From the SAL website: F is a carefully crafted subset of the most recent version of Fortran, the world's most powerful numeric language. F retains the modern features of Fortran--modules and data abstraction, for example--but discards facilities such as EQUIVALENCE, which are difficult to teach, use, or debug.
Backwards compatibility is extremely important for the Fortran crowd (who tend to be a very conservative bunch). Having to rewrite source code is not going to make them happy.
I don't live in the United States. I live in Canada. I'm glad I do, because I think your government has gone crazy.
As a totally scatagorical example, what rules exist for bleach and/or Toilet Duck? These are things many people have in their bathrooms, but they're also precursers for chlorine gas (technically a chemical weapon). Can the FBI kick down your door for having bleach and Toilet Duck in your house?
Another example of knee-jerk lawmaking, or legislating from the crotch - however you want to refer to it.
You have the wrong end, chum, for as wiser sages than I have said, somewhere in that brain is the creative genius we used to know as John Romero. Labotomize the real culprit, the flooziest of floozies, the two-legged catalyst of Romero's downfall. Stevie Case, a.k.a. Killcreek.
She used him like a ladder. Is it his fault he had a few rotten rungs?
Correct, but more so in the opposite sense. Denying funding for a project can retard it's advancement. Kennedy and LBJ basically gave NASA a carte blanche which meant they could go as fast as their scientists and engineers would allow.
ATI has historically made good hardware that was crippled by buggy, poorly written, limited drivers.
Stop complaining about your 'historical' (i.e. anecdotal) experiences with ATI hardware. I own an R200 and am extremely satisfied with it (using DRI thank you very much). I haven't used the closed source driver, but feedback seems to be very positive.
If it's Windows your referring to, then ATI has extremely competitive, high quality drivers. The CATALYST crew claim they produce the industry standard in graphics drivers. In the wake of the 3dmark controversy, they may be right.
If you have a specific complaint about modern ATI products, then share it. Otherwise, shuddup.
This article seems to reiterate what everyone has been saying (Carmack, Valve, everyone). The GeforceFX architecture can only be made competitive for 3d engines using modern shaders with herculean effort. This is to be competitive, not dominantly superior.
Honestly, I thought nVidia learned their lesson with the NV1 - don't make weird hardware.
Now, what has to be making GeforceFX owners worried is Gabe Newell's warning that the new Detonator drivers might be making illegitimate 'optimizations' and, furthermore, covering them up by rendering high quality screen captures.
Of course, the shuttle never has flown a polar orbit, and SLC-6 at Vandenberg has it's own little hard-luck story (don't build your launch site on Indian burial grounds). The short of it is, the military got spooked about the reliability of the shuttle after Challenger blew up, decided it wasn't worth it to fix the problems at Slick-6, and have used Titans ever since. For the shuttle, that was a lot of very lucrative business lost.
Were it not for Challenger, the shuttle might have operated out of Vandenberg. What would public perception of the program be like if that were the case?
Here's a listing of all military launches using the shuttle.
The TV series "The Outer Limits" (the episode was "Final Exam"), a Good-Will-Hunting type Physics student invents and builds a cold fusion bomb using everyday components.
If I remember correctly, the fictional bomb had a pretty astounding yield - over 50 megatons, larger than the biggest H-Bomb. I believe the authorities would suppress knowledge of cold fusion for this reason much more readily than for the sake of the energy industry.
Hah! All you plotters in the Open Source conspiracy will never tell your lies to the FTC! SCO is on a divine mission, supported by God and the American Way! No one may get in our path!
All I'm saying is that RMS is proposing a highly idealized, theoretical solution to a practical problem, that it will not work, and that he is wrong to recommend it. I believe it is wrong to recommend any specialized GNU/Linux distribution for general use as it is not designed to be generally useful.
A more responsible statement (similar to the one he made later on regarding WINE) would be to recommend limiting non-free software use to an absolute minimum, and to attempt to reduce non-free software use over time. Additionally, he should have said that those persons whose software is close to compliance with Free Software guidelines should be petitioned and given an opportunity to come into compliance. These are practical ways to migrate people to free software.
I stand by my statements. No idea is the ultimate solution, and anyone who thinks so is extremely arrogant. If Stallman does not truly believe so then I apologize to him for my misread, but I'm surprised you have the gall to accuse me of apathy. You speak a lot about principles, but you don't talk at all about how to reach these principled objectives.
I hope you don't have much influence in the Free Software movement. If things were done your way, you'd have us in the same paragraph as Che Guevara under the title of 'noble failures', and everyone else using Windows.
I've been somewhat less vigilant in the past little while in that regard. This was a mistake.
My lack of vigilance, however, is not the point. Posting tubgirl links on slashdot is the digital equivalent of shoveling dog shit onto the sidewalk. It's someone's disgusting and bizarre amusement, and I just don't get how they get off on it.
Today I would recommend GNU/LinEx, the distribution prepared by the government of Extremadura, because that's the only installable distribution that consists entirely of free software.
Sorry, but where is Extremadura?
Non-free programs are not officially considered "part of Debian", but Debian does distribute them. The Debian web site describes non-free programs, and their ftp server distributes them. That's why we don't have links to their site on www.gnu.org.
Debian is a general purpose system. It is meant to provide as much functionality to as many people as possible. This is not as much of an issue for distrobution like LinEx - similarily Red Escolar and the BU distro - which can be (and, I presume, are) highly customized for the locales which use them. In the case of Extremadura, I presume the powers that be simply decided they didn't want/need non-free software. In the case of Debian, non-free (and contrib) software is a compromise which is required by the larger, more varied user base it supports.
RMS should be careful where he brushes his tar. Many software packages under non-free in Debian differ from FSF guidelines in only small ways. For example, the security auditor Bass has a license which is substantively the compliant with Debian Free Software Guidelines, with the exception of having to notify (not ask permission of) the author if you use his software in a commercial product or business.
If you're a commercial entity using Bass, you might email the author, tell him that his utility is cool and what you want to use it for. This would bring you into compliance. According to FSF and DFS guidelines, and therefore RMS, this makes Bass somehow immoral. I don't know the author's motivation for this (perhaps he's just interested), but I don't think he wants to prevent anyone from using his software.
For that matter, what if I created a product and licensed it in such a way that it was free, except it couldn't be used by 'repressive' governments? This 'ethical clause' would make my product non-free software, and RMS would therefore condemn it (although Amnesty International would not). However, I would consider this to be an improvement, as it would not only promote software freedom, but possibly physical and political freedom as well. Would this be wrong?
I agree with the fundamentals of the Free Software movement. I just find RMS's general arrogance repugnant. It stinks of radicalism; frankly, radicals scare me. --- --- A note on names. GNU/Linux is a product. Names like Debian, Redhat, etc. are brands. I may use brand names at times to refer to products. This is culturally common, acceptable, and comprehensible as well. Don't hold it against me.
I will give you a practical example. Say you have a packet filtering router, and you want to run a socks proxy on it. To prevent people from using the proxy to attack you from outside, you would put a firewall rule on the external interface blocking the socks port, while most of your firewall rules exist on the internal interface.
The DMZ in this case is the region in which the proxy sits (between the two interfaces). It is an area which is not totally secure, like your private network, but not totally open either. Typically it's the region where you place web caches, externally available servers, proxies and such.
OpenBSD and IP Tables allow you to assign a firewall rule to a particular interface. Bordermanager does not. Thus it is impossible to create a DMZ on a (single) router running Bordermanager. You must use two machines as firewalls and create the DMZ between them.
Their client/server stuff seems to be OK. I haven't used it that much. I've heard the IT people complain that it generates a constant datastream, even when idle, and nobody knows why.
Bordermanager is pure crap. It doesn't allow you to create a DMZ between your two NICs, and it will drop your connection like a stone unless you reboot all the time. I don't know how Novell can sell it.
I don't think IPX has any business existing anymore.
Funny that they aren't using Unixware for their server. Maybe it isn't so good after all. Maybe they've just liquidated their IT department to the point where nobody can do a switchover. With that thought in mind, here's a movie quote:
You fought for me?!! You manipulated me! Into where I am now - staring at the Brown & Williamson building, it's all dark except for the tenth floor. That's the legal department, that's where they fuck with my life!
Slashdot Mirror
From the SAL website:
F is a carefully crafted subset of the most recent version of Fortran, the world's most powerful numeric language. F retains the modern features of Fortran--modules and data abstraction, for example--but discards facilities such as EQUIVALENCE, which are difficult to teach, use, or debug.
Backwards compatibility is extremely important for the Fortran crowd (who tend to be a very conservative bunch). Having to rewrite source code is not going to make them happy.
OK, break out those optical cables.
they last "15,000 to 20,000 miles", which sounds pretty good to me.
No, that sucks. Your tires will probably last longer (and be cheaper too).
I don't live in the United States. I live in Canada. I'm glad I do, because I think your government has gone crazy.
As a totally scatagorical example, what rules exist for bleach and/or Toilet Duck? These are things many people have in their bathrooms, but they're also precursers for chlorine gas (technically a chemical weapon). Can the FBI kick down your door for having bleach and Toilet Duck in your house?
Another example of knee-jerk lawmaking, or legislating from the crotch - however you want to refer to it.
Half life was based off a source drop between Quake and Quake2. They added a ton of functionality before release (including a DirectX backend).
The Half life 2 engine is all original.
... [rimshot]
I think you are wrong. I quite distinctly remember the little girl saying that it was UNIX.
You have the wrong end, chum, for as wiser sages than I have said, somewhere in that brain is the creative genius we used to know as John Romero. Labotomize the real culprit, the flooziest of floozies, the two-legged catalyst of Romero's downfall. Stevie Case, a.k.a. Killcreek.
She used him like a ladder. Is it his fault he had a few rotten rungs?
Correct, but more so in the opposite sense. Denying funding for a project can retard it's advancement. Kennedy and LBJ basically gave NASA a carte blanche which meant they could go as fast as their scientists and engineers would allow.
ATI has historically made good hardware that was crippled by buggy, poorly written, limited drivers.
Stop complaining about your 'historical' (i.e. anecdotal) experiences with ATI hardware. I own an R200 and am extremely satisfied with it (using DRI thank you very much). I haven't used the closed source driver, but feedback seems to be very positive.
If it's Windows your referring to, then ATI has extremely competitive, high quality drivers. The CATALYST crew claim they produce the industry standard in graphics drivers. In the wake of the 3dmark controversy, they may be right.
If you have a specific complaint about modern ATI products, then share it. Otherwise, shuddup.
This article seems to reiterate what everyone has been saying (Carmack, Valve, everyone). The GeforceFX architecture can only be made competitive for 3d engines using modern shaders with herculean effort. This is to be competitive, not dominantly superior.
Honestly, I thought nVidia learned their lesson with the NV1 - don't make weird hardware.
Now, what has to be making GeforceFX owners worried is Gabe Newell's warning that the new Detonator drivers might be making illegitimate 'optimizations' and, furthermore, covering them up by rendering high quality screen captures.
Of course, the shuttle never has flown a polar orbit, and SLC-6 at Vandenberg has it's own little hard-luck story (don't build your launch site on Indian burial grounds). The short of it is, the military got spooked about the reliability of the shuttle after Challenger blew up, decided it wasn't worth it to fix the problems at Slick-6, and have used Titans ever since. For the shuttle, that was a lot of very lucrative business lost.
Were it not for Challenger, the shuttle might have operated out of Vandenberg. What would public perception of the program be like if that were the case?
Here's a listing of all military launches using the shuttle.
Just like every good series, the Outer Limits got remade.
The TV series "The Outer Limits" (the episode was "Final Exam"), a Good-Will-Hunting type Physics student invents and builds a cold fusion bomb using everyday components.
If I remember correctly, the fictional bomb had a pretty astounding yield - over 50 megatons, larger than the biggest H-Bomb. I believe the authorities would suppress knowledge of cold fusion for this reason much more readily than for the sake of the energy industry.
Did he have a spare particle accelerator and working knowledge of nuclear physics?
Hah! All you plotters in the Open Source conspiracy will never tell your lies to the FTC! SCO is on a divine mission, supported by God and the American Way! No one may get in our path!
-- "Baghdad" McBride
All I'm saying is that RMS is proposing a highly idealized, theoretical solution to a practical problem, that it will not work, and that he is wrong to recommend it. I believe it is wrong to recommend any specialized GNU/Linux distribution for general use as it is not designed to be generally useful.
A more responsible statement (similar to the one he made later on regarding WINE) would be to recommend limiting non-free software use to an absolute minimum, and to attempt to reduce non-free software use over time. Additionally, he should have said that those persons whose software is close to compliance with Free Software guidelines should be petitioned and given an opportunity to come into compliance. These are practical ways to migrate people to free software.
I stand by my statements. No idea is the ultimate solution, and anyone who thinks so is extremely arrogant. If Stallman does not truly believe so then I apologize to him for my misread, but I'm surprised you have the gall to accuse me of apathy. You speak a lot about principles, but you don't talk at all about how to reach these principled objectives.
I hope you don't have much influence in the Free Software movement. If things were done your way, you'd have us in the same paragraph as Che Guevara under the title of 'noble failures', and everyone else using Windows.
MHTGR?
I've been somewhat less vigilant in the past little while in that regard. This was a mistake.
My lack of vigilance, however, is not the point. Posting tubgirl links on slashdot is the digital equivalent of shoveling dog shit onto the sidewalk. It's someone's disgusting and bizarre amusement, and I just don't get how they get off on it.
Today I would recommend GNU/LinEx, the distribution prepared by the government of Extremadura, because that's the only installable distribution that consists entirely of free software.
Sorry, but where is Extremadura?
Non-free programs are not officially considered "part of Debian", but Debian does distribute them. The Debian web site describes non-free programs, and their ftp server distributes them. That's why we don't have links to their site on www.gnu.org.
Debian is a general purpose system. It is meant to provide as much functionality to as many people as possible. This is not as much of an issue for distrobution like LinEx - similarily Red Escolar and the BU distro - which can be (and, I presume, are) highly customized for the locales which use them. In the case of Extremadura, I presume the powers that be simply decided they didn't want/need non-free software. In the case of Debian, non-free (and contrib) software is a compromise which is required by the larger, more varied user base it supports.
RMS should be careful where he brushes his tar. Many software packages under non-free in Debian differ from FSF guidelines in only small ways. For example, the security auditor Bass has a license which is substantively the compliant with Debian Free Software Guidelines, with the exception of having to notify (not ask permission of) the author if you use his software in a commercial product or business.
If you're a commercial entity using Bass, you might email the author, tell him that his utility is cool and what you want to use it for. This would bring you into compliance. According to FSF and DFS guidelines, and therefore RMS, this makes Bass somehow immoral. I don't know the author's motivation for this (perhaps he's just interested), but I don't think he wants to prevent anyone from using his software.
For that matter, what if I created a product and licensed it in such a way that it was free, except it couldn't be used by 'repressive' governments? This 'ethical clause' would make my product non-free software, and RMS would therefore condemn it (although Amnesty International would not). However, I would consider this to be an improvement, as it would not only promote software freedom, but possibly physical and political freedom as well. Would this be wrong?
I agree with the fundamentals of the Free Software movement. I just find RMS's general arrogance repugnant. It stinks of radicalism; frankly, radicals scare me.
--- ---
A note on names. GNU/Linux is a product. Names like Debian, Redhat, etc. are brands. I may use brand names at times to refer to products. This is culturally common, acceptable, and comprehensible as well. Don't hold it against me.
Thanks, I really needed to see that.
Again.
People like you should be culled from the human race.
Try Doom Legacy. TCP/IP multiplayer (other things too).
I will give you a practical example. Say you have a packet filtering router, and you want to run a socks proxy on it. To prevent people from using the proxy to attack you from outside, you would put a firewall rule on the external interface blocking the socks port, while most of your firewall rules exist on the internal interface.
The DMZ in this case is the region in which the proxy sits (between the two interfaces). It is an area which is not totally secure, like your private network, but not totally open either. Typically it's the region where you place web caches, externally available servers, proxies and such.
OpenBSD and IP Tables allow you to assign a firewall rule to a particular interface. Bordermanager does not. Thus it is impossible to create a DMZ on a (single) router running Bordermanager. You must use two machines as firewalls and create the DMZ between them.
Their client/server stuff seems to be OK. I haven't used it that much. I've heard the IT people complain that it generates a constant datastream, even when idle, and nobody knows why.
Bordermanager is pure crap. It doesn't allow you to create a DMZ between your two NICs, and it will drop your connection like a stone unless you reboot all the time. I don't know how Novell can sell it.
I don't think IPX has any business existing anymore.
That is all.
Funny that they aren't using Unixware for their server. Maybe it isn't so good after all. Maybe they've just liquidated their IT department to the point where nobody can do a switchover. With that thought in mind, here's a movie quote:
-- Jeffrey Wigand, The Insider