The docs only talked about 'Copy on write' mechanisms, but does it also do 'Copy on read'?
I believe it does. They call it page replication. When a CPU on node 1 fetches data from memory on node 2, the system replicates the data pages from node 2 onto node 1, if free memory is available for them. It's all automatic. In this way, main memory is almost used like a level 4 cache.
Do all NUMA machines do that?
By now, they might, but when the O2000 was first released, SGI made a very big deal about this feature.
Based on your resume, you obviously have inside information that I'm not privy to. But I think you're oversimplifying the story a bit.
The Origin 3000 series (SN-1) was designed very nearly from the beginning to accommodate either MIPS CPUs from the R10000 family or IA-64 CPUs. In 1996-1997, when I worked most closely with the SN-MIPS and SN-IA groups, there was doubt about whether the IA-64 processor would be Merced or McKinley, but there was no question about support for one or both of them.
Look inside a C-brick some time. (If you don't have one handy-- heh-- there's an illustration here.) See all that empty space at the front? The original design called for the use of either MIPS PIMMs or IA-64 PIMMs. The IA-64 PIMMs would include all the necessary hardware to make the Intel chips talk to the Bedrock memory controller. The MIPS PIMMs are pretty small, about four inches square or so. But the IA-64 PIMMs were projected to be real monsters with giant heat sinks on them. Thus all that empty space in the C-brick.
For quite a while, of course, SGI has been working on SN-2, or whatever they're calling the successor to SN-MIPS these days. I'm not associated with that group any more, so I'm not in the loop on the new design. I've heard rumors on the order of 128 MIPS processors in a rack, quadrupling the processor density of the 3000 series systems, but I don't have any real information there. It's certainly possible that SGI is preparing to roll out their IA-64 systems in the spring in the new architecture, but that would surprise me. Of course, like I said before, you seem to know more than I do on this one.
I won't argue with you about it, but I'm not going to email you either. I'll just take your word for it and admit that you learn something new every day.
During development of Windows NT 3.1, the first version (god bless starting counting at 3) MS made a strong pitch to SGI to get behind it and work with them to replace IRIX. SGI turned it down, and later signed up for limited workstation production.
If true, this is the first I've heard of this. Can you back this up with some sort of evidence?
You say that synthetic benchmarks are irrelevant. Then you go on to say that this particular synthetic benchmark is highly relevant.
No, I don't. I really can't emphasize this enough: read. I said, "SPEC is synthetic and irrelvant." Big difference.
I'd like to see this run a TPC variant, which is closer to real-world than it is synthetic.
The TPC benchmarks are measurements of database performance. Since SGI was trying to demonstrate the features and capabilities of their hardware, it would have been completely inappropriate for them to use a database benchmark. STREAM TRIAD is great because it measures only one thing: the rate at which data can be moved from memory to the CPU or vice versa. The TPCs measure aggregate systems, including hardware, storage, OS, database software, and so on. They may be relevant if you're looking for a fast database server system, but they're hardly useful for evaluating one hardware architecture over another.
What, do I have to explicitly call out the components and subcomponents? It is a memory bus, for the purpose of this discussion.
The whole point of this discussion is that the SGI system can outperform virtually everything else on STREAM TRIAD because it has no memory bus. Memory busses are bottlenecks, and pumping a lot of data through them is very hard. The SGI system eliminates the bottleneck and thus demonstrates amazing bandwidth. When you miss the whole point of the discussion, I'm going to call you on it.
Back in the early days of Windows NT, it was not know what its capabilities would be. SGI nearly bought the farm by betting NT to replace IRIX.
Then SGI realized NT wasn't going to be for big machines, and let that bad dream fade away.
Dude, that's simply not true. SGI never even built a prototype Windows system that ran any version of NT before 4.0. By the time their Windows NT workstations made it out the door, Windows 2000 was very nearly a reality. So close that rather than adding support for certain features, SGI just punted. For example, dual monitor support was never offered on the NT systems until Windows 2000 came out, because NT 4.0 didn't support running two graphics pipes with different drivers.
SGI never, ever, spent any time or effort on Windows NT for anything other than workstations.
But likewise, finding x86 hardware with more processors is probably the largest reason that x86 Linux, Windows, whatever isn't running on bigger machines.
So, let me get this straight. They don't make big x86 boxes, and that's "probably the largest reason" why there are no OSs for big x86 boxes? Brilliant!
Good. The machine looks like it has a decent memory bus, and memory modules with a good configuration and speed rating.
You know, before you piss in SGI's Cheerios, you might want to do a little reading. The Origin 3000 architecture, on which this prototype system was based, has no memory bus at all. It uses a fabric of switched multi-gigabyte-per-second interconnects to attach CPUs to RAM and to other CPU nodes.
CPU benchmarks (like SPEC) are synthetic and irrelevant, because they fit in cache. Virtually no real application fits in cache, and the sort of applications you run on a machine this big deal with data sets no the order of tens or even hundreds of gigabytes. Memory-to-CPU bandwidth is probably the only real indicator of the ability of the system to handle real-world workloads.
It's also the only thing-- other than the dimensions and the color of the plastics-- that differentiates SGI's big Itanium 2 server from everybody else's big Itanium 2 servers.
It's not clear to what extent application programs have to be aware of this. Clearly, if you lay things out in memory badly, with lots of CPUs reading and writing the same memory from all over the memory net, the system will bottleneck.
Speaking as somebody who's done his share of IRIX programming, I'd say "none at all."
In some cases, on Origin 2000 hardware with older versions of IRIX, you could see notable performance differences if you went out of your way to place memory in banks adjacent to the running processors. But the Origin 3000 architecture, with its significant reductions in memory latency, and newer versions of IRIX, with their improved page replication algorithms, have made manual memory placement almost obsolete. Almost.
SGI spent a lot of time and trouble trying to reduce the impact of accessing remote memory. The caching mechanisms and page replication stuff are really well thought-out.
Anybody else see that as the main reason this is running Linux instead of Irix?
SGI started working on porting IRIX to the IA-64 architecture back in (I think it was) 1995 or 1996. Not long after, they found that it would be easier and cheaper to get Linux to scale more efficiently and to port some key libraries and services from IRIX than it would be to port all of IRIX over to the new architecture.
Add to this of course that Itanium2 is hardly a vector processor, so what they're doing is comparing apples to oranges.
Ordinarily I'm all for spoiling benchmarks by pointing out ways in which they're not applicable, but in this case, you're wrong. This test measures bandwidth into and out of main memory. That's it. It makes no difference whether the processors have vector registers and instructions or not. Noting matters except the factors that contribute to moving data between main memory and the CPUs.
Since this machine is a standard Origin 3000 with McKinley processor modules, I'm going to assume the answer will be yes. You can partition an O3000 down to a single processor brick + base IO brick, so I imagine that SGI will implement the necessary software bits to make that happen on the SN1-IA systems. I know there are both user space bits (mkpart, partmgr) and kernel space bits (the TCP-over-NUMAlink driver).
I personally have only seen partitioning used on HA systems and lab systems. For a fully fault-tolerant N-processor system, you can buy one 2N-processor Origin and partition it down the middle. The two nodes can run in parallel, passing data back and forth over the NUMAlink via TCP/IP, until one goes down. Also, partitioning is great in a lab environment. It's nice to be able to carve up a big multiprocessor system and give each user a 4-processor (or multiple of 4) node.
I wonder what linux apps would someone run on a system this big?
Anything you'd run on an IRIX system of that size, I'd imagine. I believe-- not positive-- that MSC has already released Nastran for Itanium 2 Linux. (Nastran is a computer-aided engineering tool used extensively in the automotive industry, and other manufacturing industries. It's used for things like stress, heat transfer, and vibration analysis.)
And, as long as the Fortran compilers are worth a damn, you can run just about any other scientific, analytical, or technical software, I'd imagine.
...interesting that SGI chose the Cray C90 - a system released in *1991* - to compare against. It's nice to know that it's only taken them 10+ years to catch up.:)
If you read the STREAM TRIAD web site linked above, you'll see that SGI didn't compare itself to the C90 exactly; it just ran a benchmark and published the results. Also in that approximate rank are other machines from NEC and Cray and, further down, Sun.
But you're right. Cray was way ahead of their time when it came to things like memory bandwidth. I remember a friend (ex-Crayon) telling me once that access to main memory on the T-90 was faster than access to the on-chip cache on the Pentium III. That sounds implausible, though, so he might have been exaggerating.
I'm curious how the new X1 (nee SV2) does on the STREAM suite.
The last word I got is that X1 is still in the PCB design phase. It's only running as a simulator right now. So it'll be a while before you see those numbers.;-)
(That info is several months old, so I may be wrong.)
I would be very interested to know what version of the kernel they are using.
I tried really hard to find that info this morning before submitting, but to no avail. But the test was demonstrated at the Intel Developer's Conference, according to the press release, so maybe we could find somebody who knows somebody?
Am I wrong about what this benchmark means? Or am I missing something basic?
With no disrespect intended, I think you might be missing something basic.
Any activity that involves moving data into and out of RAM will benefit from the ability to do it faster. That includes things such disparate things as database processing (if you're lucky, you can cache your indices in RAM), media encoding, hell, even compiling. Memory bandwidth is one of the few aspects of computer design that touches just about every application, with the exception of those that are small enough-- or sufficiently well optimized-- to fit into cache.
Anyone have an educated guess of what the actual score would be?
Zero. Origin servers don't have graphics cards. Which means, unfortunately, the Slashdot community is going to have to try to wrap its collective head around a more meaningful measurement of potential performance.
I can't believe I missed this the first time through.
Likewise your approving reference to "Manifest Destiny", a discredited and equally racist concept that, in a contemporary context, would support U.S. seizure of territory beyond it's current borders.
You're damn right! Sovereignty is a privilege, not a right. Look at Germany after World War II. Did we just hand the keys over to the German people again and pack out? Hell, no. Five years of occupation by the US, along with the UK and France, resulted in the formation of the Federal Republic, which has become one of the strongest, most modern countries on Earth. Essentially the same thing happened between '45 and '52 in Japan.
Let's compare this to a recent sequence of events. We smashed the oppressive and illegitimate Taliban regime in Afghanistan. Where were our occupation forces? Where were the parades through Kabul? Who's the provisional governor of Afghanistan? Where's Douglas MacArthur when we need him?
Nope. Instead of occupation and support, all they got was a loya jirga. Now, less than three months later, they're back to assassination attempts and bombings. There's no political system in Afghanistan, because nobody has taken the time to build one. Countries with no tradition of democracy can't just be handed a get-out-of-oppression-free card and expected to build their own country with it. They have to be taught.
That's why I oppose the proposed invasion of Iraq. If my country would go all the way with it-- marching into Baghdad, removing Hussein from power, and setting up a provisional occupation government for a period of not less than five years-- I'd be all for it. But I'm afraid that's not what's going to happen. Crushing the Hussein government without building a strong new government in its place would just destabilize things even further.
If we did it right, occupying whole swaths of the Third World would be the best thing for everybody concerned. There are countries where corruption and lawlessness are the rule, not the exception. These kinds of societies can't possibly be expected to govern themselves in any effective manner. As I've said many times before in this forum, democracy can't work without a universal respect for and observance of the law. In countries where there is no law-- only anarchy or dictatorship-- the seeds of democracy will find no purchase.
Get in there with a hundred thousand troops armed with M-16s, MREs with the little packages of M&Ms in 'em, and copies of the Declaration of Independence. Force 'em to live under the strict rule of martial law for a few years, then gradually give them the ability to govern themselves. In fifty years, we'll all be buying expensive stereos and luxury cars imported from that economic superpower, Afghanistan.
Bet your ass I'm for the seizure of territory outside our current borders. Somebody's gotta show these people how it's done. I don't see the French rushing to do it. Do you?
It's great to go away from Slashdot for a few hours only to find that one or more AC's has been holding your end of the argument in your absence. That's so cool.
He was making a generalized statement about the behavior of "foreign types" based solely on their status as "foreigners". I consider the phrase itself just as racist as "white types", "black types", "Asian types", "American types", etc.
That's the kind of wooly-headed liberal thinking that leads to grandmothers being patted down at airports. Making sound, well-informed decisions based on experience means observing trends. If I burn my hand on three ovens in a row, I'll learn to be afraid of ovens. Likewise, if (for example) people of Arab extraction blow up three buildings in a row, I'll learn to keep a closer eye on people of Arab extraction. That's racial profiling, and it's very controversial. It's also smart. It's very easy to look back on events like the internment of Japanese Americans in the 1940's and shake our heads, but don't conclude that just because people went to unreasonable extremes then that their motivations were flawed.
People fall naturally into groups. There are old people, young people, white people, black people, tall people, short people. Within a group, you often-- not always, but often-- find trends of behavior and personality. I find that people who aren't from American tend to have certain opinions and beliefs in common. So I'm going to note this, and use this information.
Don't like it? Fine. Call me a racist. Doesn't bother me a bit, because I know better.
I hate to say that I agree completely. Occasionally season 3 had a weird episode here and there ("Scratch & Sniff," "Revenging Angel") but also some really over-the-top good stuff. The "Daedalus Demands"/"Icarus Abides" saga that started way back in "Eat Me" was some really amazing television.
But Season 4 has just blown, hitting what I think was its low point with "John Quixote."
But I'm easy. If they would throw me a bone with something as good as "The Flax" or "Crackers Don't Matter," I'd be right back in front of my tee vee.
(God, I'm a geek. Quoting all these episode titles, I sound like Comic Book guy. Time to go soak up some real life.)
I'd estimate that 10% of books are worth reading, and less than 1% of television programming is worth the electricity to transmit it.
De gustibus non disputandum est. I'd say the fractions are closer to one percent of one percent on both counts. If you think one out of every ten books is worth reading, you must not be looking at very many books.
This is the common American failing; confusing quantity with quality. I'd rather have 3 good channels than 2000 bad ones.
And a common European failing is to think that you can get 3 good channels. Won't happen unless you socialize television. Oh, wait....
It's better to have 2000 channels, each of which being outstanding some of the time, than to have 3 channels that are always mediocre. Let the cream rise to the top, and then skim, baby, skim.
I'd love to hear somebody-- anybody-- explain to me how Linux has any kind of edge over OS X. But please omit discussions on how Linux runs on just about any kind of hardware. I will simply argue that OS X runs on better hardware, and we'll be right back where we started.
What's Linux got going for it that OS X doesn't?
Comments of a political nature and absurd misuses of the word "free" will be pointed at, laughed at, and ignored.
Slashdot Mirror
Does anybody know if this format is used somewhere else or even documented?
Uhhh... you mean vCalendar? Yeah, I think I saw some somewhere.
The docs only talked about 'Copy on write' mechanisms, but does it also do 'Copy on read'?
I believe it does. They call it page replication. When a CPU on node 1 fetches data from memory on node 2, the system replicates the data pages from node 2 onto node 1, if free memory is available for them. It's all automatic. In this way, main memory is almost used like a level 4 cache.
Do all NUMA machines do that?
By now, they might, but when the O2000 was first released, SGI made a very big deal about this feature.
Based on your resume, you obviously have inside information that I'm not privy to. But I think you're oversimplifying the story a bit.
The Origin 3000 series (SN-1) was designed very nearly from the beginning to accommodate either MIPS CPUs from the R10000 family or IA-64 CPUs. In 1996-1997, when I worked most closely with the SN-MIPS and SN-IA groups, there was doubt about whether the IA-64 processor would be Merced or McKinley, but there was no question about support for one or both of them.
Look inside a C-brick some time. (If you don't have one handy-- heh-- there's an illustration here.) See all that empty space at the front? The original design called for the use of either MIPS PIMMs or IA-64 PIMMs. The IA-64 PIMMs would include all the necessary hardware to make the Intel chips talk to the Bedrock memory controller. The MIPS PIMMs are pretty small, about four inches square or so. But the IA-64 PIMMs were projected to be real monsters with giant heat sinks on them. Thus all that empty space in the C-brick.
For quite a while, of course, SGI has been working on SN-2, or whatever they're calling the successor to SN-MIPS these days. I'm not associated with that group any more, so I'm not in the loop on the new design. I've heard rumors on the order of 128 MIPS processors in a rack, quadrupling the processor density of the 3000 series systems, but I don't have any real information there. It's certainly possible that SGI is preparing to roll out their IA-64 systems in the spring in the new architecture, but that would surprise me. Of course, like I said before, you seem to know more than I do on this one.
I won't argue with you about it, but I'm not going to email you either. I'll just take your word for it and admit that you learn something new every day.
During development of Windows NT 3.1, the first version (god bless starting counting at 3) MS made a strong pitch to SGI to get behind it and work with them to replace IRIX. SGI turned it down, and later signed up for limited workstation production.
If true, this is the first I've heard of this. Can you back this up with some sort of evidence?
You say that synthetic benchmarks are irrelevant. Then you go on to say that this particular synthetic benchmark is highly relevant.
No, I don't. I really can't emphasize this enough: read. I said, "SPEC is synthetic and irrelvant." Big difference.
I'd like to see this run a TPC variant, which is closer to real-world than it is synthetic.
The TPC benchmarks are measurements of database performance. Since SGI was trying to demonstrate the features and capabilities of their hardware, it would have been completely inappropriate for them to use a database benchmark. STREAM TRIAD is great because it measures only one thing: the rate at which data can be moved from memory to the CPU or vice versa. The TPCs measure aggregate systems, including hardware, storage, OS, database software, and so on. They may be relevant if you're looking for a fast database server system, but they're hardly useful for evaluating one hardware architecture over another.
What, do I have to explicitly call out the components and subcomponents? It is a memory bus, for the purpose of this discussion.
The whole point of this discussion is that the SGI system can outperform virtually everything else on STREAM TRIAD because it has no memory bus. Memory busses are bottlenecks, and pumping a lot of data through them is very hard. The SGI system eliminates the bottleneck and thus demonstrates amazing bandwidth. When you miss the whole point of the discussion, I'm going to call you on it.
Back in the early days of Windows NT, it was not know what its capabilities would be. SGI nearly bought the farm by betting NT to replace IRIX.
Then SGI realized NT wasn't going to be for big machines, and let that bad dream fade away.
Dude, that's simply not true. SGI never even built a prototype Windows system that ran any version of NT before 4.0. By the time their Windows NT workstations made it out the door, Windows 2000 was very nearly a reality. So close that rather than adding support for certain features, SGI just punted. For example, dual monitor support was never offered on the NT systems until Windows 2000 came out, because NT 4.0 didn't support running two graphics pipes with different drivers.
SGI never, ever, spent any time or effort on Windows NT for anything other than workstations.
But likewise, finding x86 hardware with more processors is probably the largest reason that x86 Linux, Windows, whatever isn't running on bigger machines.
So, let me get this straight. They don't make big x86 boxes, and that's "probably the largest reason" why there are no OSs for big x86 boxes? Brilliant!
Good. The machine looks like it has a decent memory bus, and memory modules with a good configuration and speed rating.
You know, before you piss in SGI's Cheerios, you might want to do a little reading. The Origin 3000 architecture, on which this prototype system was based, has no memory bus at all. It uses a fabric of switched multi-gigabyte-per-second interconnects to attach CPUs to RAM and to other CPU nodes.
CPU benchmarks (like SPEC) are synthetic and irrelevant, because they fit in cache. Virtually no real application fits in cache, and the sort of applications you run on a machine this big deal with data sets no the order of tens or even hundreds of gigabytes. Memory-to-CPU bandwidth is probably the only real indicator of the ability of the system to handle real-world workloads.
It's also the only thing-- other than the dimensions and the color of the plastics-- that differentiates SGI's big Itanium 2 server from everybody else's big Itanium 2 servers.
It's not clear to what extent application programs have to be aware of this. Clearly, if you lay things out in memory badly, with lots of CPUs reading and writing the same memory from all over the memory net, the system will bottleneck.
Speaking as somebody who's done his share of IRIX programming, I'd say "none at all."
In some cases, on Origin 2000 hardware with older versions of IRIX, you could see notable performance differences if you went out of your way to place memory in banks adjacent to the running processors. But the Origin 3000 architecture, with its significant reductions in memory latency, and newer versions of IRIX, with their improved page replication algorithms, have made manual memory placement almost obsolete. Almost.
SGI spent a lot of time and trouble trying to reduce the impact of accessing remote memory. The caching mechanisms and page replication stuff are really well thought-out.
Anybody else see that as the main reason this is running Linux instead of Irix?
SGI started working on porting IRIX to the IA-64 architecture back in (I think it was) 1995 or 1996. Not long after, they found that it would be easier and cheaper to get Linux to scale more efficiently and to port some key libraries and services from IRIX than it would be to port all of IRIX over to the new architecture.
It's all about time and money.
Add to this of course that Itanium2 is hardly a vector processor, so what they're doing is comparing apples to oranges.
Ordinarily I'm all for spoiling benchmarks by pointing out ways in which they're not applicable, but in this case, you're wrong. This test measures bandwidth into and out of main memory. That's it. It makes no difference whether the processors have vector registers and instructions or not. Noting matters except the factors that contribute to moving data between main memory and the CPUs.
The second thought is: can it be partitioned?
Since this machine is a standard Origin 3000 with McKinley processor modules, I'm going to assume the answer will be yes. You can partition an O3000 down to a single processor brick + base IO brick, so I imagine that SGI will implement the necessary software bits to make that happen on the SN1-IA systems. I know there are both user space bits (mkpart, partmgr) and kernel space bits (the TCP-over-NUMAlink driver).
I personally have only seen partitioning used on HA systems and lab systems. For a fully fault-tolerant N-processor system, you can buy one 2N-processor Origin and partition it down the middle. The two nodes can run in parallel, passing data back and forth over the NUMAlink via TCP/IP, until one goes down. Also, partitioning is great in a lab environment. It's nice to be able to carve up a big multiprocessor system and give each user a 4-processor (or multiple of 4) node.
I wonder what linux apps would someone run on a system this big?
Anything you'd run on an IRIX system of that size, I'd imagine. I believe-- not positive-- that MSC has already released Nastran for Itanium 2 Linux. (Nastran is a computer-aided engineering tool used extensively in the automotive industry, and other manufacturing industries. It's used for things like stress, heat transfer, and vibration analysis.)
And, as long as the Fortran compilers are worth a damn, you can run just about any other scientific, analytical, or technical software, I'd imagine.
...interesting that SGI chose the Cray C90 - a system released in *1991* - to compare against. It's nice to know that it's only taken them 10+ years to catch up. :)
;-)
If you read the STREAM TRIAD web site linked above, you'll see that SGI didn't compare itself to the C90 exactly; it just ran a benchmark and published the results. Also in that approximate rank are other machines from NEC and Cray and, further down, Sun.
But you're right. Cray was way ahead of their time when it came to things like memory bandwidth. I remember a friend (ex-Crayon) telling me once that access to main memory on the T-90 was faster than access to the on-chip cache on the Pentium III. That sounds implausible, though, so he might have been exaggerating.
I'm curious how the new X1 (nee SV2) does on the STREAM suite.
The last word I got is that X1 is still in the PCB design phase. It's only running as a simulator right now. So it'll be a while before you see those numbers.
(That info is several months old, so I may be wrong.)
I would be very interested to know what version of the kernel they are using.
I tried really hard to find that info this morning before submitting, but to no avail. But the test was demonstrated at the Intel Developer's Conference, according to the press release, so maybe we could find somebody who knows somebody?
Am I wrong about what this benchmark means? Or am I missing something basic?
With no disrespect intended, I think you might be missing something basic.
Any activity that involves moving data into and out of RAM will benefit from the ability to do it faster. That includes things such disparate things as database processing (if you're lucky, you can cache your indices in RAM), media encoding, hell, even compiling. Memory bandwidth is one of the few aspects of computer design that touches just about every application, with the exception of those that are small enough-- or sufficiently well optimized-- to fit into cache.
Anyone have an educated guess of what the actual score would be?
Zero. Origin servers don't have graphics cards. Which means, unfortunately, the Slashdot community is going to have to try to wrap its collective head around a more meaningful measurement of potential performance.
I can't believe I missed this the first time through.
Likewise your approving reference to "Manifest Destiny", a discredited and equally racist concept that, in a contemporary context, would support U.S. seizure of territory beyond it's current borders.
You're damn right! Sovereignty is a privilege, not a right. Look at Germany after World War II. Did we just hand the keys over to the German people again and pack out? Hell, no. Five years of occupation by the US, along with the UK and France, resulted in the formation of the Federal Republic, which has become one of the strongest, most modern countries on Earth. Essentially the same thing happened between '45 and '52 in Japan.
Let's compare this to a recent sequence of events. We smashed the oppressive and illegitimate Taliban regime in Afghanistan. Where were our occupation forces? Where were the parades through Kabul? Who's the provisional governor of Afghanistan? Where's Douglas MacArthur when we need him?
Nope. Instead of occupation and support, all they got was a loya jirga. Now, less than three months later, they're back to assassination attempts and bombings. There's no political system in Afghanistan, because nobody has taken the time to build one. Countries with no tradition of democracy can't just be handed a get-out-of-oppression-free card and expected to build their own country with it. They have to be taught.
That's why I oppose the proposed invasion of Iraq. If my country would go all the way with it-- marching into Baghdad, removing Hussein from power, and setting up a provisional occupation government for a period of not less than five years-- I'd be all for it. But I'm afraid that's not what's going to happen. Crushing the Hussein government without building a strong new government in its place would just destabilize things even further.
If we did it right, occupying whole swaths of the Third World would be the best thing for everybody concerned. There are countries where corruption and lawlessness are the rule, not the exception. These kinds of societies can't possibly be expected to govern themselves in any effective manner. As I've said many times before in this forum, democracy can't work without a universal respect for and observance of the law. In countries where there is no law-- only anarchy or dictatorship-- the seeds of democracy will find no purchase.
Get in there with a hundred thousand troops armed with M-16s, MREs with the little packages of M&Ms in 'em, and copies of the Declaration of Independence. Force 'em to live under the strict rule of martial law for a few years, then gradually give them the ability to govern themselves. In fifty years, we'll all be buying expensive stereos and luxury cars imported from that economic superpower, Afghanistan.
Bet your ass I'm for the seizure of territory outside our current borders. Somebody's gotta show these people how it's done. I don't see the French rushing to do it. Do you?
Yes, that's true. But I fail to see what it has to do with anything at all.
Thanks for pointing it out, though.
It's great to go away from Slashdot for a few hours only to find that one or more AC's has been holding your end of the argument in your absence. That's so cool.
He was making a generalized statement about the behavior of "foreign types" based solely on their status as "foreigners". I consider the phrase itself just as racist as "white types", "black types", "Asian types", "American types", etc.
That's the kind of wooly-headed liberal thinking that leads to grandmothers being patted down at airports. Making sound, well-informed decisions based on experience means observing trends. If I burn my hand on three ovens in a row, I'll learn to be afraid of ovens. Likewise, if (for example) people of Arab extraction blow up three buildings in a row, I'll learn to keep a closer eye on people of Arab extraction. That's racial profiling, and it's very controversial. It's also smart. It's very easy to look back on events like the internment of Japanese Americans in the 1940's and shake our heads, but don't conclude that just because people went to unreasonable extremes then that their motivations were flawed.
People fall naturally into groups. There are old people, young people, white people, black people, tall people, short people. Within a group, you often-- not always, but often-- find trends of behavior and personality. I find that people who aren't from American tend to have certain opinions and beliefs in common. So I'm going to note this, and use this information.
Don't like it? Fine. Call me a racist. Doesn't bother me a bit, because I know better.
I hate to say that I agree completely. Occasionally season 3 had a weird episode here and there ("Scratch & Sniff," "Revenging Angel") but also some really over-the-top good stuff. The "Daedalus Demands"/"Icarus Abides" saga that started way back in "Eat Me" was some really amazing television.
But Season 4 has just blown, hitting what I think was its low point with "John Quixote."
But I'm easy. If they would throw me a bone with something as good as "The Flax" or "Crackers Don't Matter," I'd be right back in front of my tee vee.
(God, I'm a geek. Quoting all these episode titles, I sound like Comic Book guy. Time to go soak up some real life.)
"Well, foobar104 has given us some nice moments. But in between all those nice moments has been a high-volume sewer hose of cultural sludge."
WHOA!
See? It works for everything....
I'd estimate that 10% of books are worth reading, and less than 1% of television programming is worth the electricity to transmit it.
De gustibus non disputandum est. I'd say the fractions are closer to one percent of one percent on both counts. If you think one out of every ten books is worth reading, you must not be looking at very many books.
Just for the record, I got the joke the first time, and no offense taken. But thanks for clearing it up anyway.
This is the common American failing; confusing quantity with quality. I'd rather have 3 good channels than 2000 bad ones.
And a common European failing is to think that you can get 3 good channels. Won't happen unless you socialize television. Oh, wait....
It's better to have 2000 channels, each of which being outstanding some of the time, than to have 3 channels that are always mediocre. Let the cream rise to the top, and then skim, baby, skim.
Nobody's taking my bait.
I'd love to hear somebody-- anybody-- explain to me how Linux has any kind of edge over OS X. But please omit discussions on how Linux runs on just about any kind of hardware. I will simply argue that OS X runs on better hardware, and we'll be right back where we started.
What's Linux got going for it that OS X doesn't?
Comments of a political nature and absurd misuses of the word "free" will be pointed at, laughed at, and ignored.