Oh, and BMW drivers are invariably prats who tailgate you for as long as possible, then overtake and cut you up.
Heh-heh. I drive a BMW now and I do drive aggressively. Of course, I drove the same way in my twin-turbo Supra, my 5.0L Mustang, and every other car I've owned.
Sorry, please don't even both spending mod points to bury this. I just thought it amusing that someone would paint with such a broad brush across a fairly large segment of adults.:P
I've been teaching Linux Kernel Internals and Linux Device Driver courses (among others) since 1995 and other Un*x topics prior to that.
Many of the big names "outsource" their very technical training to third parties unless the topic is something related to an internal project (I won't name those companies here, but suffice to say that their names are abbreviations and usually three characters or less in length:)).
These training classes will cost the company $1k-$2k per student, depending on the exact nature of the course. This covers a 5-day class, 7-8 hours per day, with labs as practical for each topic. Obviously, in 5 days you're not going to get a lot of depth in any single topic, but a good instructor will be able to answer off-the-cuff questions during breaks between topics. I know that in a 10-15 minute break, I often get about 3 minutes to make a run to a restroom and spend the rest of the time going over details with one or more students. The most common areas of questions include processor scheduling (big changes in this area right now), virtual memory implementation details (especially the slab allocator and zoned memory concepts), and the block layer API. We don't get heavy into implementation details, but the student is expected to have at least some C background so that they can accomplish the lab exercises.
Some customers require detailed knowledge about specific subsystems and I will add a "chalk talk" ("dry erase talk"?) as time permits to cover those areas as much as possible. For example, a company that makes video poker machines running Linux might want details on hacking the interrupt handlers, while a company that builds disk storage units might want to talk about how best to support a custom RAID controller. Those types of things come up primarily in the Linux Device Driver course; the Internals course typically comes first in the curriculum and can be applied by application and system programmers to the code they write.
Most individuals are not going to be able to afford the associated costs, however. There are some training companies that offer "public" courses: I do those classes as well as on-site classes, but public classes for Internals don't often happen because there's not enough interest to fill the classroom with warm bodies. Send me a message if you're looking for such classes and I'll give you a list of vendors. If you opt to go the less expensive route, I suggest you get Robert Love's book on the Linux Kernel; overall it's a great book, but it does lack depth in one or two areas. After you understand what he covers there, then download the free book on the Linux VMM from
Sorry, I'm rambling. I'll now return you to your regularly scheduled pr0n viewing.:)
A few years ago I read a short blurb in my IEEE Spectrum magazine about quantum communication. Apparently, a quantum particle with a particular spin (there are three pairs: top/bottom, strange/charmed, and up/down (is that last one right?)) has an exactly opposite "anti-particle" (well, DUH!;)). The interesting thing is that a change in the spin of the first particle causes an instant change in the spin of the other.
IIRC, one particle was in Chicago somewhere (University lab?) and the other was down-under. The change in spin of the Chicago particle was determined to exactly coincide with the change in the other (well, as "exactly" as the lab instruments could measure).
If further experiments panned out we would have instantaneous communication across any distance (I'll have to Google for it now and see what happened with it).
While you might be "renting" the Windows operating system, due to its licensing scheme, I essentially own SUSE Linux, since there are no restrictions on how I can use it, transfer it to another, and so on. Of course, those privileges are assigned to me by the license and require that I obey certain aspects of the license, but overall I can continue to use the operating system indefinitely and without fear of the license being revoked.
I think the big issue with Novell/MS getting in bed together concerns updates to SUSE. It would be very easy for Novell to slide something into an update that I wouldn't necessarily notice. Hence my statement regarding being extra careful when updating.
To continue the analogy one step further: Your mileage may vary.:)
Just because Ford or Chevy moves in the wrong direction with new products, doesn't mean I should buy a new car. The same applies to operating systems.
Granted, I will be on the lookout for anything unusual going on with SUSE, and I will certainly be watching to make sure that I don't install something from a SUSE repository that I shouldn't, but I think jumping ship is a bit premature at this point.
Implementing all of these would be far cheaper than having people boil their water,
Not only that, but they might actually work.:) After all, boiling water isn't going to remove any of the heavy metals found in contaminated water. In fact, as the water boils off, the contaminates become more concentrated per measure of water.
Boiling water kills (most/some) germs and bacteria, but that's all.
But does he have an obligation to inform the FBI (I'm assuming from the OP's tone that this happened in the US)? After all, a federal crime was committed by breaching the computers and using them for unauthorized purposes (see the Wire Fraud Act).
I agree with all posters who say that he should CYA and document everything. But is he liable for not reporting that a federal law has been broken?
Another option is to run Windows exclusively inside a virtual machine session and prevent any persistence of changes to the virtual drive(s) in that session. If you are hacked, reboot the virtual machine and everything is back the way it was.
This produces the problem of where to store data that should be persistent. The obviously place is on a "network" drive that connects to the host's filesystem. If the host is non-Windows, this has a good chance of working reasonably well, except that the virtual machine can still get to it. So after saving the data, it would need to be moved outside of the hierarchy seen by the VM.
But how can you guarantee that it can be sanitized before moving it back? After all, you'll need that data at some point, or you wouldn't have saved it.
Unfortunately, our elections don't let us choose "the good one", just "the better" (or "not the worse").
Hmm. There are people I know who I wish didn't have the right to vote. Or at least, there are people who I wish did not exercise their right to vote.
While it might seem crass or crude to say such things, the reality is that many people don't understand the political system in which they live. Do you want such people voting for president? I certainly don't.
The original list of people who could vote were narrowed down to those who fit the following criteria: (1) male, (2) a land owner, and (3) between of the ages of 18 and 44 (do I have the age range correct?). In fact, it wasn't viewed as a "right" so much as a "responsibility". A responsibility to your family, a responsibility to your neighbor, and a responsibility to your country! Those who were eligible to vote and didn't -- and didn't have a good reason -- were often ostracized from the community, sometimes for a long period of time.
(I am not a historian so I could very likely have the above synopsis all wrong. If I do, please correct me! I like to be as accurate as possible and I would appreciate learning from you if you know better than I!)
Until we can vote in a way where everyone's votes count, not just the winners, we don't really have democracy.
But then, we don't have a democracy in America, we have a republic.:)
VMware wants the Linux kernel to support a closed-source interface with an API that is not subject to change.
As Greg KH pointed out in his keynote at the recent 2006 Open Linux Symposium (mid-July), closed-source kernel modules are illegal and the kernel developers will do everything they can to prevent them. Besides the fact that they're illegal (in the view of many of the top kernel developers who would be responsible for approving the patches), they also consider them unethical. So even if they were somehow persuaded to include such an API from the legal standpoint, they find it morally reprehensible and would likely refuse on those grounds.
The folks at Xen aren't much better. They think an open source implementation is good, but as you point out, they want it tailored just for them. Linux has always been about choice, so I doubt that viewpoint will prevail either.
Instead, there needs to be a middle ground between these two extremes, or the Linux kernel will never get a hypervisor.
Kernel drivers can not use virtual memory, explicit multithreading or advanced algorithms from MP, STL and hundreds of other C++ libraries.
Wow, I don't know where to start with this. First, mod parent down using -1, vague.
Have you written a filesystem? I have. The "virtual memory" you're apparently referring to is the process' virtual memory. This sounds like what you're trying to say is that the protection provided by process virtual memory isn't available in the kernel. And that's true. But that doesn't change the fact that the kernel can map any physical memory to any effective address. So the statement that the "kernel can not use virtual memory," is extremely bogus. ("Bogus" is a technical term. It means "completely wrong.")
Concerning "explicit multithreading," you must be referring to the idea that the kernel can't call pthread_create()? But there's no reason for the kernel to do so. Multi-threading in a user-level process is often done to achieve concurrency related to the delays that a single-threaded application would have if it had to wait for I/O to complete. The kernel doesn't have to do that at all. The kernel would queue up the I/O request, then continue on its merry way. When the interrupt from the device signals that the I/O has completed, a separate handler takes care of it. In a user process, threads are useful in order to modularize the I/O functions. In the kernel, they often aren't needed, since callbacks are used instead. Same functionality, different design technique. And even if they are needed for some obscure reason, all modern kernels (Linux included) support kernel threads. (My SUSE Linux 10 box currently has 19 kernel threads executing.)
The "advanced algorithms" you're referring to are probably coming out of user-space libraries. And in this regard, you're correct -- user-space libraries cannot (currently) be linked into the kernel and there is plenty of debate about whether such abilities should even be attempted. (The problem with user-space libraries inside kernel space revolves mostly around bugs and implementation deficiencies. The truth is that an algorithm that is mostly cpu-intensive probably could be loaded into kernel space using some kind of hack, and there are open source projects that are already working along these lines.)
In any case, there's no reason why those algorithms couldn't be executed inside the kernel. For example, take the find() generic algorithm from the STL (a macro from one of the libraries you mentioned). Why can't I use it? (The truth is I can.) And why can't I use the list class from the same library? I admit that linking large objects into the kernel could result in quite a bit of bloat, but there is not a technical reason that it couldn't be done. (Except in the case of C++ exceptions within the kernel. There is a group that has patches available for the Linux kernel that add support for C++ exception handling. With those patches, any STL code should be able to work in kernel space, although I've not tried it personally.)
It seems to me like the parent has read a magazine article and jumped to conclusions. Or perhaps they are even an experienced developer, but took huge liberties with the wording of their statement. But as "Captain Obvious," I felt it was my slashdot civic duty to clarify he issue.:)
Here's the text describing what the COA is (in Microsoft's mind, anyway):
A Certificate of Authenticity (COA) is a label that helps you identify genuine Microsoft software. A COA is not a software license it is a visual identifier that assists in determining whether or not the Microsoft software you are running is genuine. However, without it, you will not have a legal license to run Microsoft software. A COA should never be purchased by itself without the software it authenticates.
You're right, of course. No one has to buy music from the big record labels.
Think of the music as the tea that was imported by the East India Tea Company and delivered to the colonies via the port in Boston. And think of DRM as the tax that was assessed when the colonists bought that same tea.
And now consider what happened during the Boston Tea Party.:)
Of course, the Boston Tea Party was a bunch of hoodlums who vandalized the private property of overseas company and they should rightly have been thrown into jail (which is why they masqueraded as native Indians: to try to cover it up, although the officials in Boston knew perfectly well who they were).
More tan anything else, the Boston Tea Party became a flashpoint in a larger fight against the tyranny of British rule over the American colonies. "The straw that broke the camel's back," for lack of a better cliche.
Actually, I would use my SUSE-based VMware session, most likely, with the persistent storage option turned off (shutdown VMware and the drive reverts back to its previous state).
It does make me wonder, though. Even VMware would require some tweaking to avoid contamination outside the virtual box (I have shared folders enabled on my VMware sessions, along with host networking and bridged networking)...
Hi, charleste. This is completely OT, but I figured I'd say "hi". I live in Tampa, but work elsewhere around the country/world (where I'm at is dependent on what week it is!).
"Hi".:)
Here's the question you have to ask yourself, though... will your friends and relatives who don't use OSS and who have crashes & viruses actually do better with OSS and a fresh install of Linux? Or would their problems be fixed with a fresh install of Windows, a good firewall, and the abolition of Internet Explorer?
And this really is the question, isn't it? But the answer isn't obvious.
First, let me explain my environment in my home office: I have a Linux/Unix-only network that I call my "corporate net". It consists of AIX, HP-UX, and a half-dozen Linux machines. The mantra in my corporate net is "no Windows allowed". One of my Linux boxes is used for my accounting system: a spreadsheet with macros, some word processing, Evolution email client, etc. All of it F/OSS.
My wife is the one that does our accounting, though. So she has to leave her Windows world (her office in our house) and come into my office to do accounting work, because I won't open an entry path in the firewall from her Windows machine. Even if I trust my firewall to only allow traffic from her box, I can't trust that her machine hasn't been compromised. And I have educated her about viruses and such, and she keeps her subscription-based scanners and such up to date. And she now uses Firefox (after 12 months of my urging!) so IE is not a direct problem.
I say "direct problem", because she still uses Quicken for our personal accounting (sigh) and Quicken uses the IE ActiveX controls for HTML rendering to display help information and other such things. So her machine won't be safe to my corporate net unless I actually delete/disable those DLLs. And she's not willing to give me that level of control over her computer.
I expect to have time over the summer to install VMware on her old Dell laptop. She plans to install Linux on it first, then I'll put on VMware, and then she'll run Quicken inside the VMware session. The VMware session will be read-only, with the Quicken data being store (in the Linux filesystem). Every time she boots VMware, she'll get the exact same image to run. Even if it does become infected, rebooting VMware will clear the infection.
In summary, educating users about viruses and how to avoid them is a great first step. But users can visit sites and become infected anyway. There are plenty of vulnerabilities that a firewall can't/won't protect against, and the user can't prevent. The only solution (that I can see) that maintains Windows compatibility is a fresh install of the OS that is read-only.
They know these security programs do nothing more than waste money on pork and make certain politicians feel smug, earning brownie points with their constituents.
This is right on.
The next time you visit an airport, ask yourself what would happen if a terrorist didn't wait until they got all the way to the metal detectors and X-ray machines before detonating an explosive device. As a business traveler, I've logged a million miles on one airline and hundreds of thousands on other airlines. Any idiot who wanted to wreak havoc could tell that the place to detonate such a device would be while standing in front of the security machines: you'd take out a bunch of people and render that section of the airport unusable.
So what do you do to fix it? Obvious: move the screening center further up. (And before you hit Reply, I know that wouldn't work. I'm j/k.)
I'm not posting as an AC. Maybe that's risky -- we all know that the U.S. government snoops on its citizens -- and I really can't afford to defend myself against the likes of Gonzales. But I can trust in my country's Constitution and hope that those in power won't abuse it. (That's naive. But maybe with enough press coverage I wouldn't get shafted too bad.) But without open discussion that provokes thoughtful responses, how will anything get any better?
Wrong acronym for the organization. Instead, it's SIIA (Software Information & Industry Association), but you're right -- Mark Webbink from RedHat is the chairman of the board! Time for me to write some letters and make a phone call or two...
Part of the problem is when a n00b goes to a developer forum and asks questions that are not development questions.
This happened to me. I was interested in hacking my TiVo (standalone 540-series) which is pretty much unhackable without a PROM modification. But I had managed to look through some of the files that I'd copied off the TiVo hard drive and I had some questions about one of the file formats. I looked through the SeriesII forum as well as a lot of searching and didn't find the answers to my questions, so I posted. Big mistake.:(
Please understand, I'm not new to Linux. I've been teaching a Linux Internals course for a few years now, as well as device driver development, kernel debugging, and so forth. I know a little bit about the Linux kernel. What I didn't know was the boot process used on the TiVo. I was slammed by the forum moderator ("RTFM in the Newbies section"). I thought I had done that, after already spending hours+hours of searching and reading posts that were too old to apply to my unit.
Until... I went back to the newbie forum and read every one of the sticky articles and I found 80% of the information I was looking for (sigh). Part of the problem was that the newbie forum has about 15 sticky articles, some of them with 40 pages of posts in a single thread. Experience had taught me that most of those are too old to apply to my unit, so I would open the thread and jump to the last page, working backwards through the thread, to see if I could find something relevant. Well, the information was actually in the first few posts -- the community had been editing that post and updating it as time went by.
My point to all this is that newbies often post in an area that is for developers; they should start with places like linuxquestions.org or other generic Q&A sites, then progress from there. The people that frequent those sites want to help others, not write code. They're the newbie's best chance of having their questions answered.
It would appear that the best way for the Linux/OpenSource community to handle this is to ignore it for now. If nVidia/ATI/whoever feels that they can afford to lose the marketshare that the community represents, we're not big enough yet.:)
When we represent the largest market share, then we dictate the terms: you've been closed-source long enough. Without at least open documentation, the kernel will no longer allow binary-only drivers to be loaded. (Use a technique similar to how MODULE_LICENSE() is used right now.)
Slashdot Mirror
Oh, and BMW drivers are invariably prats who tailgate you for as long as possible, then overtake and cut you up.
Heh-heh. I drive a BMW now and I do drive aggressively. Of course, I drove the same way in my twin-turbo Supra, my 5.0L Mustang, and every other car I've owned.
Sorry, please don't even both spending mod points to bury this. I just thought it amusing that someone would paint with such a broad brush across a fairly large segment of adults. :P
I've been teaching Linux Kernel Internals and Linux Device Driver courses (among others) since 1995 and other Un*x topics prior to that.
Many of the big names "outsource" their very technical training to third parties unless the topic is something related to an internal project (I won't name those companies here, but suffice to say that their names are abbreviations and usually three characters or less in length :)).
These training classes will cost the company $1k-$2k per student, depending on the exact nature of the course. This covers a 5-day class, 7-8 hours per day, with labs as practical for each topic. Obviously, in 5 days you're not going to get a lot of depth in any single topic, but a good instructor will be able to answer off-the-cuff questions during breaks between topics. I know that in a 10-15 minute break, I often get about 3 minutes to make a run to a restroom and spend the rest of the time going over details with one or more students. The most common areas of questions include processor scheduling (big changes in this area right now), virtual memory implementation details (especially the slab allocator and zoned memory concepts), and the block layer API. We don't get heavy into implementation details, but the student is expected to have at least some C background so that they can accomplish the lab exercises.
Some customers require detailed knowledge about specific subsystems and I will add a "chalk talk" ("dry erase talk"?) as time permits to cover those areas as much as possible. For example, a company that makes video poker machines running Linux might want details on hacking the interrupt handlers, while a company that builds disk storage units might want to talk about how best to support a custom RAID controller. Those types of things come up primarily in the Linux Device Driver course; the Internals course typically comes first in the curriculum and can be applied by application and system programmers to the code they write.
Most individuals are not going to be able to afford the associated costs, however. There are some training companies that offer "public" courses: I do those classes as well as on-site classes, but public classes for Internals don't often happen because there's not enough interest to fill the classroom with warm bodies. Send me a message if you're looking for such classes and I'll give you a list of vendors. If you opt to go the less expensive route, I suggest you get Robert Love's book on the Linux Kernel; overall it's a great book, but it does lack depth in one or two areas. After you understand what he covers there, then download the free book on the Linux VMM from
Sorry, I'm rambling. I'll now return you to your regularly scheduled pr0n viewing. :)
A few years ago I read a short blurb in my IEEE Spectrum magazine about quantum communication. Apparently, a quantum particle with a particular spin (there are three pairs: top/bottom, strange/charmed, and up/down (is that last one right?)) has an exactly opposite "anti-particle" (well, DUH! ;)). The interesting thing is that a change in the spin of the first particle causes an instant change in the spin of the other.
IIRC, one particle was in Chicago somewhere (University lab?) and the other was down-under. The change in spin of the Chicago particle was determined to exactly coincide with the change in the other (well, as "exactly" as the lab instruments could measure).
If further experiments panned out we would have instantaneous communication across any distance (I'll have to Google for it now and see what happened with it).
Here's the text of the bill as it currently stands (April 17th, ~2:40PM EDT): http://www.flsenate.gov/cgi-bin/view_page.pl?Tab=s ession&Submenu=1&FT=D&File=SB1974c1491738.html&Dir ectory=session/2007/Senate/bills/amendments_com/ht ml/
Since there may be a session ID in there somewhere, try visiting this page http://www.flsenate.gov/session/index.cfm?BI_Mode= ViewBillInfo&Mode=Bills&SubMenu=1&Year=2007&billnu m=1974 and scrolling down to the section labeled Proposed Committee Substitutes and click on the latest Web Page version. Then search for the word "open" within that page. :)
It looks to me like this is much ado about nothing at this point...
It doesn't matter.
Regardless of whether the Sun's effect on the Earth is changing are temperatures abnormally, WE are having an effect!
While you might be "renting" the Windows operating system, due to its licensing scheme, I essentially own SUSE Linux, since there are no restrictions on how I can use it, transfer it to another, and so on. Of course, those privileges are assigned to me by the license and require that I obey certain aspects of the license, but overall I can continue to use the operating system indefinitely and without fear of the license being revoked.
I think the big issue with Novell/MS getting in bed together concerns updates to SUSE. It would be very easy for Novell to slide something into an update that I wouldn't necessarily notice. Hence my statement regarding being extra careful when updating.
To continue the analogy one step further: Your mileage may vary. :)
I don't agree.
Just because Ford or Chevy moves in the wrong direction with new products, doesn't mean I should buy a new car. The same applies to operating systems.
Granted, I will be on the lookout for anything unusual going on with SUSE, and I will certainly be watching to make sure that I don't install something from a SUSE repository that I shouldn't, but I think jumping ship is a bit premature at this point.
Not only that, but they might actually work. :) After all, boiling water isn't going to remove any of the heavy metals found in contaminated water. In fact, as the water boils off, the contaminates become more concentrated per measure of water.
Boiling water kills (most/some) germs and bacteria, but that's all.
But does he have an obligation to inform the FBI (I'm assuming from the OP's tone that this happened in the US)? After all, a federal crime was committed by breaching the computers and using them for unauthorized purposes (see the Wire Fraud Act).
I agree with all posters who say that he should CYA and document everything. But is he liable for not reporting that a federal law has been broken?
Another option is to run Windows exclusively inside a virtual machine session and prevent any persistence of changes to the virtual drive(s) in that session. If you are hacked, reboot the virtual machine and everything is back the way it was. This produces the problem of where to store data that should be persistent. The obviously place is on a "network" drive that connects to the host's filesystem. If the host is non-Windows, this has a good chance of working reasonably well, except that the virtual machine can still get to it. So after saving the data, it would need to be moved outside of the hierarchy seen by the VM. But how can you guarantee that it can be sanitized before moving it back? After all, you'll need that data at some point, or you wouldn't have saved it.
Hmm. There are people I know who I wish didn't have the right to vote. Or at least, there are people who I wish did not exercise their right to vote.
While it might seem crass or crude to say such things, the reality is that many people don't understand the political system in which they live. Do you want such people voting for president? I certainly don't.
The original list of people who could vote were narrowed down to those who fit the following criteria: (1) male, (2) a land owner, and (3) between of the ages of 18 and 44 (do I have the age range correct?). In fact, it wasn't viewed as a "right" so much as a "responsibility". A responsibility to your family, a responsibility to your neighbor, and a responsibility to your country! Those who were eligible to vote and didn't -- and didn't have a good reason -- were often ostracized from the community, sometimes for a long period of time.
(I am not a historian so I could very likely have the above synopsis all wrong. If I do, please correct me! I like to be as accurate as possible and I would appreciate learning from you if you know better than I!)
But then, we don't have a democracy in America, we have a republic. :)
VMware wants the Linux kernel to support a closed-source interface with an API that is not subject to change.
As Greg KH pointed out in his keynote at the recent 2006 Open Linux Symposium (mid-July), closed-source kernel modules are illegal and the kernel developers will do everything they can to prevent them. Besides the fact that they're illegal (in the view of many of the top kernel developers who would be responsible for approving the patches), they also consider them unethical. So even if they were somehow persuaded to include such an API from the legal standpoint, they find it morally reprehensible and would likely refuse on those grounds.
The folks at Xen aren't much better. They think an open source implementation is good, but as you point out, they want it tailored just for them. Linux has always been about choice, so I doubt that viewpoint will prevail either.
Instead, there needs to be a middle ground between these two extremes, or the Linux kernel will never get a hypervisor.
The proposal as quoted wouldn't apply to AMERICAN CITIZENS who decide to commit a terrorist act, would it? Is that a good thing or a bad thing?
Oh, I thought that was obvious: extrempletely wrong. :)
Wow, I don't know where to start with this. First, mod parent down using -1, vague.
Have you written a filesystem? I have. The "virtual memory" you're apparently referring to is the process' virtual memory. This sounds like what you're trying to say is that the protection provided by process virtual memory isn't available in the kernel. And that's true. But that doesn't change the fact that the kernel can map any physical memory to any effective address. So the statement that the "kernel can not use virtual memory," is extremely bogus. ("Bogus" is a technical term. It means "completely wrong.")
Concerning "explicit multithreading," you must be referring to the idea that the kernel can't call pthread_create()? But there's no reason for the kernel to do so. Multi-threading in a user-level process is often done to achieve concurrency related to the delays that a single-threaded application would have if it had to wait for I/O to complete. The kernel doesn't have to do that at all. The kernel would queue up the I/O request, then continue on its merry way. When the interrupt from the device signals that the I/O has completed, a separate handler takes care of it. In a user process, threads are useful in order to modularize the I/O functions. In the kernel, they often aren't needed, since callbacks are used instead. Same functionality, different design technique. And even if they are needed for some obscure reason, all modern kernels (Linux included) support kernel threads. (My SUSE Linux 10 box currently has 19 kernel threads executing.)
The "advanced algorithms" you're referring to are probably coming out of user-space libraries. And in this regard, you're correct -- user-space libraries cannot (currently) be linked into the kernel and there is plenty of debate about whether such abilities should even be attempted. (The problem with user-space libraries inside kernel space revolves mostly around bugs and implementation deficiencies. The truth is that an algorithm that is mostly cpu-intensive probably could be loaded into kernel space using some kind of hack, and there are open source projects that are already working along these lines.)
In any case, there's no reason why those algorithms couldn't be executed inside the kernel. For example, take the find() generic algorithm from the STL (a macro from one of the libraries you mentioned). Why can't I use it? (The truth is I can.) And why can't I use the list class from the same library? I admit that linking large objects into the kernel could result in quite a bit of bloat, but there is not a technical reason that it couldn't be done. (Except in the case of C++ exceptions within the kernel. There is a group that has patches available for the Linux kernel that add support for C++ exception handling. With those patches, any STL code should be able to work in kernel space, although I've not tried it personally.)
It seems to me like the parent has read a magazine article and jumped to conclusions. Or perhaps they are even an experienced developer, but took huge liberties with the wording of their statement. But as "Captain Obvious," I felt it was my slashdot civic duty to clarify he issue. :)
The above quote is from http://www.microsoft.com/resources/howtotell/en/co a.mspx
Think of the music as the tea that was imported by the East India Tea Company and delivered to the colonies via the port in Boston. And think of DRM as the tax that was assessed when the colonists bought that same tea.
And now consider what happened during the Boston Tea Party. :)
Of course, the Boston Tea Party was a bunch of hoodlums who vandalized the private property of overseas company and they should rightly have been thrown into jail (which is why they masqueraded as native Indians: to try to cover it up, although the officials in Boston knew perfectly well who they were).
More tan anything else, the Boston Tea Party became a flashpoint in a larger fight against the tyranny of British rule over the American colonies. "The straw that broke the camel's back," for lack of a better cliche.
Actually, I would use my SUSE-based VMware session, most likely, with the persistent storage option turned off (shutdown VMware and the drive reverts back to its previous state).
It does make me wonder, though. Even VMware would require some tweaking to avoid contamination outside the virtual box (I have shared folders enabled on my VMware sessions, along with host networking and bridged networking)...
Hi, charleste. This is completely OT, but I figured I'd say "hi". I live in Tampa, but work elsewhere around the country/world (where I'm at is dependent on what week it is!). "Hi". :)
And this really is the question, isn't it? But the answer isn't obvious.
First, let me explain my environment in my home office: I have a Linux/Unix-only network that I call my "corporate net". It consists of AIX, HP-UX, and a half-dozen Linux machines. The mantra in my corporate net is "no Windows allowed". One of my Linux boxes is used for my accounting system: a spreadsheet with macros, some word processing, Evolution email client, etc. All of it F/OSS.
My wife is the one that does our accounting, though. So she has to leave her Windows world (her office in our house) and come into my office to do accounting work, because I won't open an entry path in the firewall from her Windows machine. Even if I trust my firewall to only allow traffic from her box, I can't trust that her machine hasn't been compromised. And I have educated her about viruses and such, and she keeps her subscription-based scanners and such up to date. And she now uses Firefox (after 12 months of my urging!) so IE is not a direct problem.
I say "direct problem", because she still uses Quicken for our personal accounting (sigh) and Quicken uses the IE ActiveX controls for HTML rendering to display help information and other such things. So her machine won't be safe to my corporate net unless I actually delete/disable those DLLs. And she's not willing to give me that level of control over her computer.
I expect to have time over the summer to install VMware on her old Dell laptop. She plans to install Linux on it first, then I'll put on VMware, and then she'll run Quicken inside the VMware session. The VMware session will be read-only, with the Quicken data being store (in the Linux filesystem). Every time she boots VMware, she'll get the exact same image to run. Even if it does become infected, rebooting VMware will clear the infection.
In summary, educating users about viruses and how to avoid them is a great first step. But users can visit sites and become infected anyway. There are plenty of vulnerabilities that a firewall can't/won't protect against, and the user can't prevent. The only solution (that I can see) that maintains Windows compatibility is a fresh install of the OS that is read-only.
This is right on.
The next time you visit an airport, ask yourself what would happen if a terrorist didn't wait until they got all the way to the metal detectors and X-ray machines before detonating an explosive device. As a business traveler, I've logged a million miles on one airline and hundreds of thousands on other airlines. Any idiot who wanted to wreak havoc could tell that the place to detonate such a device would be while standing in front of the security machines: you'd take out a bunch of people and render that section of the airport unusable.
So what do you do to fix it? Obvious: move the screening center further up. (And before you hit Reply, I know that wouldn't work. I'm j/k.)
I'm not posting as an AC. Maybe that's risky -- we all know that the U.S. government snoops on its citizens -- and I really can't afford to defend myself against the likes of Gonzales. But I can trust in my country's Constitution and hope that those in power won't abuse it. (That's naive. But maybe with enough press coverage I wouldn't get shafted too bad.) But without open discussion that provokes thoughtful responses, how will anything get any better?
Wrong acronym for the organization. Instead, it's SIIA (Software Information & Industry Association), but you're right -- Mark Webbink from RedHat is the chairman of the board! Time for me to write some letters and make a phone call or two...
http://www.siia.net/membership/board.asp
This happened to me. I was interested in hacking my TiVo (standalone 540-series) which is pretty much unhackable without a PROM modification. But I had managed to look through some of the files that I'd copied off the TiVo hard drive and I had some questions about one of the file formats. I looked through the SeriesII forum as well as a lot of searching and didn't find the answers to my questions, so I posted. Big mistake. :(
Please understand, I'm not new to Linux. I've been teaching a Linux Internals course for a few years now, as well as device driver development, kernel debugging, and so forth. I know a little bit about the Linux kernel. What I didn't know was the boot process used on the TiVo. I was slammed by the forum moderator ("RTFM in the Newbies section"). I thought I had done that, after already spending hours+hours of searching and reading posts that were too old to apply to my unit.
Until... I went back to the newbie forum and read every one of the sticky articles and I found 80% of the information I was looking for (sigh). Part of the problem was that the newbie forum has about 15 sticky articles, some of them with 40 pages of posts in a single thread. Experience had taught me that most of those are too old to apply to my unit, so I would open the thread and jump to the last page, working backwards through the thread, to see if I could find something relevant. Well, the information was actually in the first few posts -- the community had been editing that post and updating it as time went by.
My point to all this is that newbies often post in an area that is for developers; they should start with places like linuxquestions.org or other generic Q&A sites, then progress from there. The people that frequent those sites want to help others, not write code. They're the newbie's best chance of having their questions answered.
When we represent the largest market share, then we dictate the terms: you've been closed-source long enough. Without at least open documentation, the kernel will no longer allow binary-only drivers to be loaded. (Use a technique similar to how MODULE_LICENSE() is used right now.)