We started experimenting with the compressed memory module (zram) in Chrome OS in 2012, and deployed it fully in 2013. It's a simple approach: the kernel uses a compressed RAM disk as a swap device. It works pretty well. On average we store about 3 compressed pages into one page.
My daughter was quite interested in this for a while, but there is one serious problem: they are making a lot of changes at once, and evaluating the results will not be easy, especially with such a small sample of students who, by self-selection, are going to be anything but representative of the rest (for one thing, they are going to be big risk takers). It will take years to see how well this works, considering how difficult it already is to evaluate the quality of the education at various colleges.
I don't know how much these considerations influenced my daughter, but she ended up picking a conventional college, partly because she applied Early Decision and got in. Minerva might have been on her list for a second round. (And yes, she is a risk taker, and not interested in Greek life or football:)
The founders are smart people and what they say makes sense, but I know many smart people who made a lot of sense, and their startups still didn't quite work out.
This may be just Italy. My experience with the other countries is skimpy at best. But my experience as a consumer in Italy is vastly inferior to the USA---except, as I said, as far as mobile phones are concerned. Never mind trying to buy overpriced aspirin on a Sunday or a Wednesday afternoon. Most store owners' attitude is that they are doing you a favor by making their merchandise available to you. Forget about returning an item for refund, two hours after you buy it, and still in the unopened original package. You're lucky if you get store credit after invoking a lifetime of friendship between the store owner's aunt and your grandmother. Store policies like Costco's are a DREAM in Italy.
I won't claim things are perfect here. With cell phones, I am as annoyed as anybody else who understands what goes on. In fact, until recently my phone was an unlocked Nokia bought in Italy which I used in both countries, swapping the SIM. Yet, I now own an iPhone. It's a tradeoff that works for me. If it eventually gets unlocked, that's good. If not, I'll keep using the Nokia in Italy---no big deal.
I don't know what the consumer laws are in Italy, but local culture is also a factor. I don't know anybody who spends time in both Italy and USA and disagrees with my view. Italy is nice for a lot of things. This isn't one of them.
With some exceptions, most notably the cell phone industry, Europe has far less consumer protection than the US, and far more anticompetitive regulations. For instance, try to buy cold medicine at the Venice airport on a Sunday.
> Could you provide technical information? Many companies have great ethics, > that doesn't prove whether or not they're violating a software license.
Ethics + smarts. You do something illegal either when you're too stupid to know the difference, or when you know it well but don't care and think you can get away with it.
> Is it possible to run the binary vmkernel blob ESX loads from the Linux > kernel, without using Linux to load it?
Again, just my opinion here, based on public information.
It seems pretty clear that ESX relies on Linux to bootstrap. But going from "ESX needs Linux to run" to "ESX is derived from Linux" is wrong. May I point out that any application running on Linux "needs" Linux to run. The difference here is that ESX (as well as the other VMware products) don't limit themselves to the standard system call interface. So what? The letter and spirit of the GPL and LGPL are that it is permissible to exploit commercially the functionality of free software, as long as one doesn't incorporate or extend its source in a non-free way. This was a calculated and reasonable compromise to help further the adoption of GNU/Linux.
The hosted products (Workstation and GSX) have pretty much the same issue. The host OS functionality they use goes beyond the standard system call interface. You can get a fairly good idea of it by looking at the (public) Linux driver. From a licensing standpoint, there is little difference between the hosted and the vmkernel-based products. They use Linux (or Windows) to bootstrap and run. None is "derived work" from Linux or Windows.
I am a former VMware programmer. Obviously I do not speak for the company, just myself.
VMware is not infringing anything. First, they have high standards of ethics. Even if they didn't, they would be too smart for that. When ESX was designed, there were other choices for the console OS, FreeBSD for instance. But they figured out that using Linux was legal and did so. Both VMware and Linux benefit from this. Yes, it is not a "standard", well-understood relationship such as running some app on top of the kernel. But it respects the technical aspects of the license and I believe its spirit as well (although my interpretation of the "spirit" may differ from yours).
One could argue that Linux benefits more from VMware than the other way around. In many cases VMware ESX introduced Linux to corporate data centers that wanted nothing to do with it. The sales people had to work hard to convince potential customers that the product was NOT running on Linux, that Linux was just running in a separate VM to help along with various tasks.
Linux is also helped by the fact that virtual machines offer a low-cost way of experimenting with new systems, and add a layer of freedom in the conservative corporate IT environment.
As to whether VMware should be free software, there are situations for which free software is just not the right model and VMware is a good example. In the early years of the company, someone tried to start a competing free-software product (at some point called Freemware) but it didn't go far. VMware is a large (huge) system. It took a lot of unglamorous work from a lot of people under the same roof to bring it to life. It was almost a miracle that it would run. It stressed CPUs in truly novel ways. (The programmers hit and had to work around previously unknown bugs in the CPU.) I, the eternal pessimist, feared that we'd never be able to make it stable enough for a viable product. Fortunately I was wrong, and in any case Windows was a lot less stable than VMware those days, so it didn't matter that much.
That theory amuses me too, but apparently there is plenty of high-energy events on Earth from cosmic rays hitting the atmosphere, so these particle accelerators aren't really doing anything particularly unusual (for now), just making things happen in places where we can observe them more easily.
I tried to figure out why this upsets me so much and I think I know why.
Believing in the natural selection mechanisms (recall that Darwin never called it "evolution") requires a certain amount of faith in the system. We can all go to museums and look at fossils. Indeed it's not that hard to go to a mountain and find fossils yourself. But most people, including me, will never be able to date geological layers based on percentages of radioactive isotopes, or other esoteric methods. For that, I trust the system. I trust that the information that comes from our universities and research labs is generally correct. Why? Because it has proven to be correct over and over. Because we wouldn't see the kind of progress that we see if the system didn't work. There is no conspiracy. There is no bias. The system seeks truth and produces truth.
When people don't believe in the system, for whatever reason, they reject the foundations of human progress. If they don't believe in natural selection, then they have no reason to believe in anything else that the system produces. When they teach others that natural selection may be wrong, they are teaching them to distrust the system. If natural selection is wrong, then anything else may be wrong too. Vaccinations, for instance, may be harmful. Chlorofluorocarbons don't really damage the ozone layer. HIV doesn't cause AIDS.
So the problem is not the waste of resources in teaching something that is wrong and useless. That is actually pretty marginal. The real damage is caused by the legitimization of the mistrust of our academic and scientific communities.
Frankly I would not be surprised if the pain and suffering caused by this mistrust exceeded that caused by fundamentalist Muslim terrorists by a large factor.
But a common use of virtual machines is for server consolidation, where several lightly loaded virtual machines run on the same CPU. This is useful when the services run on different OS versions, or use different DLLs, or cannot run under the same OS for some other weird incompatibility reason. In this case, Microsoft will be able to charge more money. I don't know which case is more common, but this is certainly a blow to server consolidation.
Trigonometric functions are far more important for their differential properties than their geometrical properties. When you'll study calculus, you will find out the surprising relationship between the exponential function and trigonometric functions (with complex numbers, they are essentially the same thing). Some of the most powerful mathematical tools for several branches of engineering and physics (Fourier transforms, for instance) use sines and cosines left and right for reasons that have nothing to do with trigonometry.
Without understanding the technical issues, I am giving this guy the benefit of the doubt. He is only going up to 30 miles (the Shuttle can reach 400 miles) and will be weightless for a fairly short time. Reentry is probably a relatively small deal. The price tag seems quite low (a small Cessna costs that much) but then again this isn't exactly space exploration, it's more like manned amateur rocketry.
Slashdot Mirror
We started experimenting with the compressed memory module (zram) in Chrome OS in 2012, and deployed it fully in 2013. It's a simple approach: the kernel uses a compressed RAM disk as a swap device. It works pretty well. On average we store about 3 compressed pages into one page.
My daughter was quite interested in this for a while, but there is one serious problem: they are making a lot of changes at once, and evaluating the results will not be easy, especially with such a small sample of students who, by self-selection, are going to be anything but representative of the rest (for one thing, they are going to be big risk takers). It will take years to see how well this works, considering how difficult it already is to evaluate the quality of the education at various colleges.
I don't know how much these considerations influenced my daughter, but she ended up picking a conventional college, partly because she applied Early Decision and got in. Minerva might have been on her list for a second round. (And yes, she is a risk taker, and not interested in Greek life or football :)
The founders are smart people and what they say makes sense, but I know many smart people who made a lot of sense, and their startups still didn't quite work out.
There is a free iPhone app called Fring that I've successfully used to make SkypeOut calls over my home WiFi network.
This may be just Italy. My experience with the other countries is skimpy at best. But my experience as a consumer in Italy is vastly inferior to the USA---except, as I said, as far as mobile phones are concerned. Never mind trying to buy overpriced aspirin on a Sunday or a Wednesday afternoon. Most store owners' attitude is that they are doing you a favor by making their merchandise available to you. Forget about returning an item for refund, two hours after you buy it, and still in the unopened original package. You're lucky if you get store credit after invoking a lifetime of friendship between the store owner's aunt and your grandmother. Store policies like Costco's are a DREAM in Italy.
I won't claim things are perfect here. With cell phones, I am as annoyed as anybody else who understands what goes on. In fact, until recently my phone was an unlocked Nokia bought in Italy which I used in both countries, swapping the SIM. Yet, I now own an iPhone. It's a tradeoff that works for me. If it eventually gets unlocked, that's good. If not, I'll keep using the Nokia in Italy---no big deal.
I don't know what the consumer laws are in Italy, but local culture is also a factor. I don't know anybody who spends time in both Italy and USA and disagrees with my view. Italy is nice for a lot of things. This isn't one of them.
With some exceptions, most notably the cell phone industry, Europe has far less consumer protection than the US, and far more anticompetitive regulations. For instance, try to buy cold medicine at the Venice airport on a Sunday.
> Could you provide technical information? Many companies have great ethics,
> that doesn't prove whether or not they're violating a software license.
Ethics + smarts. You do something illegal either when you're too stupid to know the difference, or when you know it well but don't care and think you can get away with it.
> Is it possible to run the binary vmkernel blob ESX loads from the Linux
> kernel, without using Linux to load it?
Again, just my opinion here, based on public information.
It seems pretty clear that ESX relies on Linux to bootstrap. But going from "ESX needs Linux to run" to "ESX is derived from Linux" is wrong. May I point out that any application running on Linux "needs" Linux to run. The difference here is that ESX (as well as the other VMware products) don't limit themselves to the standard system call interface. So what? The letter and spirit of the GPL and LGPL are that it is permissible to exploit commercially the functionality of free software, as long as one doesn't incorporate or extend its source in a non-free way. This was a calculated and reasonable compromise to help further the adoption of GNU/Linux.
The hosted products (Workstation and GSX) have pretty much the same issue. The host OS functionality they use goes beyond the standard system call interface. You can get a fairly good idea of it by looking at the (public) Linux driver. From a licensing standpoint, there is little difference between the hosted and the vmkernel-based products. They use Linux (or Windows) to bootstrap and run. None is "derived work" from Linux or Windows.
I am a former VMware programmer. Obviously I do not speak for the company, just myself.
VMware is not infringing anything. First, they have high standards of ethics. Even if they didn't, they would be too smart for that. When ESX was designed, there were other choices for the console OS, FreeBSD for instance. But they figured out that using Linux was legal and did so. Both VMware and Linux benefit from this. Yes, it is not a "standard", well-understood relationship such as running some app on top of the kernel. But it respects the technical aspects of the license and I believe its spirit as well (although my interpretation of the "spirit" may differ from yours).
One could argue that Linux benefits more from VMware than the other way around. In many cases VMware ESX introduced Linux to corporate data centers that wanted nothing to do with it. The sales people had to work hard to convince potential customers that the product was NOT running on Linux, that Linux was just running in a separate VM to help along with various tasks.
Linux is also helped by the fact that virtual machines offer a low-cost way of experimenting with new systems, and add a layer of freedom in the conservative corporate IT environment.
As to whether VMware should be free software, there are situations for which free software is just not the right model and VMware is a good example. In the early years of the company, someone tried to start a competing free-software product (at some point called Freemware) but it didn't go far. VMware is a large (huge) system. It took a lot of unglamorous work from a lot of people under the same roof to bring it to life. It was almost a miracle that it would run. It stressed CPUs in truly novel ways. (The programmers hit and had to work around previously unknown bugs in the CPU.) I, the eternal pessimist, feared that we'd never be able to make it stable enough for a viable product. Fortunately I was wrong, and in any case Windows was a lot less stable than VMware those days, so it didn't matter that much.
Luigi
That theory amuses me too, but apparently there is plenty of high-energy events on Earth from cosmic rays hitting the atmosphere, so these particle accelerators aren't really doing anything particularly unusual (for now), just making things happen in places where we can observe them more easily.
I tried to figure out why this upsets me so much and I think I know why.
Believing in the natural selection mechanisms (recall that Darwin never called it "evolution") requires a certain amount of faith in the system. We can all go to museums and look at fossils. Indeed it's not that hard to go to a mountain and find fossils yourself. But most people, including me, will never be able to date geological layers based on percentages of radioactive isotopes, or other esoteric methods. For that, I trust the system. I trust that the information that comes from our universities and research labs is generally correct. Why? Because it has proven to be correct over and over. Because we wouldn't see the kind of progress that we see if the system didn't work. There is no conspiracy. There is no bias. The system seeks truth and produces truth.
When people don't believe in the system, for whatever reason, they reject the foundations of human progress. If they don't believe in natural selection, then they have no reason to believe in anything else that the system produces. When they teach others that natural selection may be wrong, they are teaching them to distrust the system. If natural selection is wrong, then anything else may be wrong too. Vaccinations, for instance, may be harmful. Chlorofluorocarbons don't really damage the ozone layer. HIV doesn't cause AIDS.
So the problem is not the waste of resources in teaching something that is wrong and useless. That is actually pretty marginal. The real damage is caused by the legitimization of the mistrust of our academic and scientific communities.
Frankly I would not be surprised if the pain and suffering caused by this mistrust exceeded that caused by fundamentalist Muslim terrorists by a large factor.
But a common use of virtual machines is for server consolidation, where several lightly loaded virtual machines run on the same CPU. This is useful when the services run on different OS versions, or use different DLLs, or cannot run under the same OS for some other weird incompatibility reason. In this case, Microsoft will be able to charge more money. I don't know which case is more common, but this is certainly a blow to server consolidation.
Trigonometric functions are far more important for their differential properties than their geometrical properties. When you'll study calculus, you will find out the surprising relationship between the exponential function and trigonometric functions (with complex numbers, they are essentially the same thing). Some of the most powerful mathematical tools for several branches of engineering and physics (Fourier transforms, for instance) use sines and cosines left and right for reasons that have nothing to do with trigonometry.
My friend Boris informed me that Costco is selling one of these.
Without understanding the technical issues, I am giving this guy the benefit of the doubt. He is only going up to 30 miles (the Shuttle can reach 400 miles) and will be weightless for a fairly short time. Reentry is probably a relatively small deal. The price tag seems quite low (a small Cessna costs that much) but then again this isn't exactly space exploration, it's more like manned amateur rocketry.