Sell two versions. The first being an "upgrade for your OEM Apple hardware" at the same $129 / $199 family pack, and another "Full Retail" version for $300 or whatever.
The bigger issue is support. Apple doesn't want to support "random" hardware. It's a nightmare. Better to do a deal with HP / Dell / Etc. and only support a few "OS X compatible" models, and make the OEM offer support.
If your intent is to support Apple by giving them money for the OS, then you can still do so and ignore the EULA. If your intent is to just be "legal", well, then you are still SOL. But yeah, I hear you.
There are tools to normalize volume levels over your entire mp3 collection - no need to manually tweak each song. When I throw a party, I make a playlist. Occasionally I'll have guests that want a certain song - the ad-hoc queuing function in Amarok makes that trivial without screwing over my playlist.
Nah. It's a big enough market that there are tons of restaurants in that price range, and they are busy.
Nvidia doesn't need to sell 30M of these cards to turn a profit. If they sell 100K-200K world wide they do fine - and they will. If you can't afford a $200 card, don't buy one. But lots of hard-core gamers will save up for this kind of thing and buy it. Then you have the people buying on 22% interest store credit cards that buy them too even though they *can't* afford it.
How about that being a really bad analogy that doesn't work?
The fact is, kernel device developers have the skills, they just don't know the specific APIs. Learning those APIs is not NEARLY as hard as you seem to think it is. To claim it is so difficult is an insult to the talented people we are taking about.
This is about preference - not skills.
When I first got out of school (ancient history here,) I got a job doing industrial control systems. I didn't know jack shit about them, but I had the skills needed to learn what I didn't know. And it didn't take long. After two weeks learning the interfaces and new terminology, I was productive. Two months and I was project lead. My next job was client / server application code. Then custom TCP/IP, SNA / LU 6.2, CPIC, and 3270. In all cases, some re-learning was needed, but people with an IQ over 70 can learn new things pretty easily and get the work done.
We are not even talking the difference between embedded systems programming and mainframe networking. It's still device driver code in the SAME programming language.
"The Linux Kernel" is one thing. "A Linux based OS" is another. The fact is, the term "Linux" by itself usually (like 99% of the time) refers to "A Linux based OS" rather than just "The Linux Kernel." Sorry that you can't seem to join the 99 percentile.
The fact is, people want more devices to work on their "Linux based OS" no matter in which layer of the system the driver runs.
With 300 idle people, it seems that it would be a good use of resources to learn the interfaces necessary to write libusb drivers. And the statement that it is a different skill set is total fucking bullshit. It's the same SKILLS, but a different API. Unless you are trying to tell me that a "kernel developer", like Linus, can't write application code, such as Git. If I was a kernel device developer (and while I have played with some kernel device driver code a number of years ago, it's not what I do) I would be offended by your comment that I lack the skill set to write a user-space driver. I may PREFER kernel code, but it's not a lack of "skills" that keeps me from writing user-space code.
USB devices "can" be accessed from userspace via a common kernel shim. Think of USB as a mini-ethernet network. SSH doesn't need to be in kernel space in order to talk to other devices on the network... It uses the common kernel shim to talk to the network interface device, just like libusb and other shims talk to the USB network device.
That said, audio and video USB devices don't use these userspace shims (such as libusb.) I personally don't know why, but I'm sure there is a good reason. I'm also sure that some USB devices that currently have kernel drivers could instead use userspace drivers (IR blasters, v4l devices, audio devices, etc.) Other than "legacy" reasons, I can't see why they don't. Is it because not ALL audio devices are USB? If you moved everything beyond the very low-level access layer out of the kernel, it sure would be easier to debug and increase kernel stability / reduce kernel bloat.
I agree, but there is another issue here. the LDP (and Novell) is willing to work with manufacturers and sign NDAs in order to get info needed to write drivers. While USB scanners (and printers) don't require "kernel" drivers, they still require drivers and the same NDAs that traditional driver devices need in order to convince manufacturers to work with the developers.
Why aren't the LDP people and the SANE people working together? A device is a device no matter what the interface. The end user doesn't really care how "device X" hooks up, or how the driver is loaded. They just want it to work.
Would it? If a movie studio makes a film that has a song playing in the background of a scene, they have to get a license for that. These licenses can be expensive.
On the other hand, Mom's home video was not made as a for-profit work. I don't know if copyright law makes exceptions for that however. IMHO, copyright law is WAY too strict.
Cows put out 500 liters of methane a day (that's per cow.) The obvious solution is to stick a tube up their asses and run a generator off of it. No need to make them walk.:-)
It's not about voip versus bittorrent - it's about real-time versus bulk and latency. It's about using resources the most efficiently. With shaping, you can reliably watch streaming youtube or porn while running your bittorrent, and everyone is happy. Shaping done right will hardly impact your overall transfer speeds.
The reality is that p2p is frequently consuming 70%+ of the bandwidth. If traffic isn't shaped at all, it makes the internet useless for anything *except* p2p. What comcast is doing isn't shaping. It's severe damage and fraud.
Anyway - so you don't use VoIP yet think it's a bad idea... Hmm. Well, I DO use VoIP, AND bittorrent, and I do think VoIP is a good idea. Not only is it less expensive, call quality is far superior, and my phone has a lot more capability. Any time I can pay Verizon or AT&T less money is a good thing. I'll go with my years of experience with VoIP over your uninformed opinion any day.
Um, the reason they will switch is that they will regain control of their systems that they have lost with Windows. What's your plan for the future - hope that Microsoft reverses course and starts to think about the needs of their customers? Fat chance of that...
And some of us have been doing just fine on alternative platforms for many many years. It's quite silly to claim that you can't run a business without Windows. The fact is, you can. I grant that it can be difficult to switch, but those of use running on alternative platforms for years prove that it can be done.
Until people just start biting the bullet and switching, no competition will EVER have a larger market share no matter HOW good their product.
The modern trick is to use VM technology to ease the transition. Eventually you will get to the point where most of your users won't need windows. Then you can work with your software vendors (or NEW software vendors) to fill in the gaps. As a result, competition IS starting to become viable for the cases where it hadn't been.
Microsoft has the right to do anything it wants to your machine, including totally disabling it. They also have the right to spy on you, reporting anything they want back to MS.
Of course back when Windows 2000 came out and all those new terms were introduced into the EULA, some of us raised red flags. Others "knew better" - MS would never actually DO the things the new EULA allowed....
Some of us still just didn't have that level of trust. We started migrating all our critical systems to alternative platforms, or didn't buy into the MS crack pipe in the first place.
So no, there is no conspiracy theory. MS opened the door for this behavior long ago, and just had the patience to slowly enforce the rights they guaranteed themselves. Why is anyone surprised or angry?
The overhead for the encrypted traffic would greatly increase the amount of data being used.
Not really. If I encrypt a 1K packet, it's still 1K. There is a MINOR bit of overhead negotiating the keys up front, but it's small enough that it would not be noticeable. If you want to ensure that the wrapper isn't tampered with (the IP packet header) you can add additional checks in the data portion of the packet. This would add about 0.5% additional overhead. Still not a big deal. Authenticate all clients / trackers with a cert from cacert.org and MITM attacks are thwarted.
From what I understand, they are forging packets that make your BT client think that peers have hung up on you. Since they (comcast) are the man-in-the-middle, they can easily perform these types of attacks.
And that's what this is. An attack. QOS would just slow things down, this kills. I don't mind QOS. I do mind active damage.
It's time to take p2p to the next level - implementing some of the concepts of the old freenet (the encryption part) and make the traffic unidentifiable. Maybe move it to UDP and make it look like DNS. Or Skype.
It's against the law in Mass. to resell a ticket for more than face value plus a small fee (which is like $2 or something...) That's why they were able to go after stubhub.
Have you ever been to any kind of sporting event with tickets? Clearly not. Your idea is completely unworkable. Many tickets are sold by people with extras right at the event. People in groups may also not all go in at the same time.
The fact is that it's PERFECTLY FINE to give your tickets to someone else, or to sell them. You just can't sell them for more than face value plus 2 dollars or whatever.
I had IDSL for a few years. Besides the fact that it is slow, I can't get it anymore even if I wanted it - not offered. Period. Hell, I spent 2 weeks trying to get a simple business ISDN-BRI line (because analog POTS sucketh) and finally gave up (in Verizon land...) No CLEC's in my area offer it either. Did I mention that ILEC's suck? And so do cable companies? I tried cable here and the quality was total crap. Massive number of outages ranging from a few minutes to a day in length all the time. Not to mention that it was dialup speeds from when the kids got out of school until 2am...
The state of high-speed internet in the US is simply horrible. How the fuck are we supposed to compete in the world market this next century with this situation? Europe / Asia are kicking our asses.
Adelphia was like that (horrible) *before* they went down, at least here. I've heard it's improved now, but I'll never go back to cable when DSL is super reliable (here time warner bought up adelphia though instead of comcast.)
Slashdot Mirror
Sell two versions. The first being an "upgrade for your OEM Apple hardware" at the same $129 / $199 family pack, and another "Full Retail" version for $300 or whatever.
The bigger issue is support. Apple doesn't want to support "random" hardware. It's a nightmare. Better to do a deal with HP / Dell / Etc. and only support a few "OS X compatible" models, and make the OEM offer support.
If your intent is to support Apple by giving them money for the OS, then you can still do so and ignore the EULA. If your intent is to just be "legal", well, then you are still SOL. But yeah, I hear you.
There are tools to normalize volume levels over your entire mp3 collection - no need to manually tweak each song. When I throw a party, I make a playlist. Occasionally I'll have guests that want a certain song - the ad-hoc queuing function in Amarok makes that trivial without screwing over my playlist.
Nah. It's a big enough market that there are tons of restaurants in that price range, and they are busy.
Nvidia doesn't need to sell 30M of these cards to turn a profit. If they sell 100K-200K world wide they do fine - and they will. If you can't afford a $200 card, don't buy one. But lots of hard-core gamers will save up for this kind of thing and buy it. Then you have the people buying on 22% interest store credit cards that buy them too even though they *can't* afford it.
How about that being a really bad analogy that doesn't work?
The fact is, kernel device developers have the skills, they just don't know the specific APIs. Learning those APIs is not NEARLY as hard as you seem to think it is. To claim it is so difficult is an insult to the talented people we are taking about.
This is about preference - not skills.
When I first got out of school (ancient history here,) I got a job doing industrial control systems. I didn't know jack shit about them, but I had the skills needed to learn what I didn't know. And it didn't take long. After two weeks learning the interfaces and new terminology, I was productive. Two months and I was project lead. My next job was client / server application code. Then custom TCP/IP, SNA / LU 6.2, CPIC, and 3270. In all cases, some re-learning was needed, but people with an IQ over 70 can learn new things pretty easily and get the work done.
We are not even talking the difference between embedded systems programming and mainframe networking. It's still device driver code in the SAME programming language.
"The Linux Kernel" is one thing. "A Linux based OS" is another. The fact is, the term "Linux" by itself usually (like 99% of the time) refers to "A Linux based OS" rather than just "The Linux Kernel." Sorry that you can't seem to join the 99 percentile.
The fact is, people want more devices to work on their "Linux based OS" no matter in which layer of the system the driver runs.
With 300 idle people, it seems that it would be a good use of resources to learn the interfaces necessary to write libusb drivers. And the statement that it is a different skill set is total fucking bullshit. It's the same SKILLS, but a different API. Unless you are trying to tell me that a "kernel developer", like Linus, can't write application code, such as Git. If I was a kernel device developer (and while I have played with some kernel device driver code a number of years ago, it's not what I do) I would be offended by your comment that I lack the skill set to write a user-space driver. I may PREFER kernel code, but it's not a lack of "skills" that keeps me from writing user-space code.
USB devices "can" be accessed from userspace via a common kernel shim. Think of USB as a mini-ethernet network. SSH doesn't need to be in kernel space in order to talk to other devices on the network... It uses the common kernel shim to talk to the network interface device, just like libusb and other shims talk to the USB network device.
That said, audio and video USB devices don't use these userspace shims (such as libusb.) I personally don't know why, but I'm sure there is a good reason. I'm also sure that some USB devices that currently have kernel drivers could instead use userspace drivers (IR blasters, v4l devices, audio devices, etc.) Other than "legacy" reasons, I can't see why they don't. Is it because not ALL audio devices are USB? If you moved everything beyond the very low-level access layer out of the kernel, it sure would be easier to debug and increase kernel stability / reduce kernel bloat.
I agree, but there is another issue here. the LDP (and Novell) is willing to work with manufacturers and sign NDAs in order to get info needed to write drivers. While USB scanners (and printers) don't require "kernel" drivers, they still require drivers and the same NDAs that traditional driver devices need in order to convince manufacturers to work with the developers.
Why aren't the LDP people and the SANE people working together? A device is a device no matter what the interface. The end user doesn't really care how "device X" hooks up, or how the driver is loaded. They just want it to work.
Would it? If a movie studio makes a film that has a song playing in the background of a scene, they have to get a license for that. These licenses can be expensive.
On the other hand, Mom's home video was not made as a for-profit work. I don't know if copyright law makes exceptions for that however.
IMHO, copyright law is WAY too strict.
Cows put out 500 liters of methane a day (that's per cow.) The obvious solution is to stick a tube up their asses and run a generator off of it. No need to make them walk. :-)
It's not about voip versus bittorrent - it's about real-time versus bulk and latency. It's about using resources the most efficiently. With shaping, you can reliably watch streaming youtube or porn while running your bittorrent, and everyone is happy. Shaping done right will hardly impact your overall transfer speeds.
The reality is that p2p is frequently consuming 70%+ of the bandwidth. If traffic isn't shaped at all, it makes the internet useless for anything *except* p2p. What comcast is doing isn't shaping. It's severe damage and fraud.
Anyway - so you don't use VoIP yet think it's a bad idea... Hmm. Well, I DO use VoIP, AND bittorrent, and I do think VoIP is a good idea. Not only is it less expensive, call quality is far superior, and my phone has a lot more capability. Any time I can pay Verizon or AT&T less money is a good thing. I'll go with my years of experience with VoIP over your uninformed opinion any day.
Um, the reason they will switch is that they will regain control of their systems that they have lost with Windows. What's your plan for the future - hope that Microsoft reverses course and starts to think about the needs of their customers? Fat chance of that...
Ever file anything in small claims court against a large company? Not so easy. Just "serving notice" can be a challenge.
And some of us have been doing just fine on alternative platforms for many many years. It's quite silly to claim that you can't run a business without Windows. The fact is, you can. I grant that it can be difficult to switch, but those of use running on alternative platforms for years prove that it can be done.
Until people just start biting the bullet and switching, no competition will EVER have a larger market share no matter HOW good their product.
The modern trick is to use VM technology to ease the transition. Eventually you will get to the point where most of your users won't need windows. Then you can work with your software vendors (or NEW software vendors) to fill in the gaps. As a result, competition IS starting to become viable for the cases where it hadn't been.
Hasn't anyone read the Windows EULA?
Microsoft has the right to do anything it wants to your machine, including totally disabling it. They also have the right to spy on you, reporting anything they want back to MS.
Of course back when Windows 2000 came out and all those new terms were introduced into the EULA, some of us raised red flags. Others "knew better" - MS would never actually DO the things the new EULA allowed....
Some of us still just didn't have that level of trust. We started migrating all our critical systems to alternative platforms, or didn't buy into the MS crack pipe in the first place.
So no, there is no conspiracy theory. MS opened the door for this behavior long ago, and just had the patience to slowly enforce the rights they guaranteed themselves. Why is anyone surprised or angry?
"Oh My God, they killed init! You bastards!"
The overhead for the encrypted traffic would greatly increase the amount of data being used.
Not really. If I encrypt a 1K packet, it's still 1K. There is a MINOR bit of overhead negotiating the keys up front, but it's small enough that it would not be noticeable. If you want to ensure that the wrapper isn't tampered with (the IP packet header) you can add additional checks in the data portion of the packet. This would add about 0.5% additional overhead. Still not a big deal. Authenticate all clients / trackers with a cert from cacert.org and MITM attacks are thwarted.
All they have to do is respect the existing TOS bits. The applications set those.
Shaping is fine. Mangling / blocking is not.
Frankly, I would prefer that my ISP have some sort of QoS so that my bulk traffic is at a lower priority than VoIP. Wouldn't you???
From what I understand, they are forging packets that make your BT client think that peers have hung up on you. Since they (comcast) are the man-in-the-middle, they can easily perform these types of attacks.
And that's what this is. An attack. QOS would just slow things down, this kills. I don't mind QOS. I do mind active damage.
It's time to take p2p to the next level - implementing some of the concepts of the old freenet (the encryption part) and make the traffic unidentifiable. Maybe move it to UDP and make it look like DNS. Or Skype.
They are selling to people in MA for events in MA. CLEARLY they are doing business in MA - it doesn't matter where their offices are.
It's against the law in Mass. to resell a ticket for more than face value plus a small fee (which is like $2 or something...) That's why they were able to go after stubhub.
Have you ever been to any kind of sporting event with tickets? Clearly not. Your idea is completely unworkable. Many tickets are sold by people with extras right at the event. People in groups may also not all go in at the same time.
The fact is that it's PERFECTLY FINE to give your tickets to someone else, or to sell them. You just can't sell them for more than face value plus 2 dollars or whatever.
I had IDSL for a few years. Besides the fact that it is slow, I can't get it anymore even if I wanted it - not offered. Period. Hell, I spent 2 weeks trying to get a simple business ISDN-BRI line (because analog POTS sucketh) and finally gave up (in Verizon land...) No CLEC's in my area offer it either. Did I mention that ILEC's suck? And so do cable companies? I tried cable here and the quality was total crap. Massive number of outages ranging from a few minutes to a day in length all the time. Not to mention that it was dialup speeds from when the kids got out of school until 2am...
The state of high-speed internet in the US is simply horrible. How the fuck are we supposed to compete in the world market this next century with this situation? Europe / Asia are kicking our asses.
Adelphia was like that (horrible) *before* they went down, at least here. I've heard it's improved now, but I'll never go back to cable when DSL is super reliable (here time warner bought up adelphia though instead of comcast.)