You can write 95% of your code in a programmet friendly language. But the critical sections need to be in C, FORTRAN or Assembler and need to be very carefully optimized. That can give you x100 on the same architecture.
Your completely right that Perl, Python, and Ruby are pigs. Java looks a little like it was dropped on its head as a child. I don't know much about C#; it tastes too much like MSWindows for me to touch it.
While different programmers have different tastes in friends, modern functional languages actually have compilers that produce code competitive with C, and a garbage collector can be a performance advantage for some use cases (faster thread-local allocation and greater sharing of data structures). But for some things, ie most numerical code, you're right that you just can't beat a static language like C or Fortran.
Power is a technical term with a specific meaning, and any self-respecting nerd should know that. If you're talking about something else, you'd better qualify the term.
Your system sounds somewhat like our most of our state universities which are heavily subsidized for residents of the particular state. The federal government will then provide loans to cover most of the remaining cost which, for lower-income families, are interest free while the student is in school.
The cost your students pay does seem a little cheaper than at the average state school over here, though.
I chose street walking as my means of income
I hope that doesn't mean the same thing it does here...
Plenty of my friends co-op'ed in undergrad at a school on the semester plan (Purdue, FYI). I just made it out, and most of them should be done within the year, putting their total time at 4.5 to 5 years. I guess the only thing special about that school is that it's big enough that just about every required engineering class is offered every semester.
It's all well and good to do something just because you love doing it, but you still have to put food on the table. It's only pragmatic to consider the financial return on your education. Besides, if someone else thinks they'll be happier making 30% more in a job they don't enjoy vs. one they do, who are we to judge?
If people have no problems paying health insurance (if they can afford it), then why is it such a problem to pay taxes to fund a proper health system?
Why does "proper" mean socialized? The problem I personally have is that I'm pretty sure our government will bungle things more than the current insurance companies and hospitals who sort-of have to compete against each other
The real problem here is a horrible regulatory framework that restricts supply, minimizes competition, and promotes frivolous malpractice suits, all of which drive up costs. Why is it that when government screws up, the answer is more government?
...unfair...
Interesting choice of words. What is fair? Equal opportunity? Equal results? Is any of this even possible?
People think that s-expressions are a poor syntax. These people are wrong.
Seriously, if you give yourself the change to wrap your head around it, s-expressions are both elegant and powerful. Representing your code as a data structure is what makes lisp lisp. Take that away, and you might as well just use ML.
Better a hundred people at a billion each than zero for free.
Not at all. Currently various US government agencies estimate the value of a human life at 5-10 million USD. They use this number to figure out things like pollution limits and whether mandatory seatbelt installation is a good idea.
A billion dollars could save MANY lives if used some other way, ie converting a coal power plant to solar-thermal.
The article doesn't describe how the produced the estimate of 1%. If they just looked at the year-over-year change, the number could be meaningless as that might be within the normal variation/trend of energy consumption.
The method economists use in this situation is to look at the group that your changing (Indiana) and compare the change in energy consumption to a nearby control group (Illinois, Michigan, Ohio, Kentucky). You can then look at the RELATIVE changes to get a valid answer.
***
Ok, I just followed the link to the actual paper, and it looks like they used several Indiana counties that were on DST prior to the policy change as their control. So, yeah, their results look pretty valid. In conclusion: Down with DST!
The reason they don't create new dams or build new ecologically friendly power stations isn't because they can't - it's because it's more commercially viable to retain limited availability of these resources.
Well if clean energy can be sold for so much more than the cost to produce, get together with a couple hundred friends and dam up the local river/build a solar plant/raise 1000 windmills. If there was money to be made, then somebody would be doing it. When there's competition, capitalism works beautifully.
Now seriously, I don't know what the situation is like in Australia (where I assume you live), but there's basically nowhere left in the US for us to build hydroelectric dams. Solar is also still quite expensive. Actually though, you might come out ahead with those windmills. That's why other people 'round here are currently building them by the ass-ton.
I'm not a kernel developer, so please excuse me if any of this is wrong. I believe a spin lock is a way of saving "I need exclusive access to this part of the system".
A spinlock looks like "while(is_locked(lock));". Thus, it will "spin" until whatever other thread/process it is competing for resources with has finished using that resource. This is in contract to a mutex which will stop the thread seeking the lock in case of contention. A spin lock is good when you have multiple processors/cores and the amount of code under the lock is very short.
WRT the kernel, I guess a global spin lock means only one thread can be in a system call at a time, though there may be some more subtle distinctions than that.
Right...compared to Windows, where all I've ever had to do is plug in my monitor.
And installing a USB-serial adapter in linux is as easy as plugging it in while I still haven't figured out how to make it work in Windows. And the gnome NetworkManager applet puts wireless configuration in Windows (at least XP, I've never used Vista) to shame. Some things are easier under Linux and some are harder. Surprising huh?
I'm a linux newbie.
Back when I was a Linux newbie we had to manually edit our xfree86.conf files to get X working at all. Uphill. Both ways.
I can't believe you tried to justify that explanation with "It's not that hard".
Compared to how things worked a couple years ago, it's a piece of cake. The number of things that "Just work" under Linux is making very rapid progress, but if you want a machine that requires the least fiddling (and you can stomach Jobs-ian aesthetics) the answer is still to fork over the cash for a Mac.
Also, this is slashdot. If adding one line to a config file and typing one command sounds hard to you, you should probably be reading 4chan or commenting on youtube videos.
It's not that hard. You just add one line to your xorg.conf (something like "Virtual 2048 768" to the Display Subsection of your Screen section). Then you do a `xrandr --output VGA-0 --left-of DVI-0'. This gives you one X screen split across your two monitors. I have this working fine on multiple machines with the open source radeon drivers. The only hiccup from compiz is that if your total screen size if bigger than the maximum texture size of your card, you get some strange artifacts in the extra screen area. It's still quite usable, and a reasonable workaround is to just stick some window there pinned to all workspaces (I use an Eterm tailling/var/log/messages pinned using devilspie).
It used to be possible to run each monitor as a separate X screen using a little more hackery in the xorg.conf file. I thought that was nicer than using a single screen; however, Xorg broke that sometime in the past 10 months. Now trying to make that work (at least with the radeon drivers) will cause X to crash, which is really just pathetic.
Uh, you don't compile modules. The distribution vendor does.
actual real modularity on the binary level for true "modules"
If you want a stable kernel module ABI, that only matters for binary-only modules (which are a bad idea). See vmware for how source-distributed modules can work fairly painlessly.
meaning settings or whatnot that have to be compiled into the kernel, instead of being switches and modules that you can throw in and out of the kernel.
What are you talking about?
I don't think it's *all* modular,...as well as making it easier to swap stuff in and out.
Most vendors compile generic kernels with just about all functionality put into kernel modules. What more do you want than modprobe, rmmod? Pretty buttons?
so any increase in that helps by making the kernel easier to work on because you can have definite targets and functionality
If you want a micro-kernel, go use QNX, hack on herd, or watch as Linux slowly steps in that direction. Maybe read some of the various flame wars on the topic and consider why herd hasn't made any significant progress in 15 years.
In other words, actual driver modularity! So users can actually download and install drivers from off the intarwebz without having to compile them and Linux can actually, I dunno, be usable for 99% of users! Brilliant!
And what would be better, a kernel that you could simply include or not include certain modules without the need for compilation, making the kernel truly modular, and hot-swapping them in or out based on your needs.
If you're actually serious, (sarcasm is kind of hard to detect in plain text):
man modprobe. Since Linux 2.0.
Slashdot Mirror
Your completely right that Perl, Python, and Ruby are pigs. Java looks a little like it was dropped on its head as a child. I don't know much about C#; it tastes too much like MSWindows for me to touch it.
While different programmers have different tastes in friends, modern functional languages actually have compilers that produce code competitive with C, and a garbage collector can be a performance advantage for some use cases (faster thread-local allocation and greater sharing of data structures). But for some things, ie most numerical code, you're right that you just can't beat a static language like C or Fortran.
Power is a technical term with a specific meaning, and any self-respecting nerd should know that. If you're talking about something else, you'd better qualify the term.
That doesn't make sense. The power used by the chip will exactly equal the heat generated by the chip due to the law of conservation of energy.
Ok, that sounds reasonable.
Eh, static types can be useful sometimes, but they're not the cure-all to software bugs that some people seem to think.
There's nothing fundamental to lisp (aside from the CL spec...) the precludes using static types. These people even did it.
Your system sounds somewhat like our most of our state universities which are heavily subsidized for residents of the particular state. The federal government will then provide loans to cover most of the remaining cost which, for lower-income families, are interest free while the student is in school. The cost your students pay does seem a little cheaper than at the average state school over here, though.
I hope that doesn't mean the same thing it does here...
Plenty of my friends co-op'ed in undergrad at a school on the semester plan (Purdue, FYI). I just made it out, and most of them should be done within the year, putting their total time at 4.5 to 5 years. I guess the only thing special about that school is that it's big enough that just about every required engineering class is offered every semester.
It's all well and good to do something just because you love doing it, but you still have to put food on the table. It's only pragmatic to consider the financial return on your education. Besides, if someone else thinks they'll be happier making 30% more in a job they don't enjoy vs. one they do, who are we to judge?
Why does "proper" mean socialized? The problem I personally have is that I'm pretty sure our government will bungle things more than the current insurance companies and hospitals who sort-of have to compete against each other
The real problem here is a horrible regulatory framework that restricts supply, minimizes competition, and promotes frivolous malpractice suits, all of which drive up costs. Why is it that when government screws up, the answer is more government?
Interesting choice of words. What is fair? Equal opportunity? Equal results? Is any of this even possible?
Make that: (? (is there (part a best (of lisp))))
People think that s-expressions are a poor syntax. These people are wrong.
Seriously, if you give yourself the change to wrap your head around it, s-expressions are both elegant and powerful. Representing your code as a data structure is what makes lisp lisp. Take that away, and you might as well just use ML.
Not at all. Currently various US government agencies estimate the value of a human life at 5-10 million USD. They use this number to figure out things like pollution limits and whether mandatory seatbelt installation is a good idea.
A billion dollars could save MANY lives if used some other way, ie converting a coal power plant to solar-thermal.
Somebody has to pay for it one way or another...
That is insane. Do we sterilize people after we give them antibiotics? After they undergo surgery?
Also, your natural selection claim may not be really valid. People with AIDS do still have children.
The article doesn't describe how the produced the estimate of 1%. If they just looked at the year-over-year change, the number could be meaningless as that might be within the normal variation/trend of energy consumption.
The method economists use in this situation is to look at the group that your changing (Indiana) and compare the change in energy consumption to a nearby control group (Illinois, Michigan, Ohio, Kentucky). You can then look at the RELATIVE changes to get a valid answer.
***
Ok, I just followed the link to the actual paper, and it looks like they used several Indiana counties that were on DST prior to the policy change as their control. So, yeah, their results look pretty valid. In conclusion: Down with DST!
No one's said it yet...
So it seems like we have the following:
Well if clean energy can be sold for so much more than the cost to produce, get together with a couple hundred friends and dam up the local river/build a solar plant/raise 1000 windmills. If there was money to be made, then somebody would be doing it. When there's competition, capitalism works beautifully.
Now seriously, I don't know what the situation is like in Australia (where I assume you live), but there's basically nowhere left in the US for us to build hydroelectric dams. Solar is also still quite expensive. Actually though, you might come out ahead with those windmills. That's why other people 'round here are currently building them by the ass-ton.
A spinlock looks like "while(is_locked(lock));". Thus, it will "spin" until whatever other thread/process it is competing for resources with has finished using that resource. This is in contract to a mutex which will stop the thread seeking the lock in case of contention. A spin lock is good when you have multiple processors/cores and the amount of code under the lock is very short.
WRT the kernel, I guess a global spin lock means only one thread can be in a system call at a time, though there may be some more subtle distinctions than that.
And installing a USB-serial adapter in linux is as easy as plugging it in while I still haven't figured out how to make it work in Windows. And the gnome NetworkManager applet puts wireless configuration in Windows (at least XP, I've never used Vista) to shame. Some things are easier under Linux and some are harder. Surprising huh?
Back when I was a Linux newbie we had to manually edit our xfree86.conf files to get X working at all. Uphill. Both ways.
Compared to how things worked a couple years ago, it's a piece of cake. The number of things that "Just work" under Linux is making very rapid progress, but if you want a machine that requires the least fiddling (and you can stomach Jobs-ian aesthetics) the answer is still to fork over the cash for a Mac.
Also, this is slashdot. If adding one line to a config file and typing one command sounds hard to you, you should probably be reading 4chan or commenting on youtube videos.
It's not that hard. You just add one line to your xorg.conf (something like "Virtual 2048 768" to the Display Subsection of your Screen section). Then you do a `xrandr --output VGA-0 --left-of DVI-0'. This gives you one X screen split across your two monitors. I have this working fine on multiple machines with the open source radeon drivers. The only hiccup from compiz is that if your total screen size if bigger than the maximum texture size of your card, you get some strange artifacts in the extra screen area. It's still quite usable, and a reasonable workaround is to just stick some window there pinned to all workspaces (I use an Eterm tailling /var/log/messages pinned using devilspie).
It used to be possible to run each monitor as a separate X screen using a little more hackery in the xorg.conf file. I thought that was nicer than using a single screen; however, Xorg broke that sometime in the past 10 months. Now trying to make that work (at least with the radeon drivers) will cause X to crash, which is really just pathetic.
Hmm, of the three vowels with the right pronunciation, I managed to pick the one that was most wrong...
Uh, you don't compile modules. The distribution vendor does.
If you want a stable kernel module ABI, that only matters for binary-only modules (which are a bad idea). See vmware for how source-distributed modules can work fairly painlessly.
What are you talking about?
Most vendors compile generic kernels with just about all functionality put into kernel modules. What more do you want than modprobe, rmmod? Pretty buttons?
If you want a micro-kernel, go use QNX, hack on herd, or watch as Linux slowly steps in that direction. Maybe read some of the various flame wars on the topic and consider why herd hasn't made any significant progress in 15 years.
Yeah...[/sarcasm]
If you're actually serious, (sarcasm is kind of hard to detect in plain text): man modprobe. Since Linux 2.0.
For "local" devices, it seems to make more sense to use something like AoE vs. iSCSI
We tried that. Once we got Hoover. Next, we got Carter.
Wait, when did the Red Cross become evil?