I remember the day when I tried to run KDE over VNC. What a mess. It wouldn't even run over 8-bit color, the slightest action caused horrible delays, and when I left it alone for a bit, a *screen saver* kicked in. A remotely-displayed screen saver! Horrors!
Windowmaker works great in this scenario. Highly recommended.
My guess is that they're doing this for embedded applications. Remember that Red Hat does a fair amount of business in the embedded arena, and PowerPC processors are pretty big in embedded applications. So while their work on the compilers will benefit everyone, including people running Linux on their Macs, this doesn't mean you're going to see a PowerPC version of Red Hat Linux any time soon.
I might be more inclined to pay something if you offered an improved Slashdot for my money: let's call it "Slashdot plus": access to extra features, rejection bin, a fast server. The point is, I'd be paying more to get more than what I get today.
Instead, it's "pay more to get exactly what you get today, and maybe later we'll add some perks". Those that don't pay actually get penalized with large ads. Seems like you guys ran out with the stick and forgot the carrot.
"Now, there's nothing wrong with trying to make your web pages look good to the largest number of people. But it's a matter of priorities: if you place a higher value on the layout than on the meaning, then you don't value your words very highly. So why should I?"
Hey guys, the 36-inch color inkjet printer still seems to be out of ink. I thought you guys were going to come out and replace the cartridge two hours ago.
I hope they publish the commenters' names. I hope somebody tries to contact a statistically significant pool of them to find out why they wrote what they wrote.
I wonder what percentage of the 7500 in favor of the settlement are Microsoft employees' sisters, or work at companies with Microsoft contracts, or were somehow contacted by Microsoft PR hacks and "encouraged" to write letters. It'd be hilarious to find that 6000 of the letters were all written by some poor intern at Microsoft, using names from the phone book.
So, did these folks get source, or offers for source, for all the GPLed components? I haven't seen anybody come forward to say whether these folks are paying attention to the terms allowing them to redistribute other people's code.
I just finished building a quiet machine. I definitely recommend Directron as a nice place to buy all of the parts needed to build a machine that won't sound like a 747.
Some parts that went into my machine:
Seagate Barracuda IV hard drive. Noise level below what people can hear. 20, 40, 60, 80 GB models available.
Fong Kai FK-603 case. Includes duct to help hot CPU air out the rear case fan. This case got a nice review from Ars Technica. Good quality, good looking case.
Antec Jet-cool CPU cooler. Good cooling capability but still low noise.
Enermax power supply. Extremely low noise, with one intake and one exhaust fan to help remove CPU heat, and available in any power level you could want.
The only extra information I have is from working at a large third-party repair outfit, as a test engineer for several years:)
Your main point is correct, but the "not worth it" price point is a lot higher than you'd guess. It's just that the difficulty of testing and repairing equipment to an acceptable level (where you can say with 99% confidence that stuff is really fixed) is a lot harder than you'd guess it is. Throwing skilled engineering and technician time at a broken box only becomes viable once the value of the box is over $1000. Plus, you'd be suprised at how poor the testing/diagnostics ability of most shops is. When new product comes out every three months, there's never the time to do it right.
A refurbished product is normally gone over by someone who has some technical expertise while the "new" product is something that was put together by monkeys taught to wield a soldering iron.
You'd think that should be the case, but it's often not. Returned merchandise often breaks down like this: of ten returned units, six are perfectly good. Two have an obvious problem easily repaired by a skilled, well-trained technician, one is completely fried, and one is flaky and fails once every 37 hours.
The problems here are many:
Skilled, well trained people are expensive, and companies are cheap.
Skilled, well trained people tend not to enjoy being low status "repair" monkeys.
For manufacturers, it's usually cheaper to make new product than to employ skilled people to fix broken product.
Third parties attempting to repair stuff often don't understand it well enough to repair or test it properly.
Sometimes manufacturers farm out their "refurbing" to stupid third parties mentioned above.
Whoever does the work, it's often more expensive to test and repair boxes than the boxes were worth new. So the company doing this is likely to realize they're losing money and find something more worthwhile to do.
That last, flaky unit is probably going to pass diagnostics and look like the other good units, and get passed onto a customer who's going to be pissed at you. 10% of your "refurb" customers pissed off is a lot of people for a big company.
Units that have failed once are statistically much more likely to fail again, because automated manufacturing flaws (like poor soldering) are likely to occur in multiple places on the same board.
In the end, "refurb" units are often of low quality because it's impossible to test and repair computer equipment cheaply and well and make money doing so.
Not having read the argument before, I'd have to agree that Theo's "evidence" seems to support the argument that he doesn't play well with others.
There is a part right in the middle of his "evidence" (search for "holiday") where he receives a message saying "I'll be away for 16 days" email from one of his adversaries and for no obvious reason Theo writes a crazy ranting reply as if this harmless vacation message were some sort of personal attack.
I'm about to pick up one or two of Seagate's Barracuda ATA IV drives. They put out 20db of noise, which is, according to a friend who works with them, "inaudible". Apparently the new fluid bearings are the reason.
The 80GB ones are a little hard to find, but 40GB are available for just over a hundred bucks.
Huh. 3 people moderated that as a troll. An expressed desire for free software to spread and overtake it's patent-encumbered rival. That's pretty disappointing; I thought more folks here cared about stuff like that.
You make valid points, except who cares if they release future firmware updates that start restricting usage?
Well, you're right in that I wouldn't be prevented from using the product I'd purchased. But I think altering products via firmware updates is sleazy, and I don't want to support companies that would pull such a bait and switch. Furthermore, I may want to buy a second box for the basement, or one for my parents. For all these reasons I would prefer to buy a product whose manufacturer isn't going to try to screw me over to appease the recording industry.
I'll seriously consider one of these once I see one that supports Vorbis. Tying ourselves to patented standards is going to hurt us in the long run. What if the folks who own patents on MP3 start requiring secret copy-protection measures in their patent licenses?
Additionally, I'm not going to spend a cent unless I have assurances that the company is committed to fighting off the music industry's attempts to control digital music formats. I'm afraid that firmware updates down the road will start removing open-format functionality, and start trying to force me to use rights-eroding proprietary formats.
While it sure would be nice, and right, to give credit, I'm not convinced that it's legally necessary.
It seems likely to be that header file structure definitions are a functional description of how a piece of hardware works. And if that's the case, that information is no more copyrightable than the telephone book. And if it's not copyrightable, it's perfectly legal to remove the credits and license and redistribute however you want. Not right, mind you, but legal.
Looks to me like he's screaming about copyright infringement and/or license violations without understanding the limited scope of copyright.
This is my way of explaining to non-geeks why crypto regulations will have near-zero effect:
Imagine that somebody comes up with a way to build a bomb using sugar cookies. A building is blown up. Congress decides to regulate the sale of sugar cookies.
Now any sane person will realize that this is pointless, because any idiot can make their own sugar cookies, and bypass all the regulations. So the regulations can only work if the ingredients are also regulated or banned (flour, sugar, eggs), or perhaps all the sugar cookie recipes are destroyed.
At this point it's pretty obvious that such a scheme would never work. But somehow nobody seems to follow this logic when it comes to encryption. The only ingredients for encryption are general-purpose computers. The recipes are encryption algorithms and computer source code. The recipes can be rediscovered or recreated by smart mathematicians and computer programmers.
So what are we going to do? Regulate computers? Mathematics? Encryption algorithms, dozens of which are published in textbooks around the world?
You could no more regulate computers, mathematics, and algorithms today than you could flour, sugar, eggs, and sugar-cookie recipes. Even if you tried, it would have near-zero effect on the bad guys, and would only increase the risk that grandma's bank account gets emptied, because her password wasn't properly encrypted.
grr. non-mozilla users try this properly formatted link to the Linux Terminal Server Project.
Kinda weird, that mozilla can parse the bad one (with missing slash) into the correct URL. Didn't notice it was bad until slashcode tacked [http] onto the end instead of the hostname.
First off, I never said the SPL is a closed source license, I only said it was not GPL-compatible. So chill.
GFS will not run without kernel patches supplied by Sistina. OK, I'll agree the bulk of GFS is not distributed as a kernel patch. But it does require kernel patches to function.
Patching the kernel to allow a non-GPL compatible module to link against and run as a part of GPL software (and then distributing the resulting binary) isn't allowed under the GPL, and isn't even allowed under Linus' informal binary-modules allowance (which he's said only applies if you use standard kernel interfaces).
If the SPL disallows commercial redistribution, it's not GPL-compatible, and it's not free software. So even though I've met the Sistina guys, and think they're a cool, smart bunch, I think that this licensing change makes them look bad, and I fear it may invite legal trouble as well.
Slashdot Mirror
Windowmaker works great in this scenario. Highly recommended.
You're not supposed to have CNAMEs as MX records.
My guess is that they're doing this for embedded applications. Remember that Red Hat does a fair amount of business in the embedded arena, and PowerPC processors are pretty big in embedded applications. So while their work on the compilers will benefit everyone, including people running Linux on their Macs, this doesn't mean you're going to see a PowerPC version of Red Hat Linux any time soon.
Instead, it's "pay more to get exactly what you get today, and maybe later we'll add some perks". Those that don't pay actually get penalized with large ads. Seems like you guys ran out with the stick and forgot the carrot.
from design, by Jamie Zawinski.
From: eng
Subj: 1055CM STILL out of ink
Hey guys, the 36-inch color inkjet printer still seems to be out of ink. I thought you guys were going to come out and replace the cartridge two hours ago.
I wonder what percentage of the 7500 in favor of the settlement are Microsoft employees' sisters, or work at companies with Microsoft contracts, or were somehow contacted by Microsoft PR hacks and "encouraged" to write letters. It'd be hilarious to find that 6000 of the letters were all written by some poor intern at Microsoft, using names from the phone book.
I fear the license wars about to erupt.
Maybe you should be trying the Google cache!
Some parts that went into my machine:
Your main point is correct, but the "not worth it" price point is a lot higher than you'd guess. It's just that the difficulty of testing and repairing equipment to an acceptable level (where you can say with 99% confidence that stuff is really fixed) is a lot harder than you'd guess it is. Throwing skilled engineering and technician time at a broken box only becomes viable once the value of the box is over $1000. Plus, you'd be suprised at how poor the testing/diagnostics ability of most shops is. When new product comes out every three months, there's never the time to do it right.
You'd think that should be the case, but it's often not. Returned merchandise often breaks down like this: of ten returned units, six are perfectly good. Two have an obvious problem easily repaired by a skilled, well-trained technician, one is completely fried, and one is flaky and fails once every 37 hours.
The problems here are many:
- Skilled, well trained people are expensive, and companies are cheap.
- Skilled, well trained people tend not to enjoy being low status "repair" monkeys.
- For manufacturers, it's usually cheaper to make new product than to employ skilled people to fix broken product.
- Third parties attempting to repair stuff often don't understand it well enough to repair or test it properly.
- Sometimes manufacturers farm out their "refurbing" to stupid third parties mentioned above.
- Whoever does the work, it's often more expensive to test and repair boxes than the boxes were worth new. So the company doing this is likely to realize they're losing money and find something more worthwhile to do.
- That last, flaky unit is probably going to pass diagnostics and look like the other good units, and get passed onto a customer who's going to be pissed at you. 10% of your "refurb" customers pissed off is a lot of people for a big company.
- Units that have failed once are statistically much more likely to fail again, because automated manufacturing flaws (like poor soldering) are likely to occur in multiple places on the same board.
In the end, "refurb" units are often of low quality because it's impossible to test and repair computer equipment cheaply and well and make money doing so.There is a part right in the middle of his "evidence" (search for "holiday") where he receives a message saying "I'll be away for 16 days" email from one of his adversaries and for no obvious reason Theo writes a crazy ranting reply as if this harmless vacation message were some sort of personal attack.
Bizarre...
Ha! You think Microsoft pays taxes?
The 80GB ones are a little hard to find, but 40GB are available for just over a hundred bucks.
Huh. 3 people moderated that as a troll. An expressed desire for free software to spread and overtake it's patent-encumbered rival. That's pretty disappointing; I thought more folks here cared about stuff like that.
Additionally, I'm not going to spend a cent unless I have assurances that the company is committed to fighting off the music industry's attempts to control digital music formats. I'm afraid that firmware updates down the road will start removing open-format functionality, and start trying to force me to use rights-eroding proprietary formats.
It seems likely to be that header file structure definitions are a functional description of how a piece of hardware works. And if that's the case, that information is no more copyrightable than the telephone book. And if it's not copyrightable, it's perfectly legal to remove the credits and license and redistribute however you want. Not right, mind you, but legal.
Looks to me like he's screaming about copyright infringement and/or license violations without understanding the limited scope of copyright.
Imagine that somebody comes up with a way to build a bomb using sugar cookies. A building is blown up. Congress decides to regulate the sale of sugar cookies.
Now any sane person will realize that this is pointless, because any idiot can make their own sugar cookies, and bypass all the regulations. So the regulations can only work if the ingredients are also regulated or banned (flour, sugar, eggs), or perhaps all the sugar cookie recipes are destroyed.
At this point it's pretty obvious that such a scheme would never work. But somehow nobody seems to follow this logic when it comes to encryption. The only ingredients for encryption are general-purpose computers. The recipes are encryption algorithms and computer source code. The recipes can be rediscovered or recreated by smart mathematicians and computer programmers.
So what are we going to do? Regulate computers? Mathematics? Encryption algorithms, dozens of which are published in textbooks around the world?
You could no more regulate computers, mathematics, and algorithms today than you could flour, sugar, eggs, and sugar-cookie recipes. Even if you tried, it would have near-zero effect on the bad guys, and would only increase the risk that grandma's bank account gets emptied, because her password wasn't properly encrypted.
Kinda weird, that mozilla can parse the bad one (with missing slash) into the correct URL. Didn't notice it was bad until slashcode tacked [http] onto the end instead of the hostname.
go here: Linux Terminal Server Project
GFS will not run without kernel patches supplied by Sistina. OK, I'll agree the bulk of GFS is not distributed as a kernel patch. But it does require kernel patches to function.
Patching the kernel to allow a non-GPL compatible module to link against and run as a part of GPL software (and then distributing the resulting binary) isn't allowed under the GPL, and isn't even allowed under Linus' informal binary-modules allowance (which he's said only applies if you use standard kernel interfaces).
If the SPL disallows commercial redistribution, it's not GPL-compatible, and it's not free software. So even though I've met the Sistina guys, and think they're a cool, smart bunch, I think that this licensing change makes them look bad, and I fear it may invite legal trouble as well.