I would say the installation method on linux is more robust than any other method on any other platform.
I think what you meant to say was "easy installation method." I consider the package management system quite easy. Tell me, what exactly do you do when [your favorite software] doesn't provide a dmg that is available for download. What's that? You need to compile it yourself? For shame, how could Apple make such an unusable operating sytem.
Well, I don't think the OP was suggesting replacing an established Windows infrastructure with linux, but, yes, you can purchase phone support for your nfs server, from a number of vendors.
What about all the various backup products, such as tape backups and seamless server redundancy? Are there alternatives for this for Linux?
Yes. And I would venture to suggest that linux probably has much better support for remote backups and failover clusters than Windows.
What assurances does a large company who absolutely can not afford significant downtime have that the software is well supported by professionals on call and that bugs are constantly being fixed?
This is Red Hat's entire business. If you need that kind of support, they would be more than happy to oblige. A number of other vendors also can provide that level of support.
There are thousands of tools that are necessary for full production environments
Yes, you are correct. And those thousands of tools are all available for linux as well...with the added bonus that they will play nicely with your Windows clients. Novell puts together a distribution that provides all of this out of the box. The only things that is really lacking is an Exchange replacement, but I see that coming in a couple of years from the various Mozilla projects. Thing is, Exchange is slowly dating itself. There are a few web services out there already, like Google Apps, that let you easily integrate email and shared calendars...and you don't need an expensive and massively proprietary application (along with the expensive and massive hardware to run it on). As these mature, Exchange is going to have to evolve or die out.
if you go to a business and tell them that they will likely say, "and what happens if it goes down?"
Hand them the business card of your local Red Hat or IBM sales representative. This is why these companies are in business, and they know how to talk to and reassure PHBs.
While legal, it would probably be exceedingly impractical to actually get PostgreSQL 8.2 running to take advantage of the latest features and fixes on Redhat 4. (from 1996, kernel 2.0.18 according to wikipedia)
Actually, it probably wouldn't be that hard. It would be a bit crazy, but PostgreSQL just needs something to compile it, and the required libraries. A better example would have been Gimp 2.4 (or some other desktop app), because that relies on libraries that require more recent kernel versions.
If you want something that supports current, bleeding edge hardware and software, Linux is the only way to go.
I sympathize with your POV, but this simply isn't true. If the hardware is closed (still lots of wireless, some raid, some nic, some ACPI, some weird usb devices like webcams and such, graphics hardware), you're going to have a hell of a time getting linux to recognize it. It is certainly true, though, that linux supports various architectures (like x86, ppc, x86-64, ia64, etc) very well and much better than Windows.
*But* being published in peer reviewed journals is still perceived as being a solid indicator of one's academic status and career progression.
And not for a bad reason. Economics is driven by productivity. A carpenter or a plumber can perform a service for somebody. A physician can treat somebody. A computer programmer can write software needed to help you run your business.
A scientist does research, but what's the difference between him and the above? The above are fairly tangible and their contribution to society easily measured. A scientists contributions are not. But people try, and the most popular method is via publications. Still, the benefit to society of basic research is a long-term affair and isn't usually realized right away. So how do you determine how much of your resources to allocate to it?
I think publications, to an extent, are a measure of productivity. If a scientist can get a research project off the ground, get relevant data, analyze it, answer an important scientific question, and put it all together into a nice story, that is productivity. If he can't, for whatever reason, it doesn't mean he is unproductive, but it is a lot harder to measure his contribution. And just like the computer programmer who is fresh out of school with no real experience to demonstrate, an employer is more likely to go with somebody who has proven himself than with somebody who hasn't.
Some of what you say is true, but I wouldn't be quite so cynical. A "useful or interesting work" can have a lot of different levels. Most journal articles represent a work in progress, not a complete understanding of a topic, so there are always more experiments to do. You just have to publish what you have at some point. Now some experiments are done sloppily and not caught by peer-reviewers, but not everybody is an expert in every field. I try to know what I'm talking about when I publish something, but if it's not directly in my expertise, a real expert can almost certainly find something wrong with it. But the article doesn't have to be perfect to get something out of it. Readers will be critical--that's their job after all if they're good scientists--and not necessarily agree with the conclusions of the authors, but they can usually conclude something from the data that is presented, and I consider that a helpful contribution to science.
I also don't agree that the vast majority are reviews, at least not in experimental science. A simple search on PubMed will tell you which articles are considered reviews and which present original research. Unless you do a very general search, articles typically outnumber reviews 10 to 1.
As in every politically driven community, there is corruption in academia. Big name scientific egos can have an undue influence over a field, but that doesn't happen very often. Most scientists are fairly collegial toward each other and respect each others work, even if they don't agree with it.
But, you are right, publish or perish is a problem, and it leads to things like blocking of competitors publication, bad grant reviews, and fabrication of data. I have, unfortunately, seen some great scientists get driven away by some sort of bs, like an insufficient number of publications. But I've also seen a lot of great scientists succeed in the system. Unfortunately, given the current economics of research, it's not a system that is going to change easily. And academia is the only place with problems. Cutthroat environments in industry can create the same sort of problems.
Peer review does effectively happen online. After an article is submitted to a journal and vetted by the editors, it is sent, usually electronically, to selected reviewers. Reviewers then submit their critique electronically. There isn't a lot of mailing of manuscripts. That, like you say, is fairly pointless in an electronic age. Critique in a forum doesn't happen, but that would be fairly impractical for a scientific article. Besides, there isn't any direct communication between reviewers and submitters. It is blind, and there isn't a lot of traffic in general--just the manuscript to be sent and the review to be received.
I do think there is an important role for journals...it allows scientific themes and significant advances to be followed more easily. Somebody else (the editors) has screened a lot of submissions--looking for things like relevance to the journal, significance of data, a well-told story, etc--before it ever makes it to print, so the reader doesn't have to wade through a ton of crap to get to the interesting article he is looking for. The economics of journals will probably certainly change, but journals themselves will remain for the near future. And nothing stops a PI from publishing their findings online if it doesn't make it into a journal. It's just that fewer people are likely to see it that way.
I find that really hard to believe. From the same site,
Median annual earnings of wage-and-salary network and computer systems administrators were $62,130 in May 2006.
If you look at the breakdown, even public schools pay more than what you say your salary is.
At any rate, that's not the point. The point is that a lot of people in the tech sector think they have an entitlement to a cushy job that makes >80k/yr with great benefits, vacation time, and 40 hour work weeks. Well, you don't. In the real world you are just another trained professional offering a service, and you will probably only ever make an average wage with average benefits--which isn't necessarily very much. If you were lobbying for better pay and working conditions in general for all trained professionals, I would sympathize. But people in the CS field don't deserve special treatment "just 'cause they know them computrons."
They can't get americans to buy their crappy pay, benefits, and job security, so they want to farm out slave labor they can have deported at their whim.
Oh cry me a river!
From the US Dept. of Labor: In May 2006, median annual earnings of wage-and-salary computer applications software engineers were $79,780.
In May 2006, median annual earnings of wage-and-salary computer systems software engineers were $85,370.
How can you possibly suggest that a salary like that qualifies as "slave labor?" That's well above the median income of $46,326 in the U.S. per the U.S. Census. Are you aware that there are real cases of slave labor in the U.S? Such as those where a person has to work 2-3 jobs, gets no benefits, vacation, or job security, and still makes less than the poverty line?
The whiney upper-middle classers need to wake up and stop crying about their employers. As long as corporate abuse doesn't happen to them, they are ok with it. Well, guess what? India, China, and Japan are training top notch computer scientists and they are willing to work for less than Americans. That's called competition, and since that is what our capitalist economy thrives on (or so the Republicans/Libertarians keep saying), deal with it.
Wait...so you're equating blindly clicking "Yes" on an annoying dialog that pops up every 2 minutes to actively entering your password when a program wants to make a major change to your machine?
That issue aside, AppArmor and SELinux are doing quite a bit more than simple privilege separation for linux right now. Not only are they finding bugs in programs to help eliminate security issues, but they are restricting programs to exactly what they need and nothing more. I think that's going to do a lot more for linux security than UAC does for Windows.
The OEM system install has been the gold standard for the non-technical end user for damn near thirty years. You unpack the box. You connect the cables and you are good to go.
Isn't that exactly what I just said? Windows comes preinstalled and configured, so it is what people use and what major companies write software for. Actual usability has very little to do with it.
People don't want to be dependent on a geek who may not be around when they need help. What they want is the toll free number, the extended warranty, the in-home service contract.
Agreed. But you're make a lot of assumptions here. First of all, service contracts are available for Linux. Second, what Windows users do you know who prefer telephone support to asking somebody they know for help? Computer users like quick fixes to their problems...they don't really care how they get it. If they can just buy somebody coffee and have them come over and fix their problem, they usually much prefer that to dealing with tech support on the phone.
Is this meant to be a joke? I just set up a Heron Compiz install, and I knew-of/had no choice but to go the the terminal at least 10 times over the course of setting it up.
Not sure what you were doing then. Just go to System|Preferences|Appearance|Visual Effects and click either "Normal" or "Fancy." All done!
For the modem, it depends on the hardware support, but did you try System|Administration|Network to configure the PPP interface?
Seriously, the tools are right there if you just look for them. I agree the help documentation can be better, though. The online forums, while helpful for advanced users, don't always give you the easiest or distro-supported ways of doing things.
It is sometimes required for some operations, usually fixing things or setting a couple things up. Such as? Seriously, people keep saying this, but I can't think of any normal operation where the CLI is required. There are many cases where I prefer to use the CLI, though, because it is more efficient and logical. I just finished installing Ubuntu 8.04 on a friend's laptop with all of the basic software, codecs, wireless support, fancy graphics, etc... and didn't have to touch the CLI at all for any of that.
This line - or something very much like it - is woven into every Linux "conversion" story posted on Slashdot.
Meanwhile, a billion users worldwide somehow manage to run Windows without the free technical support of a resident geek.
O rly? I think you are looking through rose-tinted glasses...perhaps not considering yourself/friends "resident geeks" who can help out with computer problems. I know plenty of Windows users, and without "resident geeks" they would be completely helpless if something happened to their computer. The point is, Windows comes preinstalled on every computer and it is something people use as a result. They aren't using linux because they aren't the type to go and try to install it. Even MacOS X wouldn't be used by people if they had to install it. So if somebody installs it and sets it up for them to use, they can happily use it without problems (specific software needs aside, of course).
Before anyone says "but ubuntu already does this", the problem is that currently you have no choice but to learn the CLI in order to accomplish anything but basic user tasks.
Examples please. I prefer to use the CLI for a lot of things, but the GUI tools are there: hardware diagnostics, networking setup/troubleshooting, basic server configuration, VM configuration, user/group management, installing/managing software, directory services. What are you doing that requires the CLI?
Well, it's kind of a pain to find out what the chipsets are from the Best Buy website, so I'm not going to go through them all. Here is a list of all the chipsets supported by the madwifi driver in linux. If you get a card with one of these chipsets, it should work out of the box. But, you are going to have to verify that the card you are buying has the right chipset. Sometimes this means going to the manufacterer's website because the retailer doesn't always give you that information. Like another replier already said, it looks like D-link will be your best shot. But, again, verify the chipset before you buy anything.
No, he's talking about spewing everything into c:\windows\system32 instead of having programs be self-contained. There is some argument as to whether complete self-containment is good, though. Personally, I like the way linux does it (some containment, some shared, everything tracked by the package management system).
The other thing that really annoys me is having user data written to the program directory. For crying out loud, no mainstream OS has been single user for at least 5 years. Every consumer program should be multiuser aware and put data in the right places!
Think of Service Packs as analogous to kernel patches. Those have been known to screw up a few programs, haven't they?
Nope, can't say they have. At least not official ones that come from the distribution provider. Also, service packs are ridiculous. You are practically reinstalling the entire operating system every time you install a service pack. A new kernel is typically a 10 megabyte download. If you throw a few userspace updates in there as well, you might hit 15 megabytes. How big is SP3? 300M?
I must be misunderstanding your point. Silence is silence. Whether or not you have been read Miranda rights has nothing to do with it. Do you know of an actual case where silence prior to being read Miranda rights was used against a defendant in court?
Well, having actually served on a jury (in a criminal case), I can tell you that your preconceived stereotype is as far from the truth as can be. Juries are quite well-balanced...men/women, old/young, students/professionals, etc....
So you didn't get selected for jury duty. Is it because you were too smart? Or are you too arrogant to believe you may have had a bias? Biases are inherently irrational. Even if you recognize them, you can't always prevent them from influencing your decisions. That's why the jury selection process is the way it is.
Actually, your silence can be used against you in court.
No, it really can't, not in a criminal case. That's why the jury selection process asks you things like "Do you think the defendant is guilty because he has been charged?" and "Do you think the defendant is guilty is he refuses to testify?" These are both disqualifying questions. Juries are specifically instructed to consider the defendant innocent until proven guilty and that the burden of proof is on the prosecutor. The defendant does not have to speak one word in his defense. It is the job of the prosecutor to prove his case beyond a reasonable doubt.
So, I've actually served on a jury for a murder trial, and I can tell you it was a very educational process. For one thing, I learned that premeditation does not mean planned in the case of first degree murder. It simply means that the defendant had time to reflect. It does not mean the defendant had to have a lot of time to think, it just means there was a logical sequence of actions: reflection, decision to act, and act, all of which can happen in the course of a few seconds. Now, I agree, I don't know how the jury determined there was premeditation in this case, but the legal definition of premeditation is often quite different from what people believe.
Because juries aren't allowed to speculate. They must determine whether or not the defendant is guilty based on the evidence presented to them by the prosecutor. Vague suggestions that "somebody else may have done it" would bias the jury. I know it seems silly from the outside, but there is actually a lot of well-established case law that is used to determine what is admissible evidence in a murder trial.
I would normally agree except for the part where you lose the ability to use the software if you stop paying. So if you diligently pay $115/yr. for 4 years ($460 total), and then decide you don't want to pay anymore, you can no longer use Office (i.e: access your files). At least if I pay $400 up front for Office, I can use it for as long as I want, out of date or not.
The problem Microsoft has with Office is that they really want a subscription model, but they don't have subscription value to add. Think about Napster, for example. If you pay for a subscription to Napster ($13/month), you get unlimited access to all of the songs available, which is a significant savings over buying all of the music you may listen to only every once in a while. So the subscription model has value. With Office, Microsoft isn't adding any value. They're just trying to get more cash out of you.
That's funny because I thought Bender's Big Score was terrible. My roommate and I loved the old Futurama episodes and have watched them all several times over. About 30 minutes into the movie, we were looking at each other and asking, "Is this Futurama?" I had to force myself to sit through the whole thing, hoping that it would get better, but all I got was a seemingly endless stream of jokes and gags repeated over in a bad way from the series. And the fart joke level of humor was pretty excessive. I always liked Futurama because it didn't have a lot of that, and instead had a lot of subtle humor that could really be appreciated by geeks.
Slashdot Mirror
Until there is a bullet-proof installation method
I would say the installation method on linux is more robust than any other method on any other platform.
I think what you meant to say was "easy installation method." I consider the package management system quite easy. Tell me, what exactly do you do when [your favorite software] doesn't provide a dmg that is available for download. What's that? You need to compile it yourself? For shame, how could Apple make such an unusable operating sytem.
Well, I don't think the OP was suggesting replacing an established Windows infrastructure with linux, but, yes, you can purchase phone support for your nfs server, from a number of vendors.
What about all the various backup products, such as tape backups and seamless server redundancy? Are there alternatives for this for Linux?
Yes. And I would venture to suggest that linux probably has much better support for remote backups and failover clusters than Windows.
What assurances does a large company who absolutely can not afford significant downtime have that the software is well supported by professionals on call and that bugs are constantly being fixed?
This is Red Hat's entire business. If you need that kind of support, they would be more than happy to oblige. A number of other vendors also can provide that level of support.
There are thousands of tools that are necessary for full production environments
Yes, you are correct. And those thousands of tools are all available for linux as well...with the added bonus that they will play nicely with your Windows clients. Novell puts together a distribution that provides all of this out of the box. The only things that is really lacking is an Exchange replacement, but I see that coming in a couple of years from the various Mozilla projects. Thing is, Exchange is slowly dating itself. There are a few web services out there already, like Google Apps, that let you easily integrate email and shared calendars...and you don't need an expensive and massively proprietary application (along with the expensive and massive hardware to run it on). As these mature, Exchange is going to have to evolve or die out.
if you go to a business and tell them that they will likely say, "and what happens if it goes down?"
Hand them the business card of your local Red Hat or IBM sales representative. This is why these companies are in business, and they know how to talk to and reassure PHBs.
While legal, it would probably be exceedingly impractical to actually get PostgreSQL 8.2 running to take advantage of the latest features and fixes on Redhat 4. (from 1996, kernel 2.0.18 according to wikipedia)
Actually, it probably wouldn't be that hard. It would be a bit crazy, but PostgreSQL just needs something to compile it, and the required libraries. A better example would have been Gimp 2.4 (or some other desktop app), because that relies on libraries that require more recent kernel versions.
If you want something that supports current, bleeding edge hardware and software, Linux is the only way to go.
I sympathize with your POV, but this simply isn't true. If the hardware is closed (still lots of wireless, some raid, some nic, some ACPI, some weird usb devices like webcams and such, graphics hardware), you're going to have a hell of a time getting linux to recognize it. It is certainly true, though, that linux supports various architectures (like x86, ppc, x86-64, ia64, etc) very well and much better than Windows.
*But* being published in peer reviewed journals is still perceived as being a solid indicator of one's academic status and career progression.
And not for a bad reason. Economics is driven by productivity. A carpenter or a plumber can perform a service for somebody. A physician can treat somebody. A computer programmer can write software needed to help you run your business.
A scientist does research, but what's the difference between him and the above? The above are fairly tangible and their contribution to society easily measured. A scientists contributions are not. But people try, and the most popular method is via publications. Still, the benefit to society of basic research is a long-term affair and isn't usually realized right away. So how do you determine how much of your resources to allocate to it?
I think publications, to an extent, are a measure of productivity. If a scientist can get a research project off the ground, get relevant data, analyze it, answer an important scientific question, and put it all together into a nice story, that is productivity. If he can't, for whatever reason, it doesn't mean he is unproductive, but it is a lot harder to measure his contribution. And just like the computer programmer who is fresh out of school with no real experience to demonstrate, an employer is more likely to go with somebody who has proven himself than with somebody who hasn't.
Some of what you say is true, but I wouldn't be quite so cynical. A "useful or interesting work" can have a lot of different levels. Most journal articles represent a work in progress, not a complete understanding of a topic, so there are always more experiments to do. You just have to publish what you have at some point. Now some experiments are done sloppily and not caught by peer-reviewers, but not everybody is an expert in every field. I try to know what I'm talking about when I publish something, but if it's not directly in my expertise, a real expert can almost certainly find something wrong with it. But the article doesn't have to be perfect to get something out of it. Readers will be critical--that's their job after all if they're good scientists--and not necessarily agree with the conclusions of the authors, but they can usually conclude something from the data that is presented, and I consider that a helpful contribution to science.
I also don't agree that the vast majority are reviews, at least not in experimental science. A simple search on PubMed will tell you which articles are considered reviews and which present original research. Unless you do a very general search, articles typically outnumber reviews 10 to 1.
As in every politically driven community, there is corruption in academia. Big name scientific egos can have an undue influence over a field, but that doesn't happen very often. Most scientists are fairly collegial toward each other and respect each others work, even if they don't agree with it.
But, you are right, publish or perish is a problem, and it leads to things like blocking of competitors publication, bad grant reviews, and fabrication of data. I have, unfortunately, seen some great scientists get driven away by some sort of bs, like an insufficient number of publications. But I've also seen a lot of great scientists succeed in the system. Unfortunately, given the current economics of research, it's not a system that is going to change easily. And academia is the only place with problems. Cutthroat environments in industry can create the same sort of problems.
Peer review does effectively happen online. After an article is submitted to a journal and vetted by the editors, it is sent, usually electronically, to selected reviewers. Reviewers then submit their critique electronically. There isn't a lot of mailing of manuscripts. That, like you say, is fairly pointless in an electronic age. Critique in a forum doesn't happen, but that would be fairly impractical for a scientific article. Besides, there isn't any direct communication between reviewers and submitters. It is blind, and there isn't a lot of traffic in general--just the manuscript to be sent and the review to be received.
I do think there is an important role for journals...it allows scientific themes and significant advances to be followed more easily. Somebody else (the editors) has screened a lot of submissions--looking for things like relevance to the journal, significance of data, a well-told story, etc--before it ever makes it to print, so the reader doesn't have to wade through a ton of crap to get to the interesting article he is looking for. The economics of journals will probably certainly change, but journals themselves will remain for the near future. And nothing stops a PI from publishing their findings online if it doesn't make it into a journal. It's just that fewer people are likely to see it that way.
I find that really hard to believe. From the same site,
Median annual earnings of wage-and-salary network and computer systems administrators were $62,130 in May 2006.
If you look at the breakdown, even public schools pay more than what you say your salary is.
At any rate, that's not the point. The point is that a lot of people in the tech sector think they have an entitlement to a cushy job that makes >80k/yr with great benefits, vacation time, and 40 hour work weeks. Well, you don't. In the real world you are just another trained professional offering a service, and you will probably only ever make an average wage with average benefits--which isn't necessarily very much. If you were lobbying for better pay and working conditions in general for all trained professionals, I would sympathize. But people in the CS field don't deserve special treatment "just 'cause they know them computrons."
They can't get americans to buy their crappy pay, benefits, and job security, so they want to farm out slave labor they can have deported at their whim.
Oh cry me a river!
From the US Dept. of Labor:
In May 2006, median annual earnings of wage-and-salary computer applications software engineers were $79,780.
In May 2006, median annual earnings of wage-and-salary computer systems software engineers were $85,370.
How can you possibly suggest that a salary like that qualifies as "slave labor?" That's well above the median income of $46,326 in the U.S. per the U.S. Census. Are you aware that there are real cases of slave labor in the U.S? Such as those where a person has to work 2-3 jobs, gets no benefits, vacation, or job security, and still makes less than the poverty line?
The whiney upper-middle classers need to wake up and stop crying about their employers. As long as corporate abuse doesn't happen to them, they are ok with it. Well, guess what? India, China, and Japan are training top notch computer scientists and they are willing to work for less than Americans. That's called competition, and since that is what our capitalist economy thrives on (or so the Republicans/Libertarians keep saying), deal with it.
Wait...so you're equating blindly clicking "Yes" on an annoying dialog that pops up every 2 minutes to actively entering your password when a program wants to make a major change to your machine?
That issue aside, AppArmor and SELinux are doing quite a bit more than simple privilege separation for linux right now. Not only are they finding bugs in programs to help eliminate security issues, but they are restricting programs to exactly what they need and nothing more. I think that's going to do a lot more for linux security than UAC does for Windows.
I believe you are looking for the option,
client use spnego = yes
in your smb.conf file. That at least works for linux clients on our Win2k3 domain.
The OEM system install has been the gold standard for the non-technical end user for damn near thirty years. You unpack the box. You connect the cables and you are good to go.
Isn't that exactly what I just said? Windows comes preinstalled and configured, so it is what people use and what major companies write software for. Actual usability has very little to do with it.
People don't want to be dependent on a geek who may not be around when they need help. What they want is the toll free number, the extended warranty, the in-home service contract.
Agreed. But you're make a lot of assumptions here. First of all, service contracts are available for Linux. Second, what Windows users do you know who prefer telephone support to asking somebody they know for help? Computer users like quick fixes to their problems...they don't really care how they get it. If they can just buy somebody coffee and have them come over and fix their problem, they usually much prefer that to dealing with tech support on the phone.
Is this meant to be a joke? I just set up a Heron Compiz install, and I knew-of/had no choice but to go the the terminal at least 10 times over the course of setting it up.
Not sure what you were doing then. Just go to System|Preferences|Appearance|Visual Effects and click either "Normal" or "Fancy." All done!
For the modem, it depends on the hardware support, but did you try System|Administration|Network to configure the PPP interface?
Seriously, the tools are right there if you just look for them. I agree the help documentation can be better, though. The online forums, while helpful for advanced users, don't always give you the easiest or distro-supported ways of doing things.
It is sometimes required for some operations, usually fixing things or setting a couple things up. Such as? Seriously, people keep saying this, but I can't think of any normal operation where the CLI is required. There are many cases where I prefer to use the CLI, though, because it is more efficient and logical. I just finished installing Ubuntu 8.04 on a friend's laptop with all of the basic software, codecs, wireless support, fancy graphics, etc... and didn't have to touch the CLI at all for any of that.
This line - or something very much like it - is woven into every Linux "conversion" story posted on Slashdot.
O rly? I think you are looking through rose-tinted glasses...perhaps not considering yourself/friends "resident geeks" who can help out with computer problems. I know plenty of Windows users, and without "resident geeks" they would be completely helpless if something happened to their computer. The point is, Windows comes preinstalled on every computer and it is something people use as a result. They aren't using linux because they aren't the type to go and try to install it. Even MacOS X wouldn't be used by people if they had to install it. So if somebody installs it and sets it up for them to use, they can happily use it without problems (specific software needs aside, of course).Meanwhile, a billion users worldwide somehow manage to run Windows without the free technical support of a resident geek.
Before anyone says "but ubuntu already does this", the problem is that currently you have no choice but to learn the CLI in order to accomplish anything but basic user tasks.
Examples please. I prefer to use the CLI for a lot of things, but the GUI tools are there: hardware diagnostics, networking setup/troubleshooting, basic server configuration, VM configuration, user/group management, installing/managing software, directory services. What are you doing that requires the CLI?
Well, it's kind of a pain to find out what the chipsets are from the Best Buy website, so I'm not going to go through them all. Here is a list of all the chipsets supported by the madwifi driver in linux. If you get a card with one of these chipsets, it should work out of the box. But, you are going to have to verify that the card you are buying has the right chipset. Sometimes this means going to the manufacterer's website because the retailer doesn't always give you that information. Like another replier already said, it looks like D-link will be your best shot. But, again, verify the chipset before you buy anything.
No, he's talking about spewing everything into c:\windows\system32 instead of having programs be self-contained. There is some argument as to whether complete self-containment is good, though. Personally, I like the way linux does it (some containment, some shared, everything tracked by the package management system).
The other thing that really annoys me is having user data written to the program directory. For crying out loud, no mainstream OS has been single user for at least 5 years. Every consumer program should be multiuser aware and put data in the right places!
Think of Service Packs as analogous to kernel patches. Those have been known to screw up a few programs, haven't they?
Nope, can't say they have. At least not official ones that come from the distribution provider. Also, service packs are ridiculous. You are practically reinstalling the entire operating system every time you install a service pack. A new kernel is typically a 10 megabyte download. If you throw a few userspace updates in there as well, you might hit 15 megabytes. How big is SP3? 300M?
I must be misunderstanding your point. Silence is silence. Whether or not you have been read Miranda rights has nothing to do with it. Do you know of an actual case where silence prior to being read Miranda rights was used against a defendant in court?
Well, having actually served on a jury (in a criminal case), I can tell you that your preconceived stereotype is as far from the truth as can be. Juries are quite well-balanced...men/women, old/young, students/professionals, etc....
So you didn't get selected for jury duty. Is it because you were too smart? Or are you too arrogant to believe you may have had a bias? Biases are inherently irrational. Even if you recognize them, you can't always prevent them from influencing your decisions. That's why the jury selection process is the way it is.
Actually, your silence can be used against you in court.
No, it really can't, not in a criminal case. That's why the jury selection process asks you things like "Do you think the defendant is guilty because he has been charged?" and "Do you think the defendant is guilty is he refuses to testify?" These are both disqualifying questions. Juries are specifically instructed to consider the defendant innocent until proven guilty and that the burden of proof is on the prosecutor. The defendant does not have to speak one word in his defense. It is the job of the prosecutor to prove his case beyond a reasonable doubt.
So, I've actually served on a jury for a murder trial, and I can tell you it was a very educational process. For one thing, I learned that premeditation does not mean planned in the case of first degree murder. It simply means that the defendant had time to reflect. It does not mean the defendant had to have a lot of time to think, it just means there was a logical sequence of actions: reflection, decision to act, and act, all of which can happen in the course of a few seconds. Now, I agree, I don't know how the jury determined there was premeditation in this case, but the legal definition of premeditation is often quite different from what people believe.
Because juries aren't allowed to speculate. They must determine whether or not the defendant is guilty based on the evidence presented to them by the prosecutor. Vague suggestions that "somebody else may have done it" would bias the jury. I know it seems silly from the outside, but there is actually a lot of well-established case law that is used to determine what is admissible evidence in a murder trial.
I would normally agree except for the part where you lose the ability to use the software if you stop paying. So if you diligently pay $115/yr. for 4 years ($460 total), and then decide you don't want to pay anymore, you can no longer use Office (i.e: access your files). At least if I pay $400 up front for Office, I can use it for as long as I want, out of date or not.
The problem Microsoft has with Office is that they really want a subscription model, but they don't have subscription value to add. Think about Napster, for example. If you pay for a subscription to Napster ($13/month), you get unlimited access to all of the songs available, which is a significant savings over buying all of the music you may listen to only every once in a while. So the subscription model has value. With Office, Microsoft isn't adding any value. They're just trying to get more cash out of you.
That's funny because I thought Bender's Big Score was terrible. My roommate and I loved the old Futurama episodes and have watched them all several times over. About 30 minutes into the movie, we were looking at each other and asking, "Is this Futurama?" I had to force myself to sit through the whole thing, hoping that it would get better, but all I got was a seemingly endless stream of jokes and gags repeated over in a bad way from the series. And the fart joke level of humor was pretty excessive. I always liked Futurama because it didn't have a lot of that, and instead had a lot of subtle humor that could really be appreciated by geeks.
I'm really hoping the second movie is better....