That's funny, and even more so because it's somewhat true. I found that one could break LVM or other layers by using it on top of a resyncing mdadm RAID.
The key was, I broke it REPEATABLY, so I could then confirm that a certain patch fixes it. When I could no longer make it break, it was known to be fixed.
Having said that, RedHat is still stable on kernel 2.3.32, with some features backported. I WOULD be interested in knowing when the distros jump to a 3.10 or higher kernel. Bcache is going to be awesome for writeback caching.
I've been accused of being a Linux fanboy and even I agree.
Of course, I've also been accused of being a Microsoft shill and an Oracle shill. I AM a Linux kernel dev, barely, so yeah, that qualifies me as being a fan, I suppose.
* My name is in the kernel CHANGELOG exactly once.
> As you pointed out $50k is a drop in the bucket. But is $300k? $500k?
Yes, when you have a billion dollars and you're pissed, no amount ending in "k" is scary. On the extreme high side, a million dollars, that's well under 1% of his money.
Let's guess the average net worth in the US is about $50,000. 0.1% of that is $50. So a million dollars to this guy is like $50 to the average person.
* DoubleClick sold for $3 billion. I'm guessing O'Connor has about a billion.
The troll screwed up this time. $50K?!?! To O'Connor, that's like $20 is to most of us. It might get more expensive? As is $150K, less than 0.1% of his net worth? I don't think he's scared.
I'd say the year of the Linux desktop has aalready come, twice.
On one weekend in 2007, over two million *nix desktops were booted for the first time. It just so happened that *nix was BSD based, and had an Apple GUI. The year of the "Linux" desktop was the year of OSX. Not the kernel we hoped for, but a mainstream POSIX system that will run all your GPL code.
Then in 2010, millions of new systems had the Linux kernel. Today, MOST new computers have Linux installed. By 2010, the ubiquitous consumer PC had shrunk to fit in your hand. All of these lovely Linux systems had a nice GUI from Google. Since most new computers are portable, not chained to the desktop, I don't know if any future year of the Linux DESKTOP matters too much.
> if Apple was the only phone OS maker I'm sure they would license o/sX to anyone who could prove decent hardware compatibility
A few years ago, Apple had the only credible smartphone OS. They didn't license their smartphone OS when they were the only one. Instead they gave the majority of market share to a company who DID license their OS.
Is that a mistake they would only make once? For several years they had the only GUI OS for desktop computers. Rather than license it, they left every other manufacturer stuck with DOS. Had Apple licensed their OS, few of us would remember Microsoft.
CISC couldn't go that fast without using 125 watts. RISC could use 99% less power and go half as fast.
Everybody bought "Intel inside", even though it drew a hundred times more power.
Yes, mobile is one reason people now care more about power consumption. Waking up to it's effect on datacenter costs is another.
You said: "The goal was always to make devices as efficient as possible" If that were, CISC would have been dead on arrival. Intel has pretty much admitted that CISC will be dead soon unless they cut power usage by 99% because suddenly power usage is more important than brute speed.
For 20 years, RISC processors used 1/10th - 1/100th as much power, yet Intel was the big name brand because CPU speed was king. As Hognoxious pointed out, the P4 is a great example that people generally didn't care too much about power usage. 125 watts was a little high, but acceptable. Now 1 watt is considered a little too high, and companies are hyping 1.5 Ghz processors, a third the speed of existing offerings.
Certainly, and that would be a central issue in any dispute over this patent. Therefore, for either party to make public statements on that issue other than "you can read it in the patent" would be stupid, for the same reason that it's stupid for criminal suspects to converse with the police.
You have a great example when you mentioned "supermarket cashier". Cashiers ARE going the way of cotton pickers. Cotton pickers have been replaced by equipment operators who harvest 100 times as much cotton, sitting in an air conditioned cab making 50 times as much money.
A cashier used to type in prices. Machines took over that part of the job and the cashier would just point the UPC reader gun at the product, then count the money.That made cashiers faster and therefore more valuable. Next, the machines took over the gun part, so the cashie just passes the product over the machine and counts the money. Now, the machine counts the money, so one cashier can manage six checkout lanes, making them six times as productive and valuable. I note that here in Texas, cashiers now start at $11-$13 / hour. I'm pretty sure that's an improvement over the $4.25 they were paid a few years ago when they had to use the scanner gun on each item. Automation is here, and it's been good for cashiers.
The trend for the last 250 years has been for machines to do more and more menial labor. That process creates jobs like web developer, auto mechanic, and cable TV tech. Most of us work in jobs that didn't exist 100 years ago. Machines now do the jobs like "seamstress" and "cotton picker", humans do "biomedical engineer". I'm glad. I'd rather do software engineering than pick cotton.
I had a 1 GHz CPU around 10 years ago. Right now I'm using a 1.2 GHz. Before that, CPU speeds would double every few years.
Okay, I cheated because my current 1.2 GHz fits in my pocket. I do have two machines with five year old CPUs that run 3-3.5 GHz, the same speed as a new machine five years later. So there ARE some real physical limits. That's why phones are dual core and servers have eight cores - because they couldn't make faster processors they had to join together more processors running at the same old speed.
For that multimedia you speak of, a rotating drive will be about as fast. A 10K rpm platter drive makes a lot more sense for video, which is sequential access.
When SSDs get faster for sequential access, then I'll be interested in larger. I don't see any need for many TBs of tiny files, and SSD is only impressive with small files. Very large databases are about the only use case I can think of for large SSDs, and maybe media laptops. Even with 40TB of data, I only want a 128GB PCIe SSD for caching.
> Not everyone has that, many live paycheck-to-paycheck,
And most who make good money live close to paycheck-to-paycheck or worse, in debt, meaning they've spent the paycheck before they get it. How many people have a loan, a debt, on a $30,000 car. They could have bought that in cash by starting with a $1,500 car, saving up for a $3,000 car, then a $6,000 car, etc. That would cost them a lot LESS than paying interest to a finance company.
Congrats to you for not putting a down payment on a $50,000 car like many people would do.
> Oh, ok. So they're suing for a network topology that can be created by using wireless routers.
They are suing a few of the largest backbone providers - AT&T, Comcast, Qwest, Level3, Comcast. I'm pretty sure those networks aren't wifi based. In fact, the Cisco product mentioned in the CRS-1 routing platform. The CRS-1 is $80,000 each.
PS the appeals court ruled that plaintiff is not allowed to change their mind later and say either that the Cisco devices infringe or that using the devices infringes. Plaintiff has claimed that it's only infringing if they are connected in a certain infringing topology and the ruling is that they have to stick to that.
They are emphatically NOT "suing people for using Cisco devices", as evidenced by the fact that some defendants use other brands to do the same thing.
Whether or not their patent is any good I don't know, but it's not Cisco specific and they aren't claiming all Cisco users are infringing.
Their claim is more along the lines of "I'm suing you for building a bomb that blew up my car." They do not claim that using a tool is bad. They claim that doing X (with any tool) is bad.
Let's understand what their claim is before we decide if it's valid or bogus.
The same can be said for any device. Gears and levers were invented before spaceships. Spaceships are a configuration of gears and levers. Therefore no-one can invent a new type of space vehicle?
If you do something NEW with gears and levers, that's a new invention. If you do something new with wheels and suction cups, that's a new invention. If you do something new with silicon and copper, that's a new invention. , Many patents have been issued for software and other things that do not do anything new. Those patents should not have issued because patents are for new inventions. Many overly broad patents have been issued and shouldn't have because they are overly broad. If you confuse those issues with what material the invention is made of, somebody is tricking you into not thinking things through.
A hammer can be used for murder. Therefore, according to your reasoning manufacturers of hammers are liable for murder?
That's precisely analogous to your assertion that: routers can be used to build an infringing network, therefore router manufacturers are liable for infringement.
Perhaps you will say hammers are designed to build things, and that is somehow different. Okay then: Hammers can be used to build bombs. Therefore, hammer makers are responsible for bombings.
Slashdot Mirror
That's funny, and even more so because it's somewhat true.
I found that one could break LVM or other layers by using it on top of a resyncing mdadm RAID.
The key was, I broke it REPEATABLY, so I could then confirm that a certain patch fixes it. When I could no longer make it break, it was known to be fixed.
Having said that, RedHat is still stable on kernel 2.3.32, with some features backported. I WOULD be interested in knowing when the distros jump to a 3.10 or higher kernel. Bcache is going to be awesome for writeback caching.
I've been accused of being a Linux fanboy and even I agree.
Of course, I've also been accused of being a Microsoft shill and an Oracle shill. I AM a Linux kernel dev, barely, so yeah, that qualifies me as being a fan, I suppose.
* My name is in the kernel CHANGELOG exactly once.
> As you pointed out $50k is a drop in the bucket. But is $300k? $500k?
Yes, when you have a billion dollars and you're pissed, no amount ending in "k" is scary. On the extreme high side, a million dollars, that's well under 1% of his money.
Let's guess the average net worth in the US is about $50,000. 0.1% of that is $50. So a million dollars to this guy is like $50 to the average person.
* DoubleClick sold for $3 billion. I'm guessing O'Connor has about a billion.
The troll screwed up this time. $50K?!?! To O'Connor, that's like $20 is to most of us. It might get more expensive? As is $150K, less than 0.1% of his net worth? I don't think he's scared.
I'd say the year of the Linux desktop has aalready come, twice.
On one weekend in 2007, over two million *nix desktops were booted for the first time. It just so happened that *nix was BSD based, and had an Apple GUI. The year of the "Linux" desktop was the year of OSX. Not the kernel we hoped for, but a mainstream POSIX system that will run all your GPL code.
Then in 2010, millions of new systems had the Linux kernel. Today, MOST new computers have Linux installed. By 2010, the ubiquitous consumer PC had shrunk to fit in your hand. All of these lovely Linux systems had a nice GUI from Google. Since most new computers are portable, not chained to the desktop, I don't know if any future year of the Linux DESKTOP matters too much.
> if Apple was the only phone OS maker I'm sure they would license o/sX to anyone who could prove decent hardware compatibility
A few years ago, Apple had the only credible smartphone OS. They didn't license their smartphone OS when they were the only one. Instead they gave the majority of market share to a company who DID license their OS.
Is that a mistake they would only make once? For several years they had the only GUI OS for desktop computers. Rather than license it, they left every other manufacturer stuck with DOS. Had Apple licensed their OS, few of us would remember Microsoft.
CISC couldn't go that fast without using 125 watts.
RISC could use 99% less power and go half as fast.
Everybody bought "Intel inside", even though it drew a hundred times more power.
Yes, mobile is one reason people now care more about power consumption. Waking up to it's effect on datacenter costs is another.
You said:
"The goal was always to make devices as efficient as possible"
If that were, CISC would have been dead on arrival.
Intel has pretty much admitted that CISC will be dead soon unless they cut power usage by 99% because suddenly power usage is more important than brute speed.
For 20 years, RISC processors used 1/10th - 1/100th as much power, yet Intel was the big name brand because CPU speed was king. As Hognoxious pointed out, the P4 is a great example that people generally didn't care too much about power usage. 125 watts was a little high, but acceptable. Now 1 watt is considered a little too high, and companies are hyping 1.5 Ghz processors, a third the speed of existing offerings.
Certainly, and that would be a central issue in any dispute over this patent. Therefore, for either party to make public statements on that issue other than "you can read it in the patent" would be stupid, for the same reason that it's stupid for criminal suspects to converse with the police.
I'm curious how many people have actually read the patent and know what the claims are.
Sometimes, we play "we have to tar and feather them to find out why we should tar and feather them".
Will AMD respond with a beer powered processor ?
Seriously, though, it's good to see Intel is serious about, and capable of, truly low power.
Ten years ago, it was a race for the most powerful processor, and Intel won*. Now it's about competing for the lowest power. Kind of ironic.
* For single threaded applications. A web server with a $200 AMD 8-core CPU at 4GHz will beat the pants off $200 of Intel CPU.
You have a great example when you mentioned "supermarket cashier". Cashiers ARE going the way of cotton pickers. Cotton pickers have been replaced by equipment operators who harvest 100 times as much cotton, sitting in an air conditioned cab making 50 times as much money.
A cashier used to type in prices. Machines took over that part of the job and the cashier would just point the UPC reader gun at the product, then count the money.That made cashiers faster and therefore more valuable. Next, the machines took over the gun part, so the cashie just passes the product over the machine and counts the money. Now, the machine counts the money, so one cashier can manage six checkout lanes, making them six times as productive and valuable. I note that here in Texas, cashiers now start at $11-$13 / hour. I'm pretty sure that's an improvement over the $4.25 they were paid a few years ago when they had to use the scanner gun on each item. Automation is here, and it's been good for cashiers.
The trend for the last 250 years has been for machines to do more and more menial labor. That process creates jobs like web developer, auto mechanic, and cable TV tech. Most of us work in jobs that didn't exist 100 years ago. Machines now do the jobs like "seamstress" and "cotton picker", humans do "biomedical engineer". I'm glad. I'd rather do software engineering than pick cotton.
I had a 1 GHz CPU around 10 years ago. Right now I'm using a 1.2 GHz. Before that, CPU speeds would double every few years.
Okay, I cheated because my current 1.2 GHz fits in my pocket. I do have two machines with five year old CPUs that run 3-3.5 GHz, the same speed as a new machine five years later. So there ARE some real physical limits. That's why phones are dual core and servers have eight cores - because they couldn't make faster processors they had to join together more processors running at the same old speed.
For that multimedia you speak of, a rotating drive will be about as fast. A 10K rpm platter drive makes a lot more sense for video, which is sequential access.
When SSDs get faster for sequential access, then I'll be interested in larger. I don't see any need for many TBs of tiny files, and SSD is only impressive with small files. Very large databases are about the only use case I can think of for large SSDs, and maybe media laptops. Even with 40TB of data, I only want a 128GB PCIe SSD for caching.
> Not everyone has that, many live paycheck-to-paycheck,
And most who make good money live close to paycheck-to-paycheck or worse, in debt, meaning they've spent the paycheck before they get it. How many people have a loan, a debt, on a $30,000 car. They could have bought that in cash by starting with a $1,500 car, saving up for a $3,000 car, then a $6,000 car, etc. That would cost them a lot LESS than paying interest to a finance company.
Congrats to you for not putting a down payment on a $50,000 car like many people would do.
These are the "small guys" being sued:
AT&T $183 Billion
Comcast $115 Billion
Time Warner $58 Billion
Level3 $6 Billion
Still think they need Cisco ($130 B) to protect them?
> Oh, ok. So they're suing for a network topology that can be created by using wireless routers.
They are suing a few of the largest backbone providers - AT&T, Comcast, Qwest, Level3, Comcast.
I'm pretty sure those networks aren't wifi based. In fact, the Cisco product mentioned in the CRS-1 routing platform. The CRS-1 is $80,000 each.
http://www.infinity-micro.com/cisco-crs-1-series-8slot-carrier-routing-system-single-1495.html
Every defendant is worth at least a billion dollars each. This is not a case of going after little guys who can't defend themselves.
PS the appeals court ruled that plaintiff is not allowed to change their mind later and say either that the Cisco devices infringe or that using the devices infringes. Plaintiff has claimed that it's only infringing if they are connected in a certain infringing topology and the ruling is that they have to stick to that.
They are emphatically NOT "suing people for using Cisco devices", as evidenced by the fact that some defendants use other brands to do the same thing.
Whether or not their patent is any good I don't know, but it's not Cisco specific and they aren't claiming all Cisco users are infringing.
Their claim is more along the lines of "I'm suing you for building a bomb that blew up my car." They do not claim that using a tool is bad. They claim that doing X (with any tool) is bad.
Let's understand what their claim is before we decide if it's valid or bogus.
Yes, nobody is claiming that all networks infringe. The plaintiff doesn't claim it covers all networks and in fact stipulates that it does not.
The same can be said for any device. Gears and levers were invented before spaceships. Spaceships are a configuration of gears and levers. Therefore no-one can invent a new type of space vehicle?
If you do something NEW with gears and levers, that's a new invention. If you do something new with wheels and suction cups, that's a new invention. If you do something new with silicon and copper, that's a new invention.
,
Many patents have been issued for software and other things that do not do anything new. Those patents should not have issued because patents are for new inventions. Many overly broad patents have been issued and shouldn't have because they are overly broad. If you confuse those issues with what material the invention is made of, somebody is tricking you into not thinking things through.
A hammer can be used for murder. Therefore, according to your reasoning manufacturers of hammers are liable for murder?
That's precisely analogous to your assertion that: routers can be used to build an infringing network, therefore router manufacturers are liable for infringement.
Perhaps you will say hammers are designed to build things, and that is somehow different. Okay then:
Hammers can be used to build bombs. Therefore, hammer makers are responsible for bombings.
Sorry, your reasoning just doesn't work.