I think the trouble with being a critical thinker, or an atheist, or a humanist is that you're right.
And it's quite hard being right in the face of people who are wrong without sounding like a fuckwit.
People go "do you think the vast majority of the world is wrong?"
Well yes... I don't know how to say that nicely, but yes.
If guns are illegal, then anyone who has a gun is a criminal, and you can prevent crime by just arresting everyone who has guns.
You can sure try to arrest us... Best wear your vest and bring bigger guns....
A) *woosh!*
B) Would you honestly die to retain posession of your firearms? Or are you just trying to impress all these internet strangers with how much of a man you are for being willing to stand up to a bunch of guys with assault rifles?
Academia is likely your best choice for ethical work, and many of them hire developers (though usually not for HPC). And HPC is generally not an easy field to break into without an internship. But what about some of the national research institutions? They do a great deal of useful "ethical" work, and they typically have a bigger budget for HPC than academia. This assumes you're willing to move (the relocation packages are good) because they're pretty sparse. Many positions require a grad degree, but they usually only require a BS and experience for programmer/developer jobs. Some of them do defense contracts too, so ask what projects you would be working on. A few places off the top of my head:
National Center for Atmospheric Research
National Renewable Energy Laboratory
National Oceanic and Atmospheric Administration
National Institute for Standards and Technology
National Center for Supercomputing Applications
National Center for Computational Sciences
National Energy Research Scientific Computing Center
San Deigo Supercomputer Center
There are quite a few others, but these are the ones I'm familiar with. It would be worth looking into these, or developer positions at a large university.
Every kid is going to be different, but Learning Perl was a great first programming book for me (Python wasn't as popular at the time, but that seems to be the prevailing opinion today). You get to learn basic programming concepts, see results right away, and not worry about the fine-grained details. The book's examples were really good, and string parsing is dead easy. And because string parsing was easy, I was able to code little games, AIM bots, and administration scripts for my web server pretty early on. Of course, I didn't learn to program *well* until I got into compiled languages, but the goal of the first language is usually just to understand the basic concepts which can be built upon. I guess the most important thing is to pick A) something easy to learn, and B) something you can use for your own projects.
If we can find a way to speed up the process of human evolution and artificially select desirable characteristics (and/or prevent undesirable characteristics), I think it would be immoral to not do so. My biggest concern with this specific example, however, is whether we can discern between HFA / Asperger's and low functioning autism... and where that line is.
That's the Old Testement. Chrisitianity is based of the New Testament.
That's great to hear! This means Christians can finally:
1. Cut the size of the bible in half
2. Do away with the ten commandments
3. Stop believing in creationism
4. Stop talking about how Noah crammed over 1,000,000 species on a boat
If you want to denounce the Old Testament, I fully support your decision. But you're going to have to cut out a good portion of your religion to do so.
If you're not willing to denounce it as a whole, then embrace the fact that your religion encourages rape, mass murder, and stoning disobedient children. The fact that you're picking only the "good" parts out to believe in tells me, most assuredly, that we do not get our morals from the bible.
According to "A Brief History of Time", as I recall, they wanted to nominate Einstein for his work on relativity. However, since it was not yet fully verified/accepted, they just nominated him based on his work for the photoelectric effect. But it sounded to me like relativity was the unofficial reason for his nomination.
I used Ubuntu for awhile, but I just switched back to Debian. Ubuntu tends to break when you do weird stuff, or run a vanilla kernel... and Debian is a nice middle ground. Fairly bare bones, easy to configure, but doesn't require you to compile almost everything. Gentoo was fun though when I had the time to mess with it.
I have gotten very comfortable and efficient with my Gnome 2 setup over the years, and I recently switched to xfce. The setup is almost identical and I've been very happy. All I had to do was customize my panels and apt-get a few add-ons (like the Orage clock/calendar and system monitor). Xfce has come a long way since I last used it.
Err... SPARC is still alive in the HPC and I/O world. No one's running to BestBuy to buy the latest T4 processor for their gaming rig, but the fastest supercomputer listed on http://top500.org/ happens to be a SPARC machine. It's not dead, it's just a niche market. That's like saying Cray is dead.
Mod parent up... Sandy Bridge can sustain 8 double precision FLOPS/cycle, as opposed to Magny-Cours 4 per cycle. Which means, if all you're doing is an embarrassingly parallel floating point problem, this 64 core Sandy Bridge cluster would get a theoretical peak of 712 GFLOPS (8 FPU's * 2.8 Ghz * 64 cores) as opposed to a 48-core Magny-Cours node with a peak of 403 GFLOPS (4 FPU's * 2.1 Ghz * 48 cores). But even a moderate amount of communication between nodes is going to greatly reduce your computational efficiency, so real world performance might be closer than you think between the 2 solutions, depending on your application.
I'm not really following the logic of changing your OS because you don't like the direction of the window manager (not specific to you, several have mentioned this)... why not stick with Gnome 2? It's nice, it works, it's stable, and it stays out of the way. I'm currently using Ubuntu 10.10 and in no hurry to upgrade because of the Unity mess. But I understand Gnome 2 is still an option for the latest release. If they ditch support for Gnome 2, I may switch back to Debian or another distro that gives the user more WM options. Isn't that the point of having multiple WM's available?
New video games are expensive entertainment products, and in a down economy... a very easy thing to do without. Ubisoft is taking a HUGE risk by making the DRM this restrictive on such a non-essential item, and I think their sales numbers will reflect that. There are hundreds of other options for a good driving/racing game. The one does look cool, but I certainly don't feel like I "have" to have it.
You can already sorta do this with rackspace, among others. The problem is... how many bloggers, etc expect to get slashdotted enough to justify the cost?
C# is required for pretty much any computer related degree these days.
Not true, it completely depends on the degree and the university. My interests have nothing to do with Microsoft platforms, and I've never been required to use C# or VB. My BS in CS is from a small state university. Even if C# is required for your degree, surely you can opt for a C++ or Java elective class? Knowing multiple languages is handy, if not essential.
When a foreign computer wins, the benchmark needs to be changed? Now that is gaming the system, American style.
No, actually... we've needed a better metric for a long time. Linpack only measures the performance of a very small subset of problems that these machines are used to solved. There are a lot more constraints to consider than just floating point performance these days - which is why we're starting to see new stuff pop up like http://www.green500.org/ and http://www.graph500.org/. Linpack is still relevant, but it's only part of the overall picture.
There is nothing useless about a supercomputer. Oak Ridge National Lab has over a billion dollar budget each year and huge portions of that budget relies on the availability of high performance computing resources. (Not to mention all the other national labs) HPC supports research in areas like energy conservation, new power sources, bioinformatics, material science, weapons simulations, engineering, and computer science. Applications range from freeing ourselves of fossil fuel reliance to designing materials to be used in [insert next big product]. HPC is the reason we don't need to do nuclear weapon testing anymore. HPC is the reason our grandkids will have a longer average lifespan. I can guarantee that these machines wouldn't be built for tens or hundreds of millions of dollars if they weren't being used. And I can guarantee that when the US regains #1, it won't be for the sake of being #1... it will be for the necessity of furthering science that benefits us all.
Slashdot Mirror
I think the trouble with being a critical thinker, or an atheist, or a humanist is that you're right.
And it's quite hard being right in the face of people who are wrong without sounding like a fuckwit.
People go "do you think the vast majority of the world is wrong?"
Well yes... I don't know how to say that nicely, but yes.
--Tim Minchin
Jesus, put the tape measures away, gentlemen.
If guns are illegal, then anyone who has a gun is a criminal, and you can prevent crime by just arresting everyone who has guns.
You can sure try to arrest us... Best wear your vest and bring bigger guns....
A) *woosh!*
B) Would you honestly die to retain posession of your firearms? Or are you just trying to impress all these internet strangers with how much of a man you are for being willing to stand up to a bunch of guys with assault rifles?
Academia is likely your best choice for ethical work, and many of them hire developers (though usually not for HPC). And HPC is generally not an easy field to break into without an internship. But what about some of the national research institutions? They do a great deal of useful "ethical" work, and they typically have a bigger budget for HPC than academia. This assumes you're willing to move (the relocation packages are good) because they're pretty sparse. Many positions require a grad degree, but they usually only require a BS and experience for programmer/developer jobs. Some of them do defense contracts too, so ask what projects you would be working on. A few places off the top of my head:
National Center for Atmospheric Research
National Renewable Energy Laboratory
National Oceanic and Atmospheric Administration
National Institute for Standards and Technology
National Center for Supercomputing Applications
National Center for Computational Sciences
National Energy Research Scientific Computing Center
San Deigo Supercomputer Center
There are quite a few others, but these are the ones I'm familiar with. It would be worth looking into these, or developer positions at a large university.
Every kid is going to be different, but Learning Perl was a great first programming book for me (Python wasn't as popular at the time, but that seems to be the prevailing opinion today). You get to learn basic programming concepts, see results right away, and not worry about the fine-grained details. The book's examples were really good, and string parsing is dead easy. And because string parsing was easy, I was able to code little games, AIM bots, and administration scripts for my web server pretty early on. Of course, I didn't learn to program *well* until I got into compiled languages, but the goal of the first language is usually just to understand the basic concepts which can be built upon. I guess the most important thing is to pick A) something easy to learn, and B) something you can use for your own projects.
If we can find a way to speed up the process of human evolution and artificially select desirable characteristics (and/or prevent undesirable characteristics), I think it would be immoral to not do so. My biggest concern with this specific example, however, is whether we can discern between HFA / Asperger's and low functioning autism... and where that line is.
That's the Old Testement. Chrisitianity is based of the New Testament.
That's great to hear! This means Christians can finally:
1. Cut the size of the bible in half
2. Do away with the ten commandments
3. Stop believing in creationism
4. Stop talking about how Noah crammed over 1,000,000 species on a boat
If you want to denounce the Old Testament, I fully support your decision. But you're going to have to cut out a good portion of your religion to do so.
If you're not willing to denounce it as a whole, then embrace the fact that your religion encourages rape, mass murder, and stoning disobedient children. The fact that you're picking only the "good" parts out to believe in tells me, most assuredly, that we do not get our morals from the bible.
Businesses that cheat and copy... students that cram and collaborate on solo work... I think you just described the US.
I think you're right - its been awhile since I've read it and I don't have a copy handy at the moment, so I probably butchered the claim.
According to "A Brief History of Time", as I recall, they wanted to nominate Einstein for his work on relativity. However, since it was not yet fully verified/accepted, they just nominated him based on his work for the photoelectric effect. But it sounded to me like relativity was the unofficial reason for his nomination.
Seems like KDE is still very actively developed, but you have to go out of your way (Kubuntu) to even use it.
sudo apt-get install kubuntu-desktop
I used Ubuntu for awhile, but I just switched back to Debian. Ubuntu tends to break when you do weird stuff, or run a vanilla kernel... and Debian is a nice middle ground. Fairly bare bones, easy to configure, but doesn't require you to compile almost everything. Gentoo was fun though when I had the time to mess with it.
I have gotten very comfortable and efficient with my Gnome 2 setup over the years, and I recently switched to xfce. The setup is almost identical and I've been very happy. All I had to do was customize my panels and apt-get a few add-ons (like the Orage clock/calendar and system monitor). Xfce has come a long way since I last used it.
...burn it to an optical disc, then shred the disc! :)
Err... SPARC is still alive in the HPC and I/O world. No one's running to BestBuy to buy the latest T4 processor for their gaming rig, but the fastest supercomputer listed on http://top500.org/ happens to be a SPARC machine. It's not dead, it's just a niche market. That's like saying Cray is dead.
Oops, math fail, thanks for the correction! Looks like I used 4 FLOPS/Hz in my calculation even though I quoted 8.
Mod parent up... Sandy Bridge can sustain 8 double precision FLOPS/cycle, as opposed to Magny-Cours 4 per cycle. Which means, if all you're doing is an embarrassingly parallel floating point problem, this 64 core Sandy Bridge cluster would get a theoretical peak of 712 GFLOPS (8 FPU's * 2.8 Ghz * 64 cores) as opposed to a 48-core Magny-Cours node with a peak of 403 GFLOPS (4 FPU's * 2.1 Ghz * 48 cores). But even a moderate amount of communication between nodes is going to greatly reduce your computational efficiency, so real world performance might be closer than you think between the 2 solutions, depending on your application.
I'm not really following the logic of changing your OS because you don't like the direction of the window manager (not specific to you, several have mentioned this)... why not stick with Gnome 2? It's nice, it works, it's stable, and it stays out of the way. I'm currently using Ubuntu 10.10 and in no hurry to upgrade because of the Unity mess. But I understand Gnome 2 is still an option for the latest release. If they ditch support for Gnome 2, I may switch back to Debian or another distro that gives the user more WM options. Isn't that the point of having multiple WM's available?
New video games are expensive entertainment products, and in a down economy... a very easy thing to do without. Ubisoft is taking a HUGE risk by making the DRM this restrictive on such a non-essential item, and I think their sales numbers will reflect that. There are hundreds of other options for a good driving/racing game. The one does look cool, but I certainly don't feel like I "have" to have it.
You can already sorta do this with rackspace, among others. The problem is... how many bloggers, etc expect to get slashdotted enough to justify the cost?
C# is required for pretty much any computer related degree these days.
Not true, it completely depends on the degree and the university. My interests have nothing to do with Microsoft platforms, and I've never been required to use C# or VB. My BS in CS is from a small state university. Even if C# is required for your degree, surely you can opt for a C++ or Java elective class? Knowing multiple languages is handy, if not essential.
When a foreign computer wins, the benchmark needs to be changed? Now that is gaming the system, American style.
No, actually... we've needed a better metric for a long time. Linpack only measures the performance of a very small subset of problems that these machines are used to solved. There are a lot more constraints to consider than just floating point performance these days - which is why we're starting to see new stuff pop up like http://www.green500.org/ and http://www.graph500.org/. Linpack is still relevant, but it's only part of the overall picture.
Regardless of its perceived usefulness, I'd take weapons simulations over weapons testing any day.
Where is the "like" button?
There is nothing useless about a supercomputer. Oak Ridge National Lab has over a billion dollar budget each year and huge portions of that budget relies on the availability of high performance computing resources. (Not to mention all the other national labs) HPC supports research in areas like energy conservation, new power sources, bioinformatics, material science, weapons simulations, engineering, and computer science. Applications range from freeing ourselves of fossil fuel reliance to designing materials to be used in [insert next big product]. HPC is the reason we don't need to do nuclear weapon testing anymore. HPC is the reason our grandkids will have a longer average lifespan. I can guarantee that these machines wouldn't be built for tens or hundreds of millions of dollars if they weren't being used. And I can guarantee that when the US regains #1, it won't be for the sake of being #1... it will be for the necessity of furthering science that benefits us all.