You are mixing up copyrights and patents. Patents are limited to geographic regions, a US patent for example is not valid in the EU and the other way around. If you want an invention protected elsewhere you have to apply for a patent in that country as well. As a result the US has software patents, and those patents are not valid in the EU. So a Europe based computer user can play mp3 for free while a US based computer user may have to pay royalties.
Re:VLC media player and MPEG-2
on
VLC 1.0.0 Released
·
· Score: 4, Insightful
You forget to add that your reply is of course very much limited to people living under US law. Software patents are afaik not valid anywhere else in the world (luckily), nor do many countries have anti-circumvention-laws like the US has. Remember that the world is bigger than the USA.
You are assuming stock bash. It is of course quite trivial (as suggested also in TFA) to change bash's history behaviour and have the details not just written to the file given in $HISTFILE but (also) silently to some remote history file, potentially on a remote server. Just to add a little more monitoring to people that deal with high value stuff.
DRM's restrictions work both ways. That's something people tend to forget. And what at least the music companies by now seem to realise.
Allowing sales of non-DRM music by the labels breaks the virtual monopoly of Apple over digital music sales. Not only can non-iPod users play music bought from ITMS (not really interesting as this is only a small part of the market), more importantly other retailers such as Amazon can sell for the iPod, giving the content providers more say in the payments and prices.
Until the sales of non-DRM popular music (and face it: the sales of the music from the big four is like 80% or more of the market, the independent labels don't really count in this), Apple had a virtual monopoly as retailer. Both ways. They were the only one that could sell DRM enabled music to iPod holders (70+% of the market). And the only retailer that iPod holders could turn to to buy the most popular music.
Content providers always think that DRM gives them more power and control - while it's in practice the middleman that is getting the control. And when that middleman (Apple) gets like >80% of the market... that's where the real power and control ends up. It fragments the market and kills competition, in a way that is often (paradoxically) bad for the producers.
Add to that the great technical restrictions an F1 car is built under: minimum weight, minimum ground clearance, maximum engine size, etc. This to keep the sport safe: these restrictions are simply to keep the speeds down. Without those restrictions F1 would go much faster, and accidents of course would be far worse. Bugatti doesn't have those technical restrictions, so it's indeed no surprise that you can build a car that accelerates faster and has a higher top speed than a F1 car.
Don't worry, people now in Guantanamo are not going back to the US. Where they will go to is another matter of course - so send her to Gitmo and she'll be out of the country forever. Maybe ending up in some minuscule pacific ocean island state or so.
My provider (not USA) actually states my date usage for each day that I used data, if any. So you can see exactly how much data was used on which date.
9 GB is quite impossible on 3G in a day (I get about 200 kb/s down and 60 kb/s up - so saturating my line up and down at the same time would be no more than about 2.2 GB of data transfer per day). So if it were as Mr Savage claims and being charged "9GB for a few hours of web surfing", then presumably that is spread over no more than a day or two. If it is charged in a single day then it's of course even more suspect.
And indeed as other posters said, the math doesn't add up: 0.015 cents times 9 GB is no where near 11,000 - not even orders of magnitude off, it's just completely off: 1.5x9 = 13.5, and then I'd only have to move the decimal point around to get to the 11000??
Now see how much solar energy the moon catches on it's surface, and compare that to the amount of energy that is prevented from reaching the earth. And then you get to the point of my post.
That depends on where this satellite is. And how big it is.
It would have to be really big to block more than it sends to us. Think of the moon. That's quite a big satellite. Now imagine the moon would become our power generator, and beam all the solar energy it receives to the earth. That's a lot of power.
Now think of how much energy the moon blocks: next to nothing. Even at a total solar eclipse the moon blocks the sunlight only to a very tiny spot on the earth - the size of the full shadow is far smaller than the size of the moon itself. Had the moon been only a little further away from the earth we would never have had a total solar eclipse as the moon as seen from the earth would be smaller than the sun.
A solar satellite I would expect to be way smaller than the moon, even taking the closer proximity to the earth into account. So even when the satellite is right in between the earth and the sun, it will not block much of the sun's energy. So virtually all energy caught by such a satellite would indeed never have reached the earth in the first place.
I have a fixed IP address (according to my provider, BizNetvigator - I'm paying for a fixed address at least!) but according to SORBS I am in a "dynamic IP range", and they can not and will not unlist my IP address. As a result I am forced to relay my mails through the mail server of my provider. Totally unnecessary but it's the only way to assure delivery of e-mails. Many of my mails are rejected and bounce at smtp handshake level, I guess there will be plenty that are silently dropping it - both I consider bad practice, I want to receive my suspected junk, dump it in a junk folder, and look through it once a day to make sure. Greylisting takes care of 95% of the spam already, so only a dozen or so junks come in every day.
Also I do see sometimes my mails being greylisted, but as I'm running a real mail server that just causes some delays. It will try again shortly after.
Ray points out some extremely simple things that were overlooked. Here's one example:
The jury could have been instructed that no statutory damages could be awarded as to any work whose copyright registration effective date was subsequent to the date of defendant's commencement of use of Kazaa [or the Court could itself have made that determination based on the answers to the verdict form].
It was here that I got lost.
As I read it, Ray says that if she started using Kazaa at a certain date, that she is free to distribute any works with copyrights registered after that date. And that of course can't be right. If she offers a song on Kazaa after the copyright of that song is registered, it's potentially infringing. Starting to use Kazaa before that copyright registration should be moot. Indeed if she would use Kazaa and distribute works with then unregistered copyrights (something that I highly doubt: I would assume the RIAA members would register their copyrights before commencing distribution).
Now Ray is the lawyer here and I'm not even an American, so I presume to have misunderstand things.
Also I understand that the RIAA may be overlawyered, it is a bit scary to me that a Court can have the wool pulled over their eyes so easily, and accept cases that according to Ray should have been laughed out of court. Knowing court cases are often highly complex I strongly suspect a huge oversimplification here from Ray's side.
Also I don't agree with his opinion that a single song is not a work, but that only a whole album is a work. I see a single song as a work, and an album as a separate (derivative?) work based on those songs. Or a collection of works, the collection in itself being a work.
But in the end it seems his "dream" boils down to a much-discussed point: is offering for distribution the same as distribution? And aren't the statutory damages on the high side?
Join a (sports) club - the most common way to meet people. Real people. And don't worry if you're not that physically strong; join a cards or chess club or so: mind sports are also sports. As geek you probably have the brains for it. And especially when joining a mind sport club you have a good chance there are males and females playing together.
Well lets summarise this easily: it's virtually impossible to get useful memory use stats on a modern O/S.
Shared libraries may be shared between processes of the program you are looking at, and other programs. Now how much memory to account for each process? Or any at all if it is used by other programs? But what if those programs are optional as well?
And then you start talking about javascript and java. How much of that should be attributed to the browser? All? I can't think of any java/javascript that is used out of the browser environment, but still. And then how about other external programs, like video players? Only the plugin part to be counted? Or all of it?
It is for sure not easy, and saying "someone competent" indicates how much you have thought about it. Which is not much at all.
Extraction of business rules from legacy (probably COBAL) system.
Farm it out to other universities or India.
(Cost: maybe $1 million)
Basic requirements and documentation finalized
This should not be filtered from the code - it probably can not be filtered from the code.
These requirements will have to be provided by the various institutions and workgroups and contractors and tax specialists. But getting those requirements is probably a dog in itself: many high-level employees will have personalised contracts with personalised payment scales and benefits, for each of them the rules will have to be filtered from the contract. This is not going to be easy. And it is going to be really hard to convince those workgroups to go through years-old contracts and updates for replacing a payroll that (presumably, and certainly from their pov), "just works". This is the really hard part. The next step is easy. Implementing rules is easy. A lot of work probably, but no more than that.
But making the system maintainable later - that is another matter. How are you going to implement rule changes? Tax rule changes? New contracts? The worst you can tell an applicant that "this salary related request you have we can not allow you because our payroll can't handle it". It is going to have to be mighty flexible. And with that mighty complex.
Aren't extras provided by the company (car, house, food allowance, etc) not considered part of your income in the US? In many countries this is considered income - albeit not paid in cash. For example, in The Netherlands, if the company provides you a car for private use, you have to add 25% of the new value of that car to your annual income, and pay tax over that amount. Similar rules apply for company provided housing and so.
She has infringed on copyright, and should be punished for that. I think that is clear enough here. If you break the law and get caught, be prepared to be punished. That the punishment amount is way higher than the value of the direct loss, only makes sense. Imagine if you infringe on a $1 copyright, and have to later pay $1 if caught. The risk of getting caught is quite low. Then why not just run the risk of getting caught? It won't cost you extra.
It is not even a matter of how much she would have spent on music, if anything at all. That is irrelevant.
The only point you can make here is that the statutory damages are excessive punishment. I would argue that 100-200 times would be more reasonable.
The main problem there (for the rights holders) is that doing some research and issuing a lawyer's letter requesting settlement to someone costs more already than the couple thousand dollars they could get in statutory damages in such a case, as most people do not infringe on too many works at a time. This is making the whole enforcement process uneconomical at best.
There have been similar law suits against Google's image search - at least I recall some porn magazines that saw their thumbnailed images appear on Google and were not happy with it. Sorry I'm too lazy to look them up, but the very existence of Google Image Search means that there is no big problem here. Storing/showing thumbnail images while linking to the original is probably considered fair use.
Airliners are not going to give you a refund or allow you to change your booking because you happen to run a fever. Never been, probably never will be (although with swine flue they are being pushed to do so). So the choice is to just take some medicine to feel better and go, or to lose your ticket. That usually includes the return leg as well. So unless you are rich enough that money doesn't matter much to you, airliners pretty much force you to go ahead with your schedule, fever or no fever.
A proper patch would imho only be able to break existing functionality if:
it changes the behaviour of a publicly documented API (it shouldn't but it can be documented),
the software providing the functionality uses an undocumented API or uses a bug workaround, the first it shouldn't do in the first place and the second is up for debate whether it's good to do or not.
Changing a documented API should happen only between OS version changes, the second is more likely. And considering the number of bugs and undocumented API calls included in Windows that may well be a serious issue. Documenting the patch will never warn one of these issues: the undocumented API calls are, well, undocumented so technically they do not exist, and it is impossible to know beforehand which bug workarounds there are in software, if any.
So assuming MS writes their patches properly, no documented functionality will change. It may change to what the documents say it does, it may internally change giving the same end result - so no matter the documentation, testing would be the only way to make sure that your specific set of third-party or in-house software still works.
And I'm sure the above accounts for open source software as much as it does for closed source.
Massive monoculture is always dangerous. The dinosaurs seemed incredibly successful, too, but too many of them were too similar--and look what happened. In diversity there is strength.
In numbers there is strength as well. There is quite some evidence that birds are the living direct descendants of the dinosaurs - and in a way I have always been puzzled on how it would be possible that all dinosaurs would become extinct but other types of animals (mammals, crocodiles) not. Dinosaurs were often huge animals, so relative few numbers before the earth is full. That is more likely to have been their undoing. When 90% gets killed, finding a mate becomes really hard due to the huge distance between individuals.
Windows is so huge in numbers that it is almost impossible to extinct. Almost always there will be some Windows computers surviving somewhere, forgotten on grandma's table, not connected to the Internet even maybe and happily moving on alone. It is impossible to wipe them all out, there are too many of them.
OS/2 is virtually extinct - some installations hanging on for dear life but there were so few of them... BeOS saw the same fate... and so there are more. Dead branches on the tree of evolution, they could not multiply sufficiently to weather the competition.
Windows is of course at risk of disease: all individuals are so similar they can easily infect one another. Some have better immune systems (firewalls, more patches installed) and may survive longer - they may even survive the main onslaught and survive the virus which itself may die out due to not enough hosts left to infect. That is after all what happened to the Spanish Flue: this strain disappeared because in the end all hosts were either immune or had died. There were virtually no fresh hosts available for the virus to survive.
Linux is reaching sufficient numbers now to also be impossible to become extinct, and add to that the large diversity in systems giving the species great immunity. Yes some groups may be vulnerable to a certain virus, others will be immune and sit out the disease. Then the ones killed by the virus will be replaced by new, immune systems and the species as a whole becomes stronger.
At the moment actually I can not think of other operating systems that are as diverse as the Linux platform. BSD is a candidate but only three major flavours available. Windows certainly is no candidate, it's all the same.
I second that. If I'm looking for some software I first look in my distro's repository (Mandriva and Debian for me). Plus in case of Mandriva the PLF repository. Works for 95% of the cases, a few clicks and done. If not there, I check for a package for my distro (rpm resp. deb). And if that doesn't work I normally give up already. I've tried (and often succeeded) compiling from source, but it's cumbersome and usually just not worth the effort.
Installing in OS-X is also a breeze: double-click the.dmg and drag&drop the app usually. Windows on the other hand... back to the stone-age. It really all works different, and everyone is re-inventing the wheel it seems.
Oh and then I'm not even talking about the effort to find the software in the first place, especially if you don't have a name but just a general description! Such as "photo editor". You can search a Linux distro's repository for "photo editor" and get a handful options. Windows and OS-X have no such options, much more digging required...
This is quite close to my first thought about this: why does it have to be free? Just because it is for Linux or Open Source? Open Source does not necessarily mean free-of-charge, not even free-as-in-speech. That one can look at the source and modify it doesn't necessarily mean you have the right to redistribute it, or that you can obtain it for free.
Now if only they can come with a simple way to pay small amounts (and that is a big issue - without having to buy "credits" in advance or whatever) I think it can give a great boost to open-source developers. If an application is good, well yes I'd happily donate a small amount (though much rather after obtaining it; not beforehand - try before you buy). Not US$50 or so - more like a dollar or two. Let a couple hundred people do so and the developer can buy himself a nice upgrade for his computer. Always nice when your hobby gives you something real in return.
Slashdot Mirror
You are mixing up copyrights and patents. Patents are limited to geographic regions, a US patent for example is not valid in the EU and the other way around. If you want an invention protected elsewhere you have to apply for a patent in that country as well. As a result the US has software patents, and those patents are not valid in the EU. So a Europe based computer user can play mp3 for free while a US based computer user may have to pay royalties.
You forget to add that your reply is of course very much limited to people living under US law. Software patents are afaik not valid anywhere else in the world (luckily), nor do many countries have anti-circumvention-laws like the US has. Remember that the world is bigger than the USA.
You are assuming stock bash. It is of course quite trivial (as suggested also in TFA) to change bash's history behaviour and have the details not just written to the file given in $HISTFILE but (also) silently to some remote history file, potentially on a remote server. Just to add a little more monitoring to people that deal with high value stuff.
DRM's restrictions work both ways. That's something people tend to forget. And what at least the music companies by now seem to realise.
Allowing sales of non-DRM music by the labels breaks the virtual monopoly of Apple over digital music sales. Not only can non-iPod users play music bought from ITMS (not really interesting as this is only a small part of the market), more importantly other retailers such as Amazon can sell for the iPod, giving the content providers more say in the payments and prices.
Until the sales of non-DRM popular music (and face it: the sales of the music from the big four is like 80% or more of the market, the independent labels don't really count in this), Apple had a virtual monopoly as retailer. Both ways. They were the only one that could sell DRM enabled music to iPod holders (70+% of the market). And the only retailer that iPod holders could turn to to buy the most popular music.
Content providers always think that DRM gives them more power and control - while it's in practice the middleman that is getting the control. And when that middleman (Apple) gets like >80% of the market... that's where the real power and control ends up. It fragments the market and kills competition, in a way that is often (paradoxically) bad for the producers.
Add to that the great technical restrictions an F1 car is built under: minimum weight, minimum ground clearance, maximum engine size, etc. This to keep the sport safe: these restrictions are simply to keep the speeds down. Without those restrictions F1 would go much faster, and accidents of course would be far worse. Bugatti doesn't have those technical restrictions, so it's indeed no surprise that you can build a car that accelerates faster and has a higher top speed than a F1 car.
Don't worry, people now in Guantanamo are not going back to the US. Where they will go to is another matter of course - so send her to Gitmo and she'll be out of the country forever. Maybe ending up in some minuscule pacific ocean island state or so.
My provider (not USA) actually states my date usage for each day that I used data, if any. So you can see exactly how much data was used on which date.
9 GB is quite impossible on 3G in a day (I get about 200 kb/s down and 60 kb/s up - so saturating my line up and down at the same time would be no more than about 2.2 GB of data transfer per day). So if it were as Mr Savage claims and being charged "9GB for a few hours of web surfing", then presumably that is spread over no more than a day or two. If it is charged in a single day then it's of course even more suspect.
And indeed as other posters said, the math doesn't add up: 0.015 cents times 9 GB is no where near 11,000 - not even orders of magnitude off, it's just completely off: 1.5x9 = 13.5, and then I'd only have to move the decimal point around to get to the 11000??
Now see how much solar energy the moon catches on it's surface, and compare that to the amount of energy that is prevented from reaching the earth. And then you get to the point of my post.
That depends on where this satellite is. And how big it is.
It would have to be really big to block more than it sends to us. Think of the moon. That's quite a big satellite. Now imagine the moon would become our power generator, and beam all the solar energy it receives to the earth. That's a lot of power.
Now think of how much energy the moon blocks: next to nothing. Even at a total solar eclipse the moon blocks the sunlight only to a very tiny spot on the earth - the size of the full shadow is far smaller than the size of the moon itself. Had the moon been only a little further away from the earth we would never have had a total solar eclipse as the moon as seen from the earth would be smaller than the sun.
A solar satellite I would expect to be way smaller than the moon, even taking the closer proximity to the earth into account. So even when the satellite is right in between the earth and the sun, it will not block much of the sun's energy. So virtually all energy caught by such a satellite would indeed never have reached the earth in the first place.
Luckily a pound doesn't weigh that much. Otherwise with 200 pounds in your wallet you could hardly walk.
That is just the seller trying to cheat you into selling you almost 10% less than what you pay for!
2 lbs = 0.907 kg.
I have a fixed IP address (according to my provider, BizNetvigator - I'm paying for a fixed address at least!) but according to SORBS I am in a "dynamic IP range", and they can not and will not unlist my IP address. As a result I am forced to relay my mails through the mail server of my provider. Totally unnecessary but it's the only way to assure delivery of e-mails. Many of my mails are rejected and bounce at smtp handshake level, I guess there will be plenty that are silently dropping it - both I consider bad practice, I want to receive my suspected junk, dump it in a junk folder, and look through it once a day to make sure. Greylisting takes care of 95% of the spam already, so only a dozen or so junks come in every day.
Also I do see sometimes my mails being greylisted, but as I'm running a real mail server that just causes some delays. It will try again shortly after.
Ray points out some extremely simple things that were overlooked. Here's one example:
The jury could have been instructed that no statutory damages could be awarded as to any work whose copyright registration effective date was subsequent to the date of defendant's commencement of use of Kazaa [or the Court could itself have made that determination based on the answers to the verdict form].
It was here that I got lost.
As I read it, Ray says that if she started using Kazaa at a certain date, that she is free to distribute any works with copyrights registered after that date. And that of course can't be right. If she offers a song on Kazaa after the copyright of that song is registered, it's potentially infringing. Starting to use Kazaa before that copyright registration should be moot. Indeed if she would use Kazaa and distribute works with then unregistered copyrights (something that I highly doubt: I would assume the RIAA members would register their copyrights before commencing distribution).
Now Ray is the lawyer here and I'm not even an American, so I presume to have misunderstand things.
Also I understand that the RIAA may be overlawyered, it is a bit scary to me that a Court can have the wool pulled over their eyes so easily, and accept cases that according to Ray should have been laughed out of court. Knowing court cases are often highly complex I strongly suspect a huge oversimplification here from Ray's side.
Also I don't agree with his opinion that a single song is not a work, but that only a whole album is a work. I see a single song as a work, and an album as a separate (derivative?) work based on those songs. Or a collection of works, the collection in itself being a work.
But in the end it seems his "dream" boils down to a much-discussed point: is offering for distribution the same as distribution? And aren't the statutory damages on the high side?
Join a (sports) club - the most common way to meet people. Real people. And don't worry if you're not that physically strong; join a cards or chess club or so: mind sports are also sports. As geek you probably have the brains for it. And especially when joining a mind sport club you have a good chance there are males and females playing together.
Well lets summarise this easily: it's virtually impossible to get useful memory use stats on a modern O/S.
Shared libraries may be shared between processes of the program you are looking at, and other programs. Now how much memory to account for each process? Or any at all if it is used by other programs? But what if those programs are optional as well?
And then you start talking about javascript and java. How much of that should be attributed to the browser? All? I can't think of any java/javascript that is used out of the browser environment, but still. And then how about other external programs, like video players? Only the plugin part to be counted? Or all of it?
It is for sure not easy, and saying "someone competent" indicates how much you have thought about it. Which is not much at all.
Phase 1
Extraction of business rules from legacy (probably COBAL) system. Farm it out to other universities or India. (Cost: maybe $1 million) Basic requirements and documentation finalized
This should not be filtered from the code - it probably can not be filtered from the code.
These requirements will have to be provided by the various institutions and workgroups and contractors and tax specialists. But getting those requirements is probably a dog in itself: many high-level employees will have personalised contracts with personalised payment scales and benefits, for each of them the rules will have to be filtered from the contract. This is not going to be easy. And it is going to be really hard to convince those workgroups to go through years-old contracts and updates for replacing a payroll that (presumably, and certainly from their pov), "just works". This is the really hard part. The next step is easy. Implementing rules is easy. A lot of work probably, but no more than that.
But making the system maintainable later - that is another matter. How are you going to implement rule changes? Tax rule changes? New contracts? The worst you can tell an applicant that "this salary related request you have we can not allow you because our payroll can't handle it". It is going to have to be mighty flexible. And with that mighty complex.
Aren't extras provided by the company (car, house, food allowance, etc) not considered part of your income in the US? In many countries this is considered income - albeit not paid in cash. For example, in The Netherlands, if the company provides you a car for private use, you have to add 25% of the new value of that car to your annual income, and pay tax over that amount. Similar rules apply for company provided housing and so.
She has infringed on copyright, and should be punished for that. I think that is clear enough here. If you break the law and get caught, be prepared to be punished. That the punishment amount is way higher than the value of the direct loss, only makes sense. Imagine if you infringe on a $1 copyright, and have to later pay $1 if caught. The risk of getting caught is quite low. Then why not just run the risk of getting caught? It won't cost you extra.
It is not even a matter of how much she would have spent on music, if anything at all. That is irrelevant.
The only point you can make here is that the statutory damages are excessive punishment. I would argue that 100-200 times would be more reasonable.
The main problem there (for the rights holders) is that doing some research and issuing a lawyer's letter requesting settlement to someone costs more already than the couple thousand dollars they could get in statutory damages in such a case, as most people do not infringe on too many works at a time. This is making the whole enforcement process uneconomical at best.
There have been similar law suits against Google's image search - at least I recall some porn magazines that saw their thumbnailed images appear on Google and were not happy with it. Sorry I'm too lazy to look them up, but the very existence of Google Image Search means that there is no big problem here. Storing/showing thumbnail images while linking to the original is probably considered fair use.
Airliners are not going to give you a refund or allow you to change your booking because you happen to run a fever. Never been, probably never will be (although with swine flue they are being pushed to do so). So the choice is to just take some medicine to feel better and go, or to lose your ticket. That usually includes the return leg as well. So unless you are rich enough that money doesn't matter much to you, airliners pretty much force you to go ahead with your schedule, fever or no fever.
How about mining? I'm sure there is a lot of interesting stuff lying under the desert sands.
There are plenty of reasons for people wanting to live in the desert. If only because the rest of the world is getting quite full.
A proper patch would imho only be able to break existing functionality if:
Changing a documented API should happen only between OS version changes, the second is more likely. And considering the number of bugs and undocumented API calls included in Windows that may well be a serious issue. Documenting the patch will never warn one of these issues: the undocumented API calls are, well, undocumented so technically they do not exist, and it is impossible to know beforehand which bug workarounds there are in software, if any.
So assuming MS writes their patches properly, no documented functionality will change. It may change to what the documents say it does, it may internally change giving the same end result - so no matter the documentation, testing would be the only way to make sure that your specific set of third-party or in-house software still works.
And I'm sure the above accounts for open source software as much as it does for closed source.
Massive monoculture is always dangerous. The dinosaurs seemed incredibly successful, too, but too many of them were too similar--and look what happened. In diversity there is strength.
In numbers there is strength as well. There is quite some evidence that birds are the living direct descendants of the dinosaurs - and in a way I have always been puzzled on how it would be possible that all dinosaurs would become extinct but other types of animals (mammals, crocodiles) not. Dinosaurs were often huge animals, so relative few numbers before the earth is full. That is more likely to have been their undoing. When 90% gets killed, finding a mate becomes really hard due to the huge distance between individuals.
Windows is so huge in numbers that it is almost impossible to extinct. Almost always there will be some Windows computers surviving somewhere, forgotten on grandma's table, not connected to the Internet even maybe and happily moving on alone. It is impossible to wipe them all out, there are too many of them.
OS/2 is virtually extinct - some installations hanging on for dear life but there were so few of them... BeOS saw the same fate... and so there are more. Dead branches on the tree of evolution, they could not multiply sufficiently to weather the competition.
Windows is of course at risk of disease: all individuals are so similar they can easily infect one another. Some have better immune systems (firewalls, more patches installed) and may survive longer - they may even survive the main onslaught and survive the virus which itself may die out due to not enough hosts left to infect. That is after all what happened to the Spanish Flue: this strain disappeared because in the end all hosts were either immune or had died. There were virtually no fresh hosts available for the virus to survive.
Linux is reaching sufficient numbers now to also be impossible to become extinct, and add to that the large diversity in systems giving the species great immunity. Yes some groups may be vulnerable to a certain virus, others will be immune and sit out the disease. Then the ones killed by the virus will be replaced by new, immune systems and the species as a whole becomes stronger.
At the moment actually I can not think of other operating systems that are as diverse as the Linux platform. BSD is a candidate but only three major flavours available. Windows certainly is no candidate, it's all the same.
I second that. If I'm looking for some software I first look in my distro's repository (Mandriva and Debian for me). Plus in case of Mandriva the PLF repository. Works for 95% of the cases, a few clicks and done. If not there, I check for a package for my distro (rpm resp. deb). And if that doesn't work I normally give up already. I've tried (and often succeeded) compiling from source, but it's cumbersome and usually just not worth the effort.
Installing in OS-X is also a breeze: double-click the .dmg and drag&drop the app usually. Windows on the other hand... back to the stone-age. It really all works different, and everyone is re-inventing the wheel it seems.
Oh and then I'm not even talking about the effort to find the software in the first place, especially if you don't have a name but just a general description! Such as "photo editor". You can search a Linux distro's repository for "photo editor" and get a handful options. Windows and OS-X have no such options, much more digging required...
This is quite close to my first thought about this: why does it have to be free? Just because it is for Linux or Open Source? Open Source does not necessarily mean free-of-charge, not even free-as-in-speech. That one can look at the source and modify it doesn't necessarily mean you have the right to redistribute it, or that you can obtain it for free.
Now if only they can come with a simple way to pay small amounts (and that is a big issue - without having to buy "credits" in advance or whatever) I think it can give a great boost to open-source developers. If an application is good, well yes I'd happily donate a small amount (though much rather after obtaining it; not beforehand - try before you buy). Not US$50 or so - more like a dollar or two. Let a couple hundred people do so and the developer can buy himself a nice upgrade for his computer. Always nice when your hobby gives you something real in return.