1. When has Google sued or threatened to sue anyone over patent violations?
Google will sue or threaten to sue when Google is threatened. In a for-profit company, this isn't a choice -- you have to protect your shareholders' assets within the full extent of the law, or you will be held liable. They just haven't really been threatened by anyone so far, but in the long run this is inevitable for any company.
Microsoft, on the other hand, are always threatened by higher quality or cheaper software that competes with their own software and/or erodes their monopoly.
I think the idea is rather to distribute the impact over a larger area of your body or whatever is being protected. That's why for instance a shin guard is hard (though padded on the inside). The problem with hard protection is that it inhibits your movement -- you don't want to go downhill skiing in a Star Wars stormtrooper armour. So if the protection is soft whenever moving normally it's a big advantage.
Description: An implementation issue exists in the kernel's handling of fcntl system calls. A local user may overwrite kernel memory and execute arbitrary code with system privileges. This update addresses the issue through improved handling of fcntl system calls. Credit to Razvan Musaloiu-E. of Johns Hopkins University, HiNRG for reporting this issue. [Emphasis mine]
If you have the ability to alter kernel memory at an arbitrary place, you can accomplish pretty much anything. An exploit could for instance modify some critical kernel data structure, or replace a function pointer or return address, thus allowing the exploiter to inject code and make the kernel run it in privileged mode.
If you're downloading music at work, it probably is stealing of company time. And given that my taxes are paying these people's salaries (that is, you and I are "the company"), I'd really rather them not.
If you want people to be efficient and motivated, you need to give them some slack. Arguing about a couple of minutes here and there is just small-minded and counterproductive if it improves their good spirits. Besides, nobody is active 100% of their time -- if you don't download a song you'll stare out of the window now and then or do something else to take a mental break.
Ah, but I am not claiming to have a metric for which OS is better. I'm saying that if you want to compare upgrade costs, then I think the most sensible way is to do what I've described. Like you said, then you have to weight the costs against the benefits. I'm suggesting a way to find (some of) the cost part of that -- not the benefit/business value part. And for that I would expect such a number to be quite helpful, especially for businesses.
I don't disagree with you, and I'm certainly not trying to argue for buying Windows over OS X. (I'm a Linux user myself, but I'd probably own a Mac if I weren't.)
But if you're going to have any metric for the upgrade costs of operating systems, what else would it be? It doesn't mean you have to upgrade. Of course you need to weight costs and benefits at each particular instance where you are considering an upgrade, but if I want to get an idea what it's going to cost me to keep my system updated over time I need to make some sort of estimation.
For me, the argument could go like: Well, upgrading OS X is a little more expensive, but I don't expect to always need the newest version, so I will probably skip every other upgrade. Looking back at how often there has been a new release and what price it has had, I expect my cost over five years to be $X. This figure is useful to me. If the "just works" feeling of Apple products or whatever other added benefit I see is worth this money is a different question that I'll have to ask myself next.
Buying a new computer complicates it a bit, but not too much considering that the cost for the operating system is baked into the price of the computer.
there are several editions of Windows for the desktop and several versions for the server, whereas OSX has only one of each. Which do you compare to?
Well, you'd have to choose one to compare to. Depending on your needs, I guess.
I would instead suggest comparing the amount of benefit per dollar. It would also be a very complicated thing to figure out [...]
Not to mention entirely subjective. But yes of course, you should do that kind of comparison before buying either one. My point was that starting to define what a service pack is as opposed to a new version, and which version number group that means what, is a rather pointless exercise.
In each case, I would weigh the cost against the benefit I'd get from the new features, and decide whether it was worth it to me.
Yes, and the "upgrade cost per year" gives you the cost part of that equation.
I would say the relevant question if you want to compare Windows and OS X upgrade costs is: How often to you have to pay to keep using the newest version (measured in calendar days), and how much do you have to pay each time? That'll give you an average cost per year figure. The version number scheme doesn't really matter.
Furthermore, during our 40 years or so, we were not supposed to eat food containing refined sugar (candy, cereals, pastries etc.) and drink fluids containing various acids (soda, fruit juice).
Honestly, I think the semantic discussion is rather uninteresting. (And for the record, I never said Android isn't open software -- quite the opposite actually.) I'll give you that it is a fact that Android is an open Linux platform according to your definition of an open Linux platform, whatever that may be, and then we can leave that at that.
Now, let's go back to the main point:
Do you, or do you not agree that there is a noteworthy difference between Maemo and Android, in terms of developing or porting common Linux software for it?
Can you accept that the ability to reflash your individual device and replace a significant portion of its software stack is not equivalent to buying a device which in its stock configuration already supports your development environment without modification (and hence including all other units sold on the market)?
If you can agree with those two points I am satisfied.
You are pretty arrogant to assume that everyone that disagrees with you are just ignorant or inexperienced. (In fact, this is starting to feel like I'm feeding a troll.)
So basically you are saying that Android is an open Linux platform because you can, well, reflash the device, completely remove Android and put something else in there? Using the same logic I could say that Windows 7 is a Linux system, you just have to wipe the hard drive and install Ubuntu. That simply doesn't make sense to me, and it is not due to lack of experience.
I honestly have a little trouble following your argumentation here.
Can't you see that there is a difference between a common GNU/Linux distribution and a highly customized Java oriented software platform that happens to use the Linux kernel underneath? Don't you think it could be reasonable to make a distinction between the two at all?
The fact that the source code is available for Android devices doesn't automatically make it an open Linux platform in any practical sense. You are not meant to hack root on a HTC Dream, and even if you manage to it's a very non-standard thing to do.
Sure, should I hack it and effectively replace the Android distribution with a hand rolled Linux distribution of my own as you are suggesting, I would be able to develop on it like an ordinary GNU/Linux/X11 system. But that would be rather pointless because I'd be the only person in the world with such a phone. Not to mention that it would require many man-years of work to recreate all the functionality that would no longer work or to make the Andriod software stack interoperate with all the new applications.
By all means, I agree that Android is an open software platform. But failing to see that Maemo is something quite different when it comes to being a Linux device platform is just fooling yourself, in my humble but considered opinion.
Well, the fact that you can hack a some Andriod based phones doesn't exactly make the platform open in that respect. Furthermore, if I am to write software and share it with people who are not geeks and/or care for their warranty, I have to stick with the API:s provided in the "stock version".
I used to imagine desktop computers of the future as a bowl of small spheres, communicating through their contact surfaces. Whenever you need to upgrade for more processing power, just buy a bag of extra spheres and pour them into the bowl. Could be stylish as well.
As a buyer, I got screwed on what turned out to be faked brand-name items from someone in China with an invalid return address [...] Paypal never gave me a refund because my item was "never actually returned to seller" even though the seller was at fault for giving a bad address.
So, you received something you didn't order from someone in a country you never ordered from. I would say the item was "never actually sent to the buyer"...
Linux and Firefox have already shown that enough volunteers are out there to produce software that gets you to that web for FREE. When a free product will do what you want it to people won't continue to shell out mega bucks for windows over and over.
Exactly. Or to put it in other words, what happens is that operating systems become commoditized. And with such an easily reproducible product as software is, this means extremely low profit margins.
Microsoft have always been able to rely on Windows and Office for income. And, using Windows as leverage, they have managed to control other markets as well. With the current development, they could be losing a fair portion of the revenue stream from one of these cash cows.
Now, will this kill Microsoft? Probably not. Others have proven that it is possible to make good software with a fraction of Microsoft's revenue, and Microsoft have survived crises before. But they might have to change and become more efficient -- hopefully for the better.
As web pages are progressing to become more and more like applications, where the browser serves as a standardized virtual machine, speed will become an issue (if it isn't already -- slashdot is rather sluggish on my box running Firefox 3.0). This includes both java script performance, which Google have been emphasizing with Chrome, as well as rendering performance.
I mean, to a company now the choice is between "patent rejected, pay nothing" and "patent approved, pay for it", its a zero risk game that can result in significant benefits.
More like "patent rejected, $20k wasted". The cost of a patent application for a typical company is usually many times higher than the fee you pay to the patent office. Patent applications are complex legal documents, and patent attorney fees are usually a substantial portion of the total cost. You will also need to have your technical specialists to spend time consulting the attorneys. And you may need to perform a preliminary patent search, which means even more expensive hours spent.
It isn't important who is right. However, as a customer it is very important to know when you are presented with excuses rather than reasons, if you are going to be able to use your "consumer power". I can see that you think I'm nit-picking, but I do believe there is quite a significant difference between "well, they have this ability to offer you some nice services" and "they want this ability to have better control over your device".
What I suggested is not particularily complicated in the realm of DRM technology. I'm guessing whatever hardware/protocols/algorithms they're using for the remote deletion are equally complicated or expensive to implement, if not more.
Nevertheless, you said that a refund system is impossible without remote deletion, as a defense for it. That statement is decidedly false, and doesn't justify that they have such control over customers devices.
Unfortunately, I agree with you that the short-term economical incentive to do what I propose is weak at the moment, since only "geeks" really understand what they are buying and what other possibilities there are. So the reason for them choose a less intrusive technology isn't increased sales as much as simple honesty.
Sadly, the corporate ideals always seem to be to grab as much power and control as possible at every opportunity, and never to give anything away unless forced to.
- You request a refund.
- They check that they have received the payment, and send you an electronic token.
- A chip in the Kindle creates a list of file hashes and signs it together with the token using a built-in key.
- You send the signed list back.
- They now know the file didn't exist at a time after which they sent you the token, and can send you the refund.
All that is needed is that a cryptography chip is installed in the Kindle, and this chip only needs to have read access to the memory. You are the only one with write access. Also, they don't get to see the contents of your files, because all you send them is hashes. They can verify that you have or haven't got a specific file (if they already know its hash key), but that's all. And they can only do this with your assistance.
Concepts are a complex feature, and C++ is an (overly) complex language.
Concepts is also a feature that could help managing the complexity of C++, for instance making template libraries much easier to use by giving sane compiler errors. You might still have to be an expert to write a GP library, but you shouldn't have to be one to use such a library.
Slashdot Mirror
Google will sue or threaten to sue when Google is threatened. In a for-profit company, this isn't a choice -- you have to protect your shareholders' assets within the full extent of the law, or you will be held liable. They just haven't really been threatened by anyone so far, but in the long run this is inevitable for any company.
Microsoft, on the other hand, are always threatened by higher quality or cheaper software that competes with their own software and/or erodes their monopoly.
Imagine the anticlimax when the food supplies run out.
I think the idea is rather to distribute the impact over a larger area of your body or whatever is being protected. That's why for instance a shin guard is hard (though padded on the inside). The problem with hard protection is that it inhibits your movement -- you don't want to go downhill skiing in a Star Wars stormtrooper armour. So if the protection is soft whenever moving normally it's a big advantage.
If you have the ability to alter kernel memory at an arbitrary place, you can accomplish pretty much anything. An exploit could for instance modify some critical kernel data structure, or replace a function pointer or return address, thus allowing the exploiter to inject code and make the kernel run it in privileged mode.
If you want people to be efficient and motivated, you need to give them some slack. Arguing about a couple of minutes here and there is just small-minded and counterproductive if it improves their good spirits. Besides, nobody is active 100% of their time -- if you don't download a song you'll stare out of the window now and then or do something else to take a mental break.
Ah, but I am not claiming to have a metric for which OS is better. I'm saying that if you want to compare upgrade costs, then I think the most sensible way is to do what I've described. Like you said, then you have to weight the costs against the benefits. I'm suggesting a way to find (some of) the cost part of that -- not the benefit/business value part. And for that I would expect such a number to be quite helpful, especially for businesses.
I don't disagree with you, and I'm certainly not trying to argue for buying Windows over OS X. (I'm a Linux user myself, but I'd probably own a Mac if I weren't.)
But if you're going to have any metric for the upgrade costs of operating systems, what else would it be? It doesn't mean you have to upgrade. Of course you need to weight costs and benefits at each particular instance where you are considering an upgrade, but if I want to get an idea what it's going to cost me to keep my system updated over time I need to make some sort of estimation.
For me, the argument could go like: Well, upgrading OS X is a little more expensive, but I don't expect to always need the newest version, so I will probably skip every other upgrade. Looking back at how often there has been a new release and what price it has had, I expect my cost over five years to be $X. This figure is useful to me. If the "just works" feeling of Apple products or whatever other added benefit I see is worth this money is a different question that I'll have to ask myself next.
Buying a new computer complicates it a bit, but not too much considering that the cost for the operating system is baked into the price of the computer.
Well, you'd have to choose one to compare to. Depending on your needs, I guess.
Not to mention entirely subjective. But yes of course, you should do that kind of comparison before buying either one. My point was that starting to define what a service pack is as opposed to a new version, and which version number group that means what, is a rather pointless exercise.
Yes, and the "upgrade cost per year" gives you the cost part of that equation.
I would say the relevant question if you want to compare Windows and OS X upgrade costs is: How often to you have to pay to keep using the newest version (measured in calendar days), and how much do you have to pay each time? That'll give you an average cost per year figure. The version number scheme doesn't really matter.
Furthermore, during our 40 years or so, we were not supposed to eat food containing refined sugar (candy, cereals, pastries etc.) and drink fluids containing various acids (soda, fruit juice).
Honestly, I think the semantic discussion is rather uninteresting. (And for the record, I never said Android isn't open software -- quite the opposite actually.) I'll give you that it is a fact that Android is an open Linux platform according to your definition of an open Linux platform, whatever that may be, and then we can leave that at that.
Now, let's go back to the main point:
If you can agree with those two points I am satisfied.
You are pretty arrogant to assume that everyone that disagrees with you are just ignorant or inexperienced. (In fact, this is starting to feel like I'm feeding a troll.)
So basically you are saying that Android is an open Linux platform because you can, well, reflash the device, completely remove Android and put something else in there? Using the same logic I could say that Windows 7 is a Linux system, you just have to wipe the hard drive and install Ubuntu. That simply doesn't make sense to me, and it is not due to lack of experience.
I honestly have a little trouble following your argumentation here.
Can't you see that there is a difference between a common GNU/Linux distribution and a highly customized Java oriented software platform that happens to use the Linux kernel underneath? Don't you think it could be reasonable to make a distinction between the two at all?
The fact that the source code is available for Android devices doesn't automatically make it an open Linux platform in any practical sense. You are not meant to hack root on a HTC Dream, and even if you manage to it's a very non-standard thing to do.
Sure, should I hack it and effectively replace the Android distribution with a hand rolled Linux distribution of my own as you are suggesting, I would be able to develop on it like an ordinary GNU/Linux/X11 system. But that would be rather pointless because I'd be the only person in the world with such a phone. Not to mention that it would require many man-years of work to recreate all the functionality that would no longer work or to make the Andriod software stack interoperate with all the new applications.
By all means, I agree that Android is an open software platform. But failing to see that Maemo is something quite different when it comes to being a Linux device platform is just fooling yourself, in my humble but considered opinion.
Well, the fact that you can hack a some Andriod based phones doesn't exactly make the platform open in that respect. Furthermore, if I am to write software and share it with people who are not geeks and/or care for their warranty, I have to stick with the API:s provided in the "stock version".
I used to imagine desktop computers of the future as a bowl of small spheres, communicating through their contact surfaces. Whenever you need to upgrade for more processing power, just buy a bag of extra spheres and pour them into the bowl. Could be stylish as well.
So, you received something you didn't order from someone in a country you never ordered from. I would say the item was "never actually sent to the buyer"...
Exactly. Or to put it in other words, what happens is that operating systems become commoditized. And with such an easily reproducible product as software is, this means extremely low profit margins.
Microsoft have always been able to rely on Windows and Office for income. And, using Windows as leverage, they have managed to control other markets as well. With the current development, they could be losing a fair portion of the revenue stream from one of these cash cows.
Now, will this kill Microsoft? Probably not. Others have proven that it is possible to make good software with a fraction of Microsoft's revenue, and Microsoft have survived crises before. But they might have to change and become more efficient -- hopefully for the better.
As web pages are progressing to become more and more like applications, where the browser serves as a standardized virtual machine, speed will become an issue (if it isn't already -- slashdot is rather sluggish on my box running Firefox 3.0). This includes both java script performance, which Google have been emphasizing with Chrome, as well as rendering performance.
More like "patent rejected, $20k wasted". The cost of a patent application for a typical company is usually many times higher than the fee you pay to the patent office. Patent applications are complex legal documents, and patent attorney fees are usually a substantial portion of the total cost. You will also need to have your technical specialists to spend time consulting the attorneys. And you may need to perform a preliminary patent search, which means even more expensive hours spent.
It isn't important who is right. However, as a customer it is very important to know when you are presented with excuses rather than reasons, if you are going to be able to use your "consumer power". I can see that you think I'm nit-picking, but I do believe there is quite a significant difference between "well, they have this ability to offer you some nice services" and "they want this ability to have better control over your device".
What I suggested is not particularily complicated in the realm of DRM technology. I'm guessing whatever hardware/protocols/algorithms they're using for the remote deletion are equally complicated or expensive to implement, if not more.
Nevertheless, you said that a refund system is impossible without remote deletion, as a defense for it. That statement is decidedly false, and doesn't justify that they have such control over customers devices.
Unfortunately, I agree with you that the short-term economical incentive to do what I propose is weak at the moment, since only "geeks" really understand what they are buying and what other possibilities there are. So the reason for them choose a less intrusive technology isn't increased sales as much as simple honesty.
Sadly, the corporate ideals always seem to be to grab as much power and control as possible at every opportunity, and never to give anything away unless forced to.
Specifically, it could work like this:
- You request a refund.
- They check that they have received the payment, and send you an electronic token.
- A chip in the Kindle creates a list of file hashes and signs it together with the token using a built-in key.
- You send the signed list back.
- They now know the file didn't exist at a time after which they sent you the token, and can send you the refund.
All that is needed is that a cryptography chip is installed in the Kindle, and this chip only needs to have read access to the memory. You are the only one with write access. Also, they don't get to see the contents of your files, because all you send them is hashes. They can verify that you have or haven't got a specific file (if they already know its hash key), but that's all. And they can only do this with your assistance.
Sure they could. They only need a mechanism for verifying that the book is no longer on the device, and leave the deleting to the owner.
Concepts is also a feature that could help managing the complexity of C++, for instance making template libraries much easier to use by giving sane compiler errors. You might still have to be an expert to write a GP library, but you shouldn't have to be one to use such a library.
For a while I was thinking: "Heck, that guy must be writing mighty long lines when he is coding!"...