exactly as originally stated.
The entire en/de-cryption process has to take place in a "secure" environment... and a PC will never be remotely that. Now if the Ibutton could take data and en/de-crypt it and pass it back, that would be nice.
All this text manages to disguise the fact that in essence the electronic publication and peer review costs (for which you lacked to mention how it would be done) would be next to nothing and that its a money making scheme for the organisation who runs it and to a lesser extent authors.
Apart from the fact that it makes authors whore themselves out to try to get maximum page hits I only see some cosmetic differences to what we have today.
Dissemination is too integral a part of scienctific research for it to be left up to private buessinuess.
Where are the big costs with running a peer review process and putting stuff on the web? Couldn't an academic community as a whole in any given area easily carry it if they banded together?
Bandwith and storage are cheap, people are of course expensive... but if its true that peer review is usually voluntary there are no costs there, what remains is the organisation behind it. After initial costs a lot of the system should be able to be automated, leaving the task of assigning peers. I think costs should be manageable.
Like suggested here: http://www.exploit-lib.org/issue5/peer-review/
Its of course a pipe dream, the academia working together in this day and age... what a laugh.
Everyone knows it, but why do people always keep quoting it when they know perfectly well that its faults have long since been exposed and corrected in other microkernels? Is it perhaps that they use the misery that is Mach to spread FUD about microkernels?
It will probably boot from prom, and I wouldnt be surprised if the prom's contents were signed (with NVIDIA putting a check in hardware in the chipset). In which case I wouldnt give you a chance in hell of breaking it that way.
Perhaps you could do something like abuse a bufferoverflow/backdoor in an authorized application to load Linux into memory after boot though, but it aint cut and dry. Reverse engineering the hardware would be a bitch too.
Slashdot Mirror
You just write a wrapper DLL for OpenGL and catch the API calls and modify them... very easy, you could do it in VB.
It can be done easily, people out there are doing it... if we cant find a way to truly fix it we might as well spread it around and learn to live with it.
I wouldnt mind seeing some good competition for win32 OS's (Linux+Wine aint it).
What is your point? These processors do what they are designed to do... x86 arent what they are because of the ISA, the underlying architecture with any wide superscalar machine has bloated so much that a little extra ISA translation on top wont do jack diddly.
PPC is fine for embedded, but dont kid yourself it aint the be all and end all of low power... its only a little better than x86 compared to whats out there. Id never imagine using it for portable applications, theres just way too many far better alternatives.
ISA for the personal computing mass market.
:)
x86 has shown one thing, having a poorly designed ISA is only a small bump in the road towards performance. Unless you are an assembly hacker you shouldnt care about the ISA, it just makes sense to concentrate the market onto one ISA... we would not have the present price performance ratio's without it.
Its a pity it had to be x86 but its a done deal now, for very highly parallel SMT machines small register sets might actually be an advantage in the future
You can keep your One World Fragmented Market, I rather have the One World One ISA one where effective competition is not only possible but present and working.
Thats a bit of a strange arguement... if you attacked the assertion that GPL is free software Id go along with it, but only the GPL insures source in all its incarnations will stay open. When I think about open source I think about being able to see the source, not use it whichever way I like including closing it off and selling it.
Why a buesiness would hire coders to produce any software under licenses which allow free redistribution is beyond me.
Oh its easy to see how you can make more money using BSD licensed software, but wether you can make more money producing it... I doubt it.
At least with GPL, if the software is your own entirely, there's an incentive for others to come to you for a traditional license. With BSD they just jack it off the net and run with it if they need it for a closed source product.
I think software patents are a bit of a misnomer... algorithm patents would better describe the concept, wether they are implemented in hardware or software is entirely besides the point (and courts will not recognise the difference IMO, because that would make it trivial to avoid any algorithm patent by implementing enough of it in low level configurable hardware).
Social security is not a pyramid scheme, if you really want to stretch the metaphor you could call it an inverted pyramid scheme. I dont think its a scam, Im just a socialist.
Nuff Said.
How about 447 THz? :)
They are making high bandwith profiles of MPEG-4 to replace even MPEG-2 for archival use ya know.
Its not about how much data you discard, its about how visually important that data is. MPEG-4 discriminates a lot better in what to throw away as MPEG-1.
No it isnt... divx is the preferred format for ripping. OpenDivx is just a basic implementation of MPEG-4 v1. The rest of the world is moving on, m$ (of which divx was ripped) and 3ivx are at version 3 for instance.
OpenDivx is missing a lot of advanced compression features, and its going to get sued to hell for contributory patent infringement if they get succesfull.
Yet another chapter in the never ending conspiracy to give microkernels a bad name... please stop reviving that mess.
The wording of challenge allows it to be easily solved... rewording it after the fact is indeed slimey.
Even with two files and without hiding data in the names you can split data between the decompressor and the file and the the ratio of sizes between the two is an extra bit of information not expressed in the sum of sizes, with a sufficiently large data file this will always allow you to solve the challenge exactly as originally stated.
The entire en/de-cryption process has to take place in a "secure" environment... and a PC will never be remotely that. Now if the Ibutton could take data and en/de-crypt it and pass it back, that would be nice.
All this text manages to disguise the fact that in essence the electronic publication and peer review costs (for which you lacked to mention how it would be done) would be next to nothing and that its a money making scheme for the organisation who runs it and to a lesser extent authors.
Apart from the fact that it makes authors whore themselves out to try to get maximum page hits I only see some cosmetic differences to what we have today.
Dissemination is too integral a part of scienctific research for it to be left up to private buessinuess.
Please tell me this is a troll...
Its a correlation not a direct relation.
Where are the big costs with running a peer review process and putting stuff on the web? Couldn't an academic community as a whole in any given area easily carry it if they banded together?
Bandwith and storage are cheap, people are of course expensive... but if its true that peer review is usually voluntary there are no costs there, what remains is the organisation behind it. After initial costs a lot of the system should be able to be automated, leaving the task of assigning peers. I think costs should be manageable.
Like suggested here: http://www.exploit-lib.org/issue5/peer-review/
Its of course a pipe dream, the academia working together in this day and age... what a laugh.
You dont have to run all services for a microkernel in different protection domains, but its nice to have the choice of doing it when you want to.
Everyone knows it, but why do people always keep quoting it when they know perfectly well that its faults have long since been exposed and corrected in other microkernels? Is it perhaps that they use the misery that is Mach to spread FUD about microkernels?
It will probably boot from prom, and I wouldnt be surprised if the prom's contents were signed (with NVIDIA putting a check in hardware in the chipset). In which case I wouldnt give you a chance in hell of breaking it that way.
Perhaps you could do something like abuse a bufferoverflow/backdoor in an authorized application to load Linux into memory after boot though, but it aint cut and dry. Reverse engineering the hardware would be a bitch too.
Things like Handel-C and Esterel will allow ordinary programmers to have something up and running relatively quickly though.
Although explicitly parallel programming occam style is not something most programmers are very good at either.
Has anyone found a way to crack that with quantum computers?
Can anything be proven about the possibility of such a "quantum algorithm" existing if not?