Obviously, "no license" is the most free. That actually depends... if it's in the public domain, then yes, but if it's copyrighted then "no license" is actually fairly restrictive.
Well he's not too happy about the BSD license either, which leaves you free to go non-free. In many ways he wants to protect the end-users from themselves by making them only use software that's free and will remain free (aka copyleft)
If a piece of software is free, it will always be free. If it's under BSD, new versions might not be free, but that doesn't affect the copy you have.
Installing proprietary software, even freeware or device blobs means you've given up the four freedoms, which in his world view you should never do.
So what is it that I can no longer do? There are things that I can't do with that software, but then I never could do those things with that software in the first place.
There's what they (FSF) call freedom 2, "The freedom to redistribute copies so you can help your neighbor". They are now saying that some forms of that are not helpful enough (despite people being willing to pay money for this help), and therefore forbidden.
Doing stupid things like renaming Firefox to IceWeezel to because of silly trademark issues, is just dumb I thought they did this because the Mozilla people said they had to if they wanted to include patches that hadn't been approved thru Mozilla? But then I didn't follow this that closely, so maybe I misunderstood something...
Debian is not strictly free as in freedom, at least not to the point to be recommended by the fsf. The main reason is, it's still too easy to install non-free stuff through debian repos. Debian isn't free because it leaves you free to go non-free?
Of course, "free" here no longer means you can distribute it however you want to, if it's part of a system then it imposes certain rules on that system nowadays...
I'd assume that a large part of the price difference is that a take-pretty-pictures camera has a *flash* for low-light conditions, or can alternatively use long exposure times. Surveillance cameras tend to be video cameras and not have flashes, which probably make low-light operation harder.
Of course, this is only relevant for shared systems. If there's only one user (or everyone shares the same login), the difference between removing ~ and reinstalling is just a couple hours.
Linux and MacOS do not get attacked, because normal users don't run with the sort of privileges that would allow the virus (or trojan as in your example) to do very much damage or replicate itself.
WTF? Any program I run has +rw access to ~ (can start itself from.profile, do arbitrary damage to all the files I actually care about, and steal passwords and the like) and the ability to connect(2) to random parts of the internet (ability to replicate, send passwords, and fetch ads). No privileges beyond this are needed to cause trouble.
The real reason is probably more to do with the size and average competency of the userbase.
But it *does* have a solution: don't give all your apps full access to ~ . Sandbox everything, and let programs fork and shrink their own sandboxes if they want to. The only part of your web browser that needs arbitrary filesystem access are the save/open dialogs, but there's no way to forbid other filesystem access outside of your profile directory. Maybe if you could do, say, "SaveDialogGenerator foo = new SaveDialogGenerator(System.Filesystem); System.Filesystem = null;" or similar, but no....
Like most absurd debates where both sides are vehemently opposed, the answer actually lies in the middle. I fail to see how ed can be considered to be "in the middle" between emacs and vi.
Ideas are not worthless. I mean, in the sense that air is worthless. Ideas may be necessary, but they're also everywhere. The problem is not finding ideas, it's putting in the work to develop the ideas enough to tell the good from the bad.
Shutting out small inventors and those who lack the funding to fully develop an idea into a marketable product is not the answer. There is a rather large difference between a working prototype and a marketable product. Models don't have to be pretty, or durable, or easily mass-producable. They just have to work.
If the poor guy can't bring the idea to market, he shouldn't be in a position to prevent others. If he comes up with the idea, he should still be able to patent it, and then he could sell it to someone that can implement it. Pricing him out of being able to patent it just ensure that he won't bother even publishing the idea in the first place, or even telling anyone about it since they'll probably just screw him over on it. Now I do think that if you own a patent, then you should have to bring a working implementation to market within a reasonable amount of time in order to keep that patent. If not, you're just obstructing progress. Ideas are worthless, and letting someone claim a monopoly on an idea is rather dumb. Working out all the details (the stuff needed to be able to provide a working model, plus the stuff needed to be make it mass producable) is the hard part, and is what's important.
2/ (In the US, at least) These companies tend to have government-granted monopoly status, where you're not allowed to compete with them. This is why US broadband sucks so much.
That's how it's convenient to sell bandwidth, sure. But it's not how people like to buy bandwidth. Say I use 1Mb/day, to check email and read the Onion. That works out to something like 11 bits/second. But you'd have to be completely insane to think that an 11 bits/second connection would be what I want. And why should I be forced to pay for 64kb/s of dedicated bandwidth, when I'll use only a tiny fraction of that?
The real costs for ISPs are not variable with use. Sure the are.
They have a fixed cost to maintain the infrastructure, and capital investments to build new infrastructure. That's it, that's all. And those costs depend on how much and how expensive equipment they need, which depends on how much bandwidth they need to provide. Faster networks are more expensive.
It doesn't get "used up", but it is limited and there is a per-byte-capacity cost. The network infrastructure costs $X/month to maintain, and provides Y Mb/s of bandwidth. That works out to $Z/Mb of capacity, which can be turned into a $A/Mb transfer (preferably with prices varying based on current network utilization, and a way to query the current price).
Most likely from a flamewar over RMS saying that, which got covered on a number of nerd-news sites.
Really, a particular work can be removed from the public domain? Or is it just that people can make proprietary new versions of it?
If a piece of software is free, it will always be free. If it's under BSD, new versions might not be free, but that doesn't affect the copy you have.
Installing proprietary software, even freeware or device blobs means you've given up the four freedoms, which in his world view you should never do.So what is it that I can no longer do? There are things that I can't do with that software, but then I never could do those things with that software in the first place.
There's what they (FSF) call freedom 2, "The freedom to redistribute copies so you can help your neighbor". They are now saying that some forms of that are not helpful enough (despite people being willing to pay money for this help), and therefore forbidden.
Of course, "free" here no longer means you can distribute it however you want to, if it's part of a system then it imposes certain rules on that system nowadays...
If the specs are open you can just write your own. At that point, why does it really matter what license their implementation has?
noexec just mean you can't execute anything *directly*. "perl nastyscript.pl" works just fine with nastyscript.pl on a noexec partition.
I'd assume that a large part of the price difference is that a take-pretty-pictures camera has a *flash* for low-light conditions, or can alternatively use long exposure times. Surveillance cameras tend to be video cameras and not have flashes, which probably make low-light operation harder.
Of course, this is only relevant for shared systems. If there's only one user (or everyone shares the same login), the difference between removing ~ and reinstalling is just a couple hours.
WTF? Any program I run has +rw access to ~ (can start itself from .profile, do arbitrary damage to all the files I actually care about, and steal passwords and the like) and the ability to connect(2) to random parts of the internet (ability to replicate, send passwords, and fetch ads). No privileges beyond this are needed to cause trouble.
The real reason is probably more to do with the size and average competency of the userbase.
But it *does* have a solution: don't give all your apps full access to ~ . Sandbox everything, and let programs fork and shrink their own sandboxes if they want to. The only part of your web browser that needs arbitrary filesystem access are the save/open dialogs, but there's no way to forbid other filesystem access outside of your profile directory. Maybe if you could do, say, "SaveDialogGenerator foo = new SaveDialogGenerator(System.Filesystem); System.Filesystem = null;" or similar, but no....
I'd guess that's why the ISPs don't seem to like this.
1/ This isn't in the US.
2/ (In the US, at least) These companies tend to have government-granted monopoly status, where you're not allowed to compete with them. This is why US broadband sucks so much.
I would assume you just leave it in the pockets on your pants when you launder them...
That's how it's convenient to sell bandwidth, sure. But it's not how people like to buy bandwidth. Say I use 1Mb/day, to check email and read the Onion. That works out to something like 11 bits/second. But you'd have to be completely insane to think that an 11 bits/second connection would be what I want. And why should I be forced to pay for 64kb/s of dedicated bandwidth, when I'll use only a tiny fraction of that?
It doesn't get "used up", but it is limited and there is a per-byte-capacity cost. The network infrastructure costs $X/month to maintain, and provides Y Mb/s of bandwidth. That works out to $Z/Mb of capacity, which can be turned into a $A/Mb transfer (preferably with prices varying based on current network utilization, and a way to query the current price).
The people who waste bandwidth on them by not installing something like adblock.