You have just perfectly described religion when it comes to software.
No, I described the scientific method. If you want to call it religion, go ahead, I won't argue with you over semantics.
Please address this comment: non-free software cannot be studied, cannot be audited, and so is incapable of providing scientific results, as long as it makes even a part of methodology unknowable and unverifiable.
Here's an example of a valid scientific study: (1) collect data (2) apply a specific, known algorithm to process data (2) draw a pretty scatter-plot based on the processed data. When you use Mathematica, you have to replace step (2) with "???". Where one expects proof, you appeal to authority (Mathematica is popular and respected, therefore its results are correct). In truth, you don't even know which algorithm you used. You know where else they use appeals to authority? Traditional religions.
If you cannot refute the thesis just above, you must agree that software like Mathematica is useless for doing science. And if you agree with that, you will also agree it is useless in science education.
Look, you keep blaming me for "religious zealotry", of which I have none. While it may appear otherwise to people who don't read what I write, I am not persuaded by Stallman's ethical argument, which is also his only argument. None of the reasons for using a free OS I gave you have anything to do with religion or ethics. They have to do with security, privacy, total cost of ownership, the ability to make the software to do what you want, the freedom to study the software itself, and the very nature of the scientific method. Why won't you address just this one thesis:
Non-free software should not be used in either science or science education because the methodology is secret, the correctness of results is unverifiable, and therefore no valid scientific results can ever be obtained with the non-trivial help from non-free software. To help you with non-trivial: a paper typed in MS Word may be scientific, but a histogram plotted from a data set processed in Mathematica is just voodoo.
Still, this article is talking about choosing Linux-based OS. I am curious to hear which distributions you deem to have "software vendor is spying on me, wasting my cycles on ads, and leaving the back door open" that got you so riled up?
Ubuntu is one of the worst offenders. They think it's OK to peddle non-free software. More than that: that's how they get paid. They are now openly spying on users by default. The fact that they are open about it does little to repair the damage it does to the users. If you know that FBI is legally filming your apartment windows 24/7 and has a tail on you wherever you go, your privacy is still invaded.
And if the search bar was not enough to convince me, Ubuntu also hosts Adobe Flash in its own repository and suggests it to its users. So they get paid for letting Adobe spy on your files, your computer use, and your Web browsing, and who knows what else, and for all we know Canonical itself spies through Flash by proxy (Adobe could as well give them an information cut). And what do you get for that abuse? A shitty Trisquel-like OS that is capable of running on unmaintainable hardware and showing low-res movies of a cat playing piano.
You link to FSF page that lists only nine distributions, which honestly are not very widely used.
This is irrelevant. Slackware, which is arguably one of the best distributions out there (from the engineering point of view), is not very widely used.
To recommend a newcomer starts from a more obscure, restrictive distribution is counter-productive
Trisquel is neither obscure nor restrictive. It's basically free Ubuntu. How can you even call free distributions restrictive? They are much easier to get help with, too, because every bug or behavior can be traced to the source code, and most of them have been by now. I know what you mean though: you seem to think that excluding non-free drivers and firmware is "restrictive". But that is nonsense. This is like saying that a car dealership free of lemons is "restrictive". Trisquel actually helps its users by cutting all of the buggy and sneaky proprietary crap.
In this particular case, where the OP is getting 3 to 4 year old desktops for free, this "restriction" does not exist in any form. All he needs to do is pick machines that are supported by a free kernel (given the age and the form factor, it's pretty much all of them). He will immediately see improvement in stability, if nothing else.
The purpose of a lab is to educate your students, not indoctrinate them in your in your religious beliefs.
Utter nonsense. Choosing to use free software in education is not a religious belief and it doesn't indoctrinate anyone. The reason for purging non-free software may be religious, but that's kind of irrelevant, no? In this case, however, it's not religious: free software should be the only option in education and science, since it can be studied. Non-free software cannot be studied, cannot be audited, and so is incapable of providing scientific results, since the methodology is voodoo.
Even further, the OP is a mathematician. How do you even start justifying the use of non-free software in math education? What's next? Secret proofs?
I am a firm believer in the Open Source philosophy...
I stopped reading right there.
Weird, that's where I started paying attention, may be because I believe that non-free software does not belong in education. Because one can't study non-free software. Because hiding the code and the methods is exactly the opposite of science. But most of all because you are fatally wrong in your assessment of the free software influence. Religious indoctrination is what non-free software does: it asks you to accept the vendor's good will with blind faith, and it asks you to run out to the store and pick up some coke while you are at it. Free software just does math. If FSF's mission is indoctrination, they are really lousy at it. Proprietary software vendors, OTOH, are all about indoctrination: they spend billions of US dollars on ad campaigns to convince users to buy non-free software, ostensibly because sharing files and code is bad for "content developers".
Anyways, OP, stay away from Edubuntu: it's non-free by design. If you can, run a fully free distribution such as Trisquel, a Ubuntu offshoot and an apparent flagship of FSF. It is not too hard to find old desktops that are fully supported by the linux-libre kernel, and even incomplete support (for example, slow 3D graphics and no wireless) may be quite sufficient for your needs.
Yeah, I need the best solution for my problem, very wise, so let me break it down for you. My problem is that my software vendor is spying on me, wasting my cycles on ads, and leaving the back door open. My solution--using free software. If you think you don't have the above-mentioned problems, you are just a sap. You paid premium for a vastly inferior product that was designed to exploit you, while perfectly good free substitutes are already on the table.
Slackware and Debian are both excellent recommendations, but one has to understand that Slackware, precisely because it's so KISS, can only be maintained by a seasoned OS administrator. Coincidentally, Slackware is widely regarded as one of the best distributions to learn OS administration from, so if that's your goal, then by all means check it out.
Debian stable would be a safe bet, but keep in mind that they are about to ding.
In general, I would stay away from distributions that peddle non-free software: they are made by people who think that harming you is OK, because that's how they get paid. Ubuntu is one of the worst offenders. Debian is OK because it is free by default, and will stay that way unless you manually add evil repositories. Slackware is OK because it's trivial to purge. But also take a look at the list of fully free, FSF-approved distributions. I have no experience with any of them, but I hear that Trisquel is probably the most user-friendly, and I know that RMS is actively looking people to improve it, so it's likely to be their flagship for a while.
Alright, let's play the analogy game:) If they did what you say, then it would be closer to grey hat territory, but they didn't.
What they did was more like walking down the street and trying doors. If unlocked, they go inside, steal some valuables, and fund "research" with the proceeds. Grey hats my ass. They say they took care to make it as gentle as possible and put things back where they were, but that's like a house thief saying: I only stole $1 from each house, and I closed doors behind me.
I don't care about the legality. Considering how little harm they did, prosecution is unnecessary, IMHO (a fine would be OK). It is more pertinent that their "research" and conclusions are total trash. If they think it's OK to trespass, steal resources, potentially harm, and then present it as a "hack" and a valid research methodology (they are obviously proud of themselves), why should I believe in their academic integrity?
This is a black hat project because computers and resources were used without owners' knowledge or consent. They said they reverted them to the pre-hack state, but they can't even begin to justify this claim, since they have not a slightest idea about the respective OS configurations. The motive had a selfish component: fame. I would call it a grey-hat hack if it provided significant benefit to people whose computers got hacked, but this is not the case here.
Applications that do not support one of those are clearly not intended to be remotely run.
Fuck you.
Is this a personal issue or something? Besides, I never claimed you can't run these remotely. Indeed, you can use X till hell freezes over and run anything you want remotely until the end of time. That doesn't mean that the software you are running on top was intended for remote use. Most software is not. And no matter how good NX is, XTHML will destroy it in interface latency, bandwidth, and security categories, and match in everything else.
Don't be mad at me for telling you this. Be mad at the developers who fail to provide the modern level of remote access for applications that obviously need it.
But no one is writing network-friendly apps for X anymore, and so network transparency is not nearly as useful as it used to be. All modern UIs are super-cluttered with pixmaps, 3d-borders, and shadows, and take forever to render over the Internet. There are vastly more efficient ways to do network transparency, they are available today, and they do not involve X at all: text-based interface, Web server + XHTML, remote procedure calls. Applications that do not support one of those are clearly not intended to be remotely run. I used to run Mozilla thunderbird over X, and can attest that it is a nightmare. High-speed Boston to Boston was almost unbearable, and T1 in Boston to high-speed cable in Sacramento was truly at a standstill. On the other side of the spectrum are applications such as transmission, with built-in Web-server and Web interface, that are pretty and highly responsive over the worst connections.
You think pay-wall is worse then ad-wall?? Pay-walls make it hard to access the information, but ad-walls corrupt the information, form as well as substance. An ad-supported newspaper is hugely inferior to one driven by micro-payments, as the former is filled with undetectable marketing lies.
No. Annoying ads have a single feature, and it has nothing to do with layout. They "inform" how to pay top dollar for an inferior product. That's 99.9% of ads. That is also why spending money on marketing works at all: for a small price, one can mislead consumers into making irrational choices. You want examples of annoyance-free ads?
Wikimedia asking for money on wikipedia.com. A Wikipedia user already found Wikipedia useful, so the ad correctly assumes that many will be glad to contribute.
An ad for legislation aimed to cut medical costs by instituting single-payer healthcare, assuming it's effective. This can be put anywhere, as everyone but a handful of private insurers is benefited by this.
An ad telling smokers to quit smoking, placed on a cigarette pack. You get the idea. An annoyance-free ad is designed to benefit the viewer, and is only shown to people it is likely to benefit.
Enjoying the oldies subtracts from the time you could've spent watching the latest and greatest premium content, and is tantamount to theft. But you already knew that...
You are right about students' uptake of information being important, and that games should be given a prominent share of the curriculum, but some of your game examples are bizarre. People already have the real life to teach them real life skills like shopping, filling out forms, and caring grandparents and pets. Building transport networks is super-cool though. What kids mostly need the school for is to teach them skills that are impossible to obtain from the real life in one human lifetime, like rudimentary statistics, physics, history, or creative writing.
I am not surprised by Kapersky saying what he does.
Nor am I. After all, Kapersky should know all about the dangers of professionally written malware, since how he's been producing and distributing it for many years. How else can I characterize a closed-sourced package resistant to any kind of audit that claims to be an "antivirus" and gobbles up the resources to perform tasks on Kapersky's behalf?
Slashdot Mirror
You seem to think that an appeal to authority is a valid rule of inference. In mathematics, too. Oookay.
Does it occur to you that just because some way of doing science is very popular, it doesn't mean it's valid, or even superior?
Please, don't reply (unless you wont to agree for a change), just ponder.
You have just perfectly described religion when it comes to software.
No, I described the scientific method. If you want to call it religion, go ahead, I won't argue with you over semantics.
Please address this comment: non-free software cannot be studied, cannot be audited, and so is incapable of providing scientific results, as long as it makes even a part of methodology unknowable and unverifiable.
Here's an example of a valid scientific study: (1) collect data (2) apply a specific, known algorithm to process data (2) draw a pretty scatter-plot based on the processed data. When you use Mathematica, you have to replace step (2) with "???". Where one expects proof, you appeal to authority (Mathematica is popular and respected, therefore its results are correct). In truth, you don't even know which algorithm you used. You know where else they use appeals to authority? Traditional religions.
If you cannot refute the thesis just above, you must agree that software like Mathematica is useless for doing science. And if you agree with that, you will also agree it is useless in science education.
Look, you keep blaming me for "religious zealotry", of which I have none. While it may appear otherwise to people who don't read what I write, I am not persuaded by Stallman's ethical argument, which is also his only argument. None of the reasons for using a free OS I gave you have anything to do with religion or ethics. They have to do with security, privacy, total cost of ownership, the ability to make the software to do what you want, the freedom to study the software itself, and the very nature of the scientific method. Why won't you address just this one thesis:
Non-free software should not be used in either science or science education because the methodology is secret, the correctness of results is unverifiable, and therefore no valid scientific results can ever be obtained with the non-trivial help from non-free software. To help you with non-trivial: a paper typed in MS Word may be scientific, but a histogram plotted from a data set processed in Mathematica is just voodoo.
Still, this article is talking about choosing Linux-based OS. I am curious to hear which distributions you deem to have "software vendor is spying on me, wasting my cycles on ads, and leaving the back door open" that got you so riled up?
Ubuntu is one of the worst offenders. They think it's OK to peddle non-free software. More than that: that's how they get paid. They are now openly spying on users by default. The fact that they are open about it does little to repair the damage it does to the users. If you know that FBI is legally filming your apartment windows 24/7 and has a tail on you wherever you go, your privacy is still invaded.
And if the search bar was not enough to convince me, Ubuntu also hosts Adobe Flash in its own repository and suggests it to its users. So they get paid for letting Adobe spy on your files, your computer use, and your Web browsing, and who knows what else, and for all we know Canonical itself spies through Flash by proxy (Adobe could as well give them an information cut). And what do you get for that abuse? A shitty Trisquel-like OS that is capable of running on unmaintainable hardware and showing low-res movies of a cat playing piano.
You link to FSF page that lists only nine distributions, which honestly are not very widely used.
This is irrelevant. Slackware, which is arguably one of the best distributions out there (from the engineering point of view), is not very widely used.
To recommend a newcomer starts from a more obscure, restrictive distribution is counter-productive
Trisquel is neither obscure nor restrictive. It's basically free Ubuntu. How can you even call free distributions restrictive? They are much easier to get help with, too, because every bug or behavior can be traced to the source code, and most of them have been by now. I know what you mean though: you seem to think that excluding non-free drivers and firmware is "restrictive". But that is nonsense. This is like saying that a car dealership free of lemons is "restrictive". Trisquel actually helps its users by cutting all of the buggy and sneaky proprietary crap.
In this particular case, where the OP is getting 3 to 4 year old desktops for free, this "restriction" does not exist in any form. All he needs to do is pick machines that are supported by a free kernel (given the age and the form factor, it's pretty much all of them). He will immediately see improvement in stability, if nothing else.
The purpose of a lab is to educate your students, not indoctrinate them in your in your religious beliefs.
Utter nonsense. Choosing to use free software in education is not a religious belief and it doesn't indoctrinate anyone. The reason for purging non-free software may be religious, but that's kind of irrelevant, no? In this case, however, it's not religious: free software should be the only option in education and science, since it can be studied. Non-free software cannot be studied, cannot be audited, and so is incapable of providing scientific results, since the methodology is voodoo.
Even further, the OP is a mathematician. How do you even start justifying the use of non-free software in math education? What's next? Secret proofs?
I am a firm believer in the Open Source philosophy...
I stopped reading right there.
Weird, that's where I started paying attention, may be because I believe that non-free software does not belong in education. Because one can't study non-free software. Because hiding the code and the methods is exactly the opposite of science. But most of all because you are fatally wrong in your assessment of the free software influence. Religious indoctrination is what non-free software does: it asks you to accept the vendor's good will with blind faith, and it asks you to run out to the store and pick up some coke while you are at it. Free software just does math. If FSF's mission is indoctrination, they are really lousy at it. Proprietary software vendors, OTOH, are all about indoctrination: they spend billions of US dollars on ad campaigns to convince users to buy non-free software, ostensibly because sharing files and code is bad for "content developers".
Anyways, OP, stay away from Edubuntu: it's non-free by design. If you can, run a fully free distribution such as Trisquel, a Ubuntu offshoot and an apparent flagship of FSF. It is not too hard to find old desktops that are fully supported by the linux-libre kernel, and even incomplete support (for example, slow 3D graphics and no wireless) may be quite sufficient for your needs.
Yeah, I need the best solution for my problem, very wise, so let me break it down for you. My problem is that my software vendor is spying on me, wasting my cycles on ads, and leaving the back door open. My solution--using free software. If you think you don't have the above-mentioned problems, you are just a sap. You paid premium for a vastly inferior product that was designed to exploit you, while perfectly good free substitutes are already on the table.
Slackware and Debian are both excellent recommendations, but one has to understand that Slackware, precisely because it's so KISS, can only be maintained by a seasoned OS administrator. Coincidentally, Slackware is widely regarded as one of the best distributions to learn OS administration from, so if that's your goal, then by all means check it out.
Debian stable would be a safe bet, but keep in mind that they are about to ding.
In general, I would stay away from distributions that peddle non-free software: they are made by people who think that harming you is OK, because that's how they get paid. Ubuntu is one of the worst offenders. Debian is OK because it is free by default, and will stay that way unless you manually add evil repositories. Slackware is OK because it's trivial to purge. But also take a look at the list of fully free, FSF-approved distributions. I have no experience with any of them, but I hear that Trisquel is probably the most user-friendly, and I know that RMS is actively looking people to improve it, so it's likely to be their flagship for a while.
And when you get snail spam, you blacklist the area code for 6 months? Feels like an overreaction to me.
They never say it's a "hack". But they clearly mean it.
Alright, let's play the analogy game :) If they did what you say, then it would be closer to grey hat territory, but they didn't.
What they did was more like walking down the street and trying doors. If unlocked, they go inside, steal some valuables, and fund "research" with the proceeds. Grey hats my ass. They say they took care to make it as gentle as possible and put things back where they were, but that's like a house thief saying: I only stole $1 from each house, and I closed doors behind me.
I don't care about the legality. Considering how little harm they did, prosecution is unnecessary, IMHO (a fine would be OK). It is more pertinent that their "research" and conclusions are total trash. If they think it's OK to trespass, steal resources, potentially harm, and then present it as a "hack" and a valid research methodology (they are obviously proud of themselves), why should I believe in their academic integrity?
It's a very scary grey hat project.
This is a black hat project because computers and resources were used without owners' knowledge or consent. They said they reverted them to the pre-hack state, but they can't even begin to justify this claim, since they have not a slightest idea about the respective OS configurations. The motive had a selfish component: fame. I would call it a grey-hat hack if it provided significant benefit to people whose computers got hacked, but this is not the case here.
Dunno about that, but it does have Spype, the ultimate answer in targeted advertizing.
Applications that do not support one of those are clearly not intended to be remotely run.
Fuck you.
Is this a personal issue or something? Besides, I never claimed you can't run these remotely. Indeed, you can use X till hell freezes over and run anything you want remotely until the end of time. That doesn't mean that the software you are running on top was intended for remote use. Most software is not. And no matter how good NX is, XTHML will destroy it in interface latency, bandwidth, and security categories, and match in everything else.
Don't be mad at me for telling you this. Be mad at the developers who fail to provide the modern level of remote access for applications that obviously need it.
But no one is writing network-friendly apps for X anymore, and so network transparency is not nearly as useful as it used to be. All modern UIs are super-cluttered with pixmaps, 3d-borders, and shadows, and take forever to render over the Internet. There are vastly more efficient ways to do network transparency, they are available today, and they do not involve X at all: text-based interface, Web server + XHTML, remote procedure calls. Applications that do not support one of those are clearly not intended to be remotely run. I used to run Mozilla thunderbird over X, and can attest that it is a nightmare. High-speed Boston to Boston was almost unbearable, and T1 in Boston to high-speed cable in Sacramento was truly at a standstill. On the other side of the spectrum are applications such as transmission, with built-in Web-server and Web interface, that are pretty and highly responsive over the worst connections.
I am curious, what do you use remote X for?
You think pay-wall is worse then ad-wall?? Pay-walls make it hard to access the information, but ad-walls corrupt the information, form as well as substance. An ad-supported newspaper is hugely inferior to one driven by micro-payments, as the former is filled with undetectable marketing lies.
No. Annoying ads have a single feature, and it has nothing to do with layout. They "inform" how to pay top dollar for an inferior product. That's 99.9% of ads. That is also why spending money on marketing works at all: for a small price, one can mislead consumers into making irrational choices. You want examples of annoyance-free ads?
Wikimedia asking for money on wikipedia.com. A Wikipedia user already found Wikipedia useful, so the ad correctly assumes that many will be glad to contribute.
An ad for legislation aimed to cut medical costs by instituting single-payer healthcare, assuming it's effective. This can be put anywhere, as everyone but a handful of private insurers is benefited by this.
An ad telling smokers to quit smoking, placed on a cigarette pack. You get the idea. An annoyance-free ad is designed to benefit the viewer, and is only shown to people it is likely to benefit.
Enjoying the oldies subtracts from the time you could've spent watching the latest and greatest premium content, and is tantamount to theft. But you already knew that...
You are right about students' uptake of information being important, and that games should be given a prominent share of the curriculum, but some of your game examples are bizarre. People already have the real life to teach them real life skills like shopping, filling out forms, and caring grandparents and pets. Building transport networks is super-cool though. What kids mostly need the school for is to teach them skills that are impossible to obtain from the real life in one human lifetime, like rudimentary statistics, physics, history, or creative writing.
A dubious honor, considering the damage they do.
It should really read "Onanymous Coward".
Whatever. At least they will be playable by free software.
If you cheat and fail you’re a cheater. If you cheat and succeed, you’re savvy. ~Eric Cartmanez, The White Person Method
All of the above and more will get absorbed by MomCorp (TM)
I think even a nearby long GRB would only fry half the planet, being 30 seconds long.
I am not surprised by Kapersky saying what he does.
Nor am I. After all, Kapersky should know all about the dangers of professionally written malware, since how he's been producing and distributing it for many years. How else can I characterize a closed-sourced package resistant to any kind of audit that claims to be an "antivirus" and gobbles up the resources to perform tasks on Kapersky's behalf?