I never buy stuff with logos. I do wear logo stuff if it was given to me for free... and I don't have anything else clean to wear. My only exception is maybe shoes, and even then, I go for the subtle stuff.
In one sense you're sort of delcaring your "tribe" by wearing logo'ed products. As far as I'm concerned, if you're paying Nike to advertise Nike products, you are part of the stupid tribe.
If you settle a lawsuit out of court, that doesn't create legal precedent. That only happens when the court itself actually decides the case.
Now having a few companies cave into the demands by BT may help get other companies nervous enough to cave, but that doesn't actually hurt the case of a company that decides to fight BT's patent.
It may not be apparent from the changelog, but there are also some important fixes to NFS version 2 (along with the v3 updates mentioned elsewhere) in the 2.2.18 kernel.
If you are using NFS at all with a 2.2.x kernel, then you really should upgrade. Yes, even if you're using NFSv2 over UDP.
Props out to trond, dhiggen, hjl, ac and all the guys on the NFS list.
Huh, the reason they are using Linux is because it's free. That and bunch of schmuks are willing to to work on the OS for free. And it has been my experience that you get what you pay for, ask yourself "what kind of job would you do if you weren't getting paid".
Looking at it from the reverse angle, shouldn't you be worried about projects where people work on them only because they are getting paid? If it is not a pleasant project to work on, isn't it likely that they won't try their very best? After all, their concerns might be keeping their jobs and making money instead of producing good code.
The above shouldn't be considered a rule of course, there are many professional coders that produce good quality work. And working for free certainly isn't a guanantee of good work.
It is just that there are no absolutes either way.
Usually, I'd build two collectors, and one controller. I'd have the controller guard the collectors, but since it doesn't have guns, it just followed them.
I'd also usually build a few multi-gun corvettes to guard the resource controller.
I mean, sure, our asteroid belt seems pretty empty now because we don't see anything on radar. But we haven't built the long-range sensor platform yet either.
Just something to think about.
Re:Correcting the failure of software copyright
on
Embracing Insanity
·
· Score: 4
Good post, BTW.
The fact that over 99% of the people who use Free and Open Source software will never modify it is irrelevant. What is important is that the tiny fraction of young people who are curious and want to learn how software works so that they can write their own, finally have the opportunity to examine and play with full fledged, working, professional quality software. And in the case of Free software, they have the right to reuse and redistribute their own work -- the modified code.
I think it's important to emphasize that nobody knows ahead of time who that 1% will be. Sometimes the best code can come from the strangest little corners; from unexpected people.
When code is Open Source, we (as a society) get the greatest chance that somebody out there will get some insight from existing code, and come up with something else brilliant. Those flashes can potentially advance the state of the art.
Doesn't the purpose of your job have something to do with this?
The grad students I knew were not paid by the university to write their thesis work, they were paid because of the other work done as part of being a grad student: lecturing, tutoring, grading papers, etc.
If, on the other hand, a grad student is employed by the network administrators, and writes a backup utility, then the University might have dibs on it. On the gripping hand, if [s]he was researching efficent protocols to backup systems over a network, then maybe not.
At any rate, it's all being driven by money. Universities want to cash in.
Fortunately for Computer Science, a lot of the most interesting work (to me anyway) is moving onto the Internet. People like me get interested in particular areas, and just start working together. I don't have a thesis advisor, nor do I take classes. I read about stuff, and start thinking about it. I don't need to go to school anymore. Most of the stuff I need to learn is on the Internet anyway.
That's why I like using window managers like kwm under X. I disable the task bar, and just use the center button menu to select windows. Nice big titles there.
Now if someone did that as part of a replacement shell for Windows, I'd buy it. It would have to support multiple desktops of course...
I'd think it wouldn't take too much effort to get faster Internet access on the DC than this method.
I've never seen the inside of a DC, but it would be likely to have a separate modem chip. The specs for just about any commercial chip are out there on the Internet. If you can find the modem, it wouldn't be too hard to disconnect it, and get at the RS-232 directly.
You'd need a level converter; the onboard RS-232 is prolly TTL level where low = 0V and high = +3V, so you'd convert it to low = -5V and high = +5V. Wire up a DB-9 serial connector, get a null modem cable, and you're done with the hardware.
On the host side, you'd need to whip up a chat script that looked (from the DC's point of view) as modem, which then starts up your PPP connection. The serial port probably runs at 115.2kbps.
Might make a nice little hardware project for some hardware hackers out there.
Yup, it goes to show how strongly our pattern recognition firmware is geared towards social interaction.
It always makes me amused to hear about the "Face on Mars", which is a somewhat symmetrical mountain that does look vaguely like a human face. Some people take this as evidence of intelligent life trying to contact us. Sheesh.
Occam's Razor, people. Just because you see some lights in the sky doesn't mean we're being visited by aliens. Just because the pyramids in Egypt are made out of big, heavy blocks doesn't mean that we had construction help from "Ancient Astronauts". And some long scratches in the ground do not mean it's an spaceship runway.
What gets me the most are the people who believe crop circles are done by aliens. And this is after the original guys showed how they did it on TV!
OK, that drifted a bit. I must be in rant mode today.
And, for the most part we still have some of our rights given to us by the constitution, but that list is getting shorter.
You have it backwards. Unfortunately, this belief that the Constitution gives us (the people) rights is not correct.
The Constitution does not give us rights, it is supposed to sharply define what (hopefully few) rights the government has. In the view of the Founding Fathers, we already have the rights (God-given and all that), and it's the government that has to be given some rights for the benefit of society as a whole.
The government and the special interest groups have come to believe that the government and the Constitution are the source of our rights, and have been acting in that fashion for quite some time now.
No, the original intent was that the government take the minimally necessary steps to ensure the safety and well-being of all. However, it has become an instrument of special interest groups to try to carve out areas of interest and profits for themselves.
It's well past time for everyone to see the folly of such an attitude. It is like with a community park, where the neighbors don't get along with each other any more. First one person stakes a claim on "his" portion of the park. Then everyone now becomes intent on stakings claims to "their" parts of the park. Soon the park is broken up into all these little areas, all clearly marked. Sure, now you've got your own space, and no one will disturb you without your permission. But you can't play frisbee in the park, nor have a picnic, or just go for a stroll without a lot of negotiation. This is tedious and time-consuming. The park as a whole as lost much of it's value.
Contrast that with a more community-minded attitude (such as with OSS). Everyone shares the park, and helps maintain it. You don't have all these boundries to worry about. You just go ahead and have that frisbee game, or walk your dog, or have a picnic. Of course, this system is subject to abuse. Some people may leave litter after their picnic, or some dog crap after a walk. But the answer to this is education, not legislation. That's what we strive for in the free/OSS community, and that's what we should strive for in the rest of the world too.
Aside: Hrmmmm... I'm in a very Libertarian mood today. I wonder why?
The purpose of burn-in testing is to find parts that could fail on the customer. It has no effect on stability other than to eliminate faulty parts.
Most electronic (or at least semiconductor) components follow the "bathtub curve" for failures. This means that components are likely to fail very early after manufacture, or at the end of their operational life (*), with a long stable period in between. You want to try to catch the ones that will fail early before your customer complains about it.
From what I understand of semiconductor electronics, running the system for a while will not help it stabilize or anything else.
(*) For older semiconductor parts, their lifetime was 15-20 years. But for stuff produced now, it's about 5. Constant causes the gates to "wear out". The fancy phrase is "electron migration". The metal in the junctions gets eroded away by the current, weaking the gate. High current, and especially high temperatures accelerate this process.
The problem is that the transmissions from Pioneer 10 are already just a smidge above the noise threshold for the receiving equipment. Even if things had gone exactly as planned, we'd have still lost contact with it next year.
It's amazing to think about the probe. It's the furthest any of our technology has travelled away from us.
It's still out there now, in the cold, cold, cold of space. It is truly swimming in a sea of stars. The Sun is barely brighter than the other stars in the sharp blackness. It still listening for the whispers of its masters, still waiting at their command. However, it can no longer hear from the people who have cared about it so much. It is all alone now. So far, far away from home.
As the years go on, it's heart, the RTG, will slowly cool, and the bus voltage will drop. At some point, the heartbeat of it's system clock will stop, and the little probe will sleep for eternity. Asleep among the stars.
It has a 16MB flash disk, and it runs QNX. There have been various hacks reported on Slashdot about installing Linux, as well as attaching IDE drives, etc.
Sure, it boots fast, and for mobile or embedded applications the hardware can be more reliable. But flash is still way more expensive than magnetic storage. Eventually the speed and cost of solid state storage will fall below magnetic and optical, but that won't be for at least a decade.
I sorta like it. I find it convenient to download freely available texts (like OS documentation or really old fiction) in HTML form, and stick it on the eBook to read at my convenience later.
However, I don't use it at all for it's intended purpose. I've never purchased any books for it. I find the idea sorta annoying that if I download a book, it's tied to a specific reader. I know that if my book is damaged, it is possible to re-register all my purchased books onto another reader. But that sounds like a hassle. Not very convenient.
I understand the desire of copyright holders to try to earn money on sales. But the new electronic formats aren't convenient. I can't loan a book to a friend, unless I loan that friend my reader too. But then I can't read anything! Ugh. Very annoying. It also doesn't help that the e-book prices are about the same as the paper books.
I have no idea how these e-books are going to work in a library. The readers are over $100 each, so you can't afford one for each separate book.
Publishers don't yet seem to realize how important a friend's recommendation is. Often being able to sample a piece (music, literature, etc.) will lead to future sales. In the end, I think they'll just be hurting themselves sales-wise, and hurting the rest of us by reducing the average quality of literature easily available to the general public.
OK. So what's stopping people from buying support contracts for OSS? I don't tend to do that myself, because I've been a C programmer and Unix administrator for over a decade.
... it's nice to have someone to blame, someone who has a responsibility to back up their product.
OK, so I blame M$ all the time for the problems I have using Windows. But does assigning blame fix the problem? Does it necessarily fix the problem any faster? No. If I submit a bug report to M$, is it likely that they'll look at it before next year? The only time a big software company like M$ pays attention to you is if you have a big, fat support contract.
In short, it's not MS's OS that everyone's buying - it's the support.
That's exactly right. However, with OSS, you get to pick who supports you. With proprietary software, you have no choice.
Sounds like a slam-dunk for OSS to me. Of course, it'll take the marketplace a while to come around. I believe the change is inevitable.
That's exactly right. If you're going to an LPD capable printer, you can just set up a queue to the printer directly. You then don't need the HP 4000 print server at all!
I do wish people would apply a little more brainpower before getting their undies in a bunch over a triviality.
Re:unix is NOT an operating system
on
Is UNIX An OS?
·
· Score: 3
Yup. It seems like the author has only been using computers for a few years. He sounds like a spoiled brat that's never had to go five minutes without on-line context sensitive help or pulldown menus.
He seems to forget that one of the first uses for Unix was for text processing. Real People(tm) like secretaries used Unix to write documents, store and organize them, and print them out. They were productive computer users waaay before GUI interfaces had even been invented yet.
Even in the Good Old Days(tm) of the Apple II and the original IBM PC, we had regular people (who were not computer hackers) learn how to use computers and be productive with them.
The author of the article also didn't study up on his history very well. Sun Microsystems started out on the Motorola 68000 line of microprocessors. They didn't move to RISC until much later.
The problem with a caching controller is that unless it's well engineered (with it's own battery backup), you more likely to run into filesystem corruption in the case of a power failure or OS crash.
A standard filesystem (such as ext2) on top of RAID5 will never be fast for small writes.
NetApps get around this because the WAFL filesystem is explicitly designed to sit atop a RAID4 drive array.
And there is a difference between RAID10 and RAID0+1.
RAID10 is a stripe of mirrors. Each pair of disks stores the same information (RAID1), and a stripe is created over those mirrors. This can tolerate multiple drive failures as long as at least one drive from each mirror is working.
RAID0+1 is a mirror of stripes. Two stripes are created(RAID0), each with half the total of disks. These stripes are then mirrored(RAID1). The problem here is that if a drive goes out, it takes out the entire stripe. If a drive in the other stripe goes out before the rebuild is complete, you're hosed.
Normally RAID systems (like RAID5) can't tolerate more than 1 drive failing at the same time. However, RAID10 provides more protection than RAID0+1, at the same price.
Perhaps this'd be possible under Berlin, but I'd like a Homeworld-style Interface.
Relic has made it relatively easy to navigate through a 3-D space with 2-D input device (a mouse). You can zoom back, rotate the view of your fleet, and then easily select and zoom in to look at details and give orders.
If you had a really high resolution display, you'd still be able to see the characters in terminal windows, even if they weren't full size. This would be good for getting an overview of your data space to see what's going on. And you'd be able to logically group things together in the shape you want instead of being limited to virtual desktops.
The game itself is pretty cool, I'd encourage y'all to check it out.
I guess it all depends on what your attitude towards what an operating system should be.
I believe that an OS should just provide a good, reliable, secure set of abstractions to deal with hardare and network communication. This allows you maximum flexibility to do whatever you want, which is what Michael is apparently slightly peeved about at this moment.
It is unfortunate that there are no standard libraries for other common functions such as printing. There's been stuff out there, such as TeX's dvi, but the major UNIX corporations in the '80s didn't adopt it. The major UNIX software vendors back then didn't have a decent, cross-platform library to code to, so they all went their own way.
As a counter-example, look at libcurses. It emerged early on as the way to code terminal applications. And now just about any text application you can point to (OSS or proprietary) uses it.
I think the lack of standards in things like printing is more an accident of history than any kind of fundamental design problem with UNIX.
Slashdot Mirror
I never buy stuff with logos. I do wear logo stuff if it was given to me for free... and I don't have anything else clean to wear. My only exception is maybe shoes, and even then, I go for the subtle stuff.
In one sense you're sort of delcaring your "tribe" by wearing logo'ed products. As far as I'm concerned, if you're paying Nike to advertise Nike products, you are part of the stupid tribe.
Ah, there may be a flaw in your thinking.
If you settle a lawsuit out of court, that doesn't create legal precedent. That only happens when the court itself actually decides the case.
Now having a few companies cave into the demands by BT may help get other companies nervous enough to cave, but that doesn't actually hurt the case of a company that decides to fight BT's patent.
It may not be apparent from the changelog, but there are also some important fixes to NFS version 2 (along with the v3 updates mentioned elsewhere) in the 2.2.18 kernel.
If you are using NFS at all with a 2.2.x kernel, then you really should upgrade. Yes, even if you're using NFSv2 over UDP.
Props out to trond, dhiggen, hjl, ac and all the guys on the NFS list.
P.S. Oh yeah, upgrade your NFS-utils too.
Huh, the reason they are using Linux is because it's free. That and bunch of schmuks are willing to to work on the OS for free. And it has been my experience that you get what you pay for, ask yourself "what kind of job would you do if you weren't getting paid".
Looking at it from the reverse angle, shouldn't you be worried about projects where people work on them only because they are getting paid? If it is not a pleasant project to work on, isn't it likely that they won't try their very best? After all, their concerns might be keeping their jobs and making money instead of producing good code.
The above shouldn't be considered a rule of course, there are many professional coders that produce good quality work. And working for free certainly isn't a guanantee of good work.
It is just that there are no absolutes either way.
I never let my resource collectors go out alone.
Usually, I'd build two collectors, and one controller. I'd have the controller guard the collectors, but since it doesn't have guns, it just followed them.
I'd also usually build a few multi-gun corvettes to guard the resource controller.
I mean, sure, our asteroid belt seems pretty empty now because we don't see anything on radar. But we haven't built the long-range sensor platform yet either.
Just something to think about.
Good post, BTW.
The fact that over 99% of the people who use Free and Open Source software will never modify it is irrelevant. What is important is that the tiny fraction of young people who are curious and want to learn how software works so that they can write their own, finally have the opportunity to examine and play with full fledged, working, professional quality software. And in the case of Free software, they have the right to reuse and redistribute their own work -- the modified code.
I think it's important to emphasize that nobody knows ahead of time who that 1% will be. Sometimes the best code can come from the strangest little corners; from unexpected people.
When code is Open Source, we (as a society) get the greatest chance that somebody out there will get some insight from existing code, and come up with something else brilliant. Those flashes can potentially advance the state of the art.
Doesn't the purpose of your job have something to do with this?
The grad students I knew were not paid by the university to write their thesis work, they were paid because of the other work done as part of being a grad student: lecturing, tutoring, grading papers, etc.
If, on the other hand, a grad student is employed by the network administrators, and writes a backup utility, then the University might have dibs on it. On the gripping hand, if [s]he was researching efficent protocols to backup systems over a network, then maybe not.
At any rate, it's all being driven by money. Universities want to cash in.
Fortunately for Computer Science, a lot of the most interesting work (to me anyway) is moving onto the Internet. People like me get interested in particular areas, and just start working together. I don't have a thesis advisor, nor do I take classes. I read about stuff, and start thinking about it. I don't need to go to school anymore. Most of the stuff I need to learn is on the Internet anyway.
That's why I like using window managers like kwm under X. I disable the task bar, and just use the center button menu to select windows. Nice big titles there.
Now if someone did that as part of a replacement shell for Windows, I'd buy it. It would have to support multiple desktops of course...
I'd think it wouldn't take too much effort to get faster Internet access on the DC than this method.
I've never seen the inside of a DC, but it would be likely to have a separate modem chip. The specs for just about any commercial chip are out there on the Internet. If you can find the modem, it wouldn't be too hard to disconnect it, and get at the RS-232 directly.
You'd need a level converter; the onboard RS-232 is prolly TTL level where low = 0V and high = +3V, so you'd convert it to low = -5V and high = +5V. Wire up a DB-9 serial connector, get a null modem cable, and you're done with the hardware.
On the host side, you'd need to whip up a chat script that looked (from the DC's point of view) as modem, which then starts up your PPP connection. The serial port probably runs at 115.2kbps.
Might make a nice little hardware project for some hardware hackers out there.
Yup, it goes to show how strongly our pattern recognition firmware is geared towards social interaction.
It always makes me amused to hear about the "Face on Mars", which is a somewhat symmetrical mountain that does look vaguely like a human face. Some people take this as evidence of intelligent life trying to contact us. Sheesh.
Occam's Razor, people. Just because you see some lights in the sky doesn't mean we're being visited by aliens. Just because the pyramids in Egypt are made out of big, heavy blocks doesn't mean that we had construction help from "Ancient Astronauts". And some long scratches in the ground do not mean it's an spaceship runway.
What gets me the most are the people who believe crop circles are done by aliens. And this is after the original guys showed how they did it on TV!
OK, that drifted a bit. I must be in rant mode today.
And, for the most part we still have some of our rights given to us by the constitution, but that list is getting shorter.
You have it backwards. Unfortunately, this belief that the Constitution gives us (the people) rights is not correct.
The Constitution does not give us rights, it is supposed to sharply define what (hopefully few) rights the government has. In the view of the Founding Fathers, we already have the rights (God-given and all that), and it's the government that has to be given some rights for the benefit of society as a whole.
The government and the special interest groups have come to believe that the government and the Constitution are the source of our rights, and have been acting in that fashion for quite some time now.
No, the original intent was that the government take the minimally necessary steps to ensure the safety and well-being of all. However, it has become an instrument of special interest groups to try to carve out areas of interest and profits for themselves.
It's well past time for everyone to see the folly of such an attitude. It is like with a community park, where the neighbors don't get along with each other any more. First one person stakes a claim on "his" portion of the park. Then everyone now becomes intent on stakings claims to "their" parts of the park. Soon the park is broken up into all these little areas, all clearly marked. Sure, now you've got your own space, and no one will disturb you without your permission. But you can't play frisbee in the park, nor have a picnic, or just go for a stroll without a lot of negotiation. This is tedious and time-consuming. The park as a whole as lost much of it's value.
Contrast that with a more community-minded attitude (such as with OSS). Everyone shares the park, and helps maintain it. You don't have all these boundries to worry about. You just go ahead and have that frisbee game, or walk your dog, or have a picnic. Of course, this system is subject to abuse. Some people may leave litter after their picnic, or some dog crap after a walk. But the answer to this is education, not legislation. That's what we strive for in the free/OSS community, and that's what we should strive for in the rest of the world too.
Aside: Hrmmmm... I'm in a very Libertarian mood today. I wonder why?
The purpose of burn-in testing is to find parts that could fail on the customer. It has no effect on stability other than to eliminate faulty parts.
Most electronic (or at least semiconductor) components follow the "bathtub curve" for failures. This means that components are likely to fail very early after manufacture, or at the end of their operational life (*), with a long stable period in between. You want to try to catch the ones that will fail early before your customer complains about it.
From what I understand of semiconductor electronics, running the system for a while will not help it stabilize or anything else.
(*) For older semiconductor parts, their lifetime was 15-20 years. But for stuff produced now, it's about 5. Constant causes the gates to "wear out". The fancy phrase is "electron migration". The metal in the junctions gets eroded away by the current, weaking the gate. High current, and especially high temperatures accelerate this process.
Well, if you're running NFS, you'd better pay attention to kernel traffic and the kernel-NFS mailling list.
There are some significant fixes for NFS stuff (client and server) coming out in 2.2.18.
Chances are not very good.
The problem is that the transmissions from Pioneer 10 are already just a smidge above the noise threshold for the receiving equipment. Even if things had gone exactly as planned, we'd have still lost contact with it next year.
It's amazing to think about the probe. It's the furthest any of our technology has travelled away from us.
It's still out there now, in the cold, cold, cold of space. It is truly swimming in a sea of stars. The Sun is barely brighter than the other stars in the sharp blackness. It still listening for the whispers of its masters, still waiting at their command. However, it can no longer hear from the people who have cared about it so much. It is all alone now. So far, far away from home.
As the years go on, it's heart, the RTG, will slowly cool, and the bus voltage will drop. At some point, the heartbeat of it's system clock will stop, and the little probe will sleep for eternity. Asleep among the stars.
As a long-time owner of two SGI 1600SW panels, I have only one response to the news of the Multilink adapter:
Yeeeeeeeeeessssssssssssssssss!!!!!!!!!!!!!!!!!!!
The Nine is just fine for 2D graphics, but sucks rocks for 3D (especially OpenGL).
I recently assembled a game machine with a GeForce card, and I'd love to be able to move the SGI to it.
Ummm... The iOpener?
It has a 16MB flash disk, and it runs QNX. There have been various hacks reported on Slashdot about installing Linux, as well as attaching IDE drives, etc.
Sure, it boots fast, and for mobile or embedded applications the hardware can be more reliable. But flash is still way more expensive than magnetic storage. Eventually the speed and cost of solid state storage will fall below magnetic and optical, but that won't be for at least a decade.
I sorta like it. I find it convenient to download freely available texts (like OS documentation or really old fiction) in HTML form, and stick it on the eBook to read at my convenience later.
However, I don't use it at all for it's intended purpose. I've never purchased any books for it. I find the idea sorta annoying that if I download a book, it's tied to a specific reader. I know that if my book is damaged, it is possible to re-register all my purchased books onto another reader. But that sounds like a hassle. Not very convenient.
I understand the desire of copyright holders to try to earn money on sales. But the new electronic formats aren't convenient. I can't loan a book to a friend, unless I loan that friend my reader too. But then I can't read anything! Ugh. Very annoying. It also doesn't help that the e-book prices are about the same as the paper books.
I have no idea how these e-books are going to work in a library. The readers are over $100 each, so you can't afford one for each separate book.
Publishers don't yet seem to realize how important a friend's recommendation is. Often being able to sample a piece (music, literature, etc.) will lead to future sales. In the end, I think they'll just be hurting themselves sales-wise, and hurting the rest of us by reducing the average quality of literature easily available to the general public.
Businesses like dealing with other businesses.
OK. So what's stopping people from buying support contracts for OSS? I don't tend to do that myself, because I've been a C programmer and Unix administrator for over a decade.
OK, so I blame M$ all the time for the problems I have using Windows. But does assigning blame fix the problem? Does it necessarily fix the problem any faster? No. If I submit a bug report to M$, is it likely that they'll look at it before next year? The only time a big software company like M$ pays attention to you is if you have a big, fat support contract.
In short, it's not MS's OS that everyone's buying - it's the support.
That's exactly right. However, with OSS, you get to pick who supports you. With proprietary software, you have no choice.
Sounds like a slam-dunk for OSS to me. Of course, it'll take the marketplace a while to come around. I believe the change is inevitable.
That's exactly right. If you're going to an LPD capable printer, you can just set up a queue to the printer directly. You then don't need the HP 4000 print server at all!
I do wish people would apply a little more brainpower before getting their undies in a bunch over a triviality.
Of course, if you're setting up a new network, you'd be better to start thinking about the Common Unix Printing Ssytem or maybe LPR Next Generation instead of dealing much with LPD.
Yup. It seems like the author has only been using computers for a few years. He sounds like a spoiled brat that's never had to go five minutes without on-line context sensitive help or pulldown menus.
He seems to forget that one of the first uses for Unix was for text processing. Real People(tm) like secretaries used Unix to write documents, store and organize them, and print them out. They were productive computer users waaay before GUI interfaces had even been invented yet.
Even in the Good Old Days(tm) of the Apple II and the original IBM PC, we had regular people (who were not computer hackers) learn how to use computers and be productive with them.
The author of the article also didn't study up on his history very well. Sun Microsystems started out on the Motorola 68000 line of microprocessors. They didn't move to RISC until much later.
The problem with a caching controller is that unless it's well engineered (with it's own battery backup), you more likely to run into filesystem corruption in the case of a power failure or OS crash.
A standard filesystem (such as ext2) on top of RAID5 will never be fast for small writes.
NetApps get around this because the WAFL filesystem is explicitly designed to sit atop a RAID4 drive array.
And there is a difference between RAID10 and RAID0+1.
RAID10 is a stripe of mirrors. Each pair of disks stores the same information (RAID1), and a stripe is created over those mirrors. This can tolerate multiple drive failures as long as at least one drive from each mirror is working.
RAID0+1 is a mirror of stripes. Two stripes are created(RAID0), each with half the total of disks. These stripes are then mirrored(RAID1). The problem here is that if a drive goes out, it takes out the entire stripe. If a drive in the other stripe goes out before the rebuild is complete, you're hosed.
Normally RAID systems (like RAID5) can't tolerate more than 1 drive failing at the same time. However, RAID10 provides more protection than RAID0+1, at the same price.
Technocrat ran the hoax story yesterday.
What made it a good hoax is that it sounded like something people might want to do, even though it was ridiculous.
What's the point of processing blocks faster anyway, doesn't SETI@Home have more processing power than they can practically use anyway?
Perhaps this'd be possible under Berlin, but I'd like a Homeworld-style Interface.
Relic has made it relatively easy to navigate through a 3-D space with 2-D input device (a mouse). You can zoom back, rotate the view of your fleet, and then easily select and zoom in to look at details and give orders.
If you had a really high resolution display, you'd still be able to see the characters in terminal windows, even if they weren't full size. This would be good for getting an overview of your data space to see what's going on. And you'd be able to logically group things together in the shape you want instead of being limited to virtual desktops.
The game itself is pretty cool, I'd encourage y'all to check it out.
I guess it all depends on what your attitude towards what an operating system should be.
I believe that an OS should just provide a good, reliable, secure set of abstractions to deal with hardare and network communication. This allows you maximum flexibility to do whatever you want, which is what Michael is apparently slightly peeved about at this moment.
It is unfortunate that there are no standard libraries for other common functions such as printing. There's been stuff out there, such as TeX's dvi, but the major UNIX corporations in the '80s didn't adopt it. The major UNIX software vendors back then didn't have a decent, cross-platform library to code to, so they all went their own way.
As a counter-example, look at libcurses. It emerged early on as the way to code terminal applications. And now just about any text application you can point to (OSS or proprietary) uses it.
I think the lack of standards in things like printing is more an accident of history than any kind of fundamental design problem with UNIX.
Don't forget Robot Wars which may be airing on your local PBS station.