Why is font licensing any different from image licensing? The page directs you to (optionally) download font information. Your computer either does or does not. If it does, it uses the font information to render something on the page. As the server gave you this information when your computer asked for it, you legitimately have a copy. However, you are not allowed to redistribute this copy to a third party unless you have a license to do so, else you are in breach of copyright.
It's just a bunch more bits that you've downloaded off of a server. How are these bits any different from any other bits?
Google Web Toolkit contains a Java to Javascript compiler. It is the automated translation by a program of one computer language into another, just like translating C to assembly language, or assembly language to machine code, or Java to Java Bytecode, or Java Bytecode to machine code. All of the programs which do those translations are "compilers"[0]. A program which takes Java and spits out Javascript is no different. It's just another compiler, albeit with a very unusual target.
[0] Although not all are necessarily called compilers all the time. The most common type of Java Bytecode to machine language compiler, because of when it is run, is often referred to as a "JITter", which is simply short for "Just In Time Compiler". And assembly language to machine code compilers are often called "assemblers", but again that does not stop them from *being* a compiler.
Actually, I'm fuzzy on which of Java or Javascript would be considered higher- or lower-level. It's not clear-cut, and could probably be considered more of a "sideways" shift than "downwards".
HTML 5 + CSS 3 is 10 years out. Are they just whistling in the dark?
If Microsoft choose to not implement HTML5 + CSS3 for 10 years, then HTML5 + CSS3 is, to all intents and purposes, 10 years out.
There's no point in using the parts of CSS2 that IE6/7 doesn't support on your web page, because the number of people using those browsers is so big and un-ignorable that you have to find a way to do what you want using the bits that it does support. But once you've done that, what you've done generally looks correct in modern browsers too. Using the bits of CSS that IE doesn't support is therefore mostly redundant and will add to your maintenance headaches. Hence, as a developer, you're generally stuck with whatever IE chooses to support.
Being a square, boring, mold-conformer is your idea of "living it up"? Well, if that's what "living it up" is, I, for one, don't really see the appeal. If it's what you want then I hope you have fun with it. Just don't assume it's necessarily for everyone.
"So in other words, using your examples, a jew, woman, homosexual, or disabled person has more rights than a white male."
Where the fuck did the GP write that? He didn't explicitly say "if they refuse to sell to white males, that's illegal" but it's implied, just like "if they refuse to sell to black people, that's illegal" is also implied, even though he didn't explicitly say that either. Or are you calling him a racist?
Are you really so fucking dumb you can't tell the difference between a list containing a few random examples, and an authoritative complete enumeration?
We do all have equal rights. The point is that the "somebody" in "if they refuse to sell to somebody who doesn't want to prove that they came by it legally, then that's well within their rights" is irrespective of age, race, sex, sexual orientation, or any other physical characteristic you care to name. The "somebody" includes whites, blacks, jews, men, women, straights, gays, bis or whatever.
"You make it sound so simple, but you ignored the part where the company blacklists you and/or gives bad references, because you refused to accept the cut in pay."
I don't see how that would follow. If they give you option A or option B, and you select one of those options calmly and rationally, without throwing a hissy fit, whining, bad-mouthing people or generally burning bridges, why on earth would they blacklist you?
Would a company blacklist you or give you bad references if you'd just quit because you felt you weren't getting enough out of that job, or felt you could do something more rewarding elsewhere, or just realised you couldn't hack the commute? I mean, unless they're just a bunch of faceless asshats who are vindictive just for the heck of.... oh, yes, I see what you mean now.:-)
I always find it interesting when you have an ongoing agreement between a company and an employee, whereby the employee works X hours and gets paid $Y, and then the company decides it can simply ask the employee to work longer for less money.
Why? You have the exact same deal. You can decide to simply ask your employer to pay you more for reduced hours, and terminate the current deal if they don't agree to it. Nothing wrong with that.
A contract of employment is between two parties, where both sides agree to give the other something, in return for the thing the other side is giving them. You're giving the company your time, effort, expertise, and ultimately the wealth you create, which they value, while they give you money which you can exchange for wealth of types you are unable to create for yourself (food, shelter, clothing, shiny toys), which you value.
If either side feels they are not getting enough of what they value in return for what they're giving, why shouldn't they try to renegotiate for what they perceive to be a better deal? And if no mutually agreeable deal can be come to, why should the parties continue their unfair (at least to one side) exchange?
Meh. I hardly ever use tabs anymore, and have gone back to using multiple windows for all my browsing sessions.
Of course, it helps that I have a vertical taskbar on the left-hand side of my monitor. It's about 200px wide and has plenty of horizontal space to display the first few words of each window title, even with 15-20 windows open. The "Taskbar thumbnails" effect (display scaled down window in giant tooltip, along with whole window title, when hovering the taskbar item) is just a nice cherry on top.
Given that I rarely need any windows to be the full width of my screen, but find myself scrolling vertically a fair amount, the extra vertical space is really useful.
I find it's much easier to go up against a lot of people. I haven't fought just one person for so long... I've been specializing in groups, fighting gangs for local charities... that kind of thing. You see, you use different moves when you're fighting half a dozen people than when you only have to worry about one. So just one person would give me so much trouble.
"The main problem wasn't backwards compatibility. It was that installers routinely (i.e. - always) overwrote critical DLLs with their own versions and that broke other applications."
Surely it wouldn't matter if installers overwrote critical DLLs, if the DLLs maintained backwards compatibility properly like good shared libraries should! (Like good shared libraries *do* on other OSs)
*sigh*
I think our difference is down to the mindset of the system we use. I (and many other Free Software types) expect shared libraries to maintain backwards compatibility because that's just what they do. Any library that doesn't generally gets patches so that it does in future, or people don't bother using that functionality, fork the library, or just write their own version.
You, and many other Windows types, seem to expect libraries to not maintain backwards compatibility, and assume that compatibility problems will need to be worked around. This probably has the effect of making library authors for Windows not care about or pay any attention to backwards compatibility, because they know that any such problems will be worked around anyway. And without caring or paying attention to it, the workarounds *are* then needed just for the system to work.
Each of us is then looking at what the other is doing and claiming it doesn't make sense in our environment. Well, that might be correct, but you aren't doing your thing in my environment, you're doing it in yours. And in that environment, I guess it does make sense. No matter how alien it looks from over here.:-)
"Hence the Ignorance + FUD"
OK, I still don't get this. Ignorance, yes. I've not used Windows in a while, and many parts of it didn't make that much sense when I did.
But, FUD? Fear, Uncertainty and Doubt? I just don't get how this applies to a technical critique, which is what I'm trying to write.
My goal is not to scare users into thinking that Windows might not be the system for them because of vague technical or commercial demons which might come back to haunt them. Rather, it is to point out concrete problems as I see them, and to highlight ways that those problems can be solved, and have been solved (better, as far as I can tell) in other environments.
How on earth were my points FUD? Are you sure you're using the right word?
The main cause of DLL hell is shared library authors who are unable, either through incompetence or lack of discipline, to maintain backwards compatibility between minor updates. (A minor cause is bad installers overwriting newer versions of libraries with older versions. But bad installers can overwrite or just delete other files as well, as well as mess up your system in countless other ways, so that's not really a problem with the DLLs per se.)
The only reason I am aware of (other than the system/user distinction that I already made) that you might want to install multiple copies, or private copies, of shared libraries is to work around this lack of backwards compatibility. If you are aware of another reason, please state it instead of just claiming FUD without anything to back that up.
The.NET article linked to does state that maintaining backwards compatibility "is extremely difficult, if not impossible." That's a direct quote, the crux of the article, and the problem it tries to describe a solution to.
However, while maintaining backwards compatibility might require a modicum of discipline, the claim that is made simply does not stand up under scrutiny. Nearly every single shared library written for every operating system other than Windows *does* manage to maintain backwards compatibility between minor revisions. The libraries I listed, as examples of some of the largest, oldest, most complex, and most widely ported shared libraries I could think of, are concrete evidence of that. Or are you claiming that factual counterexamples to over-stated claims are "FUD"?
Having private copies of shared libraries does cause security problems when security bugs are found in those libraries. Take a look at the fallout from DSA 122-1. The 8 packages which included private copies of the zlib library were still vulnerable to the zlib bug, even after the system-wide shared library had been patched and updated. In contrast, the thousands of packages that use the shared version (1715 packages depend on zlib according to apt on my system) were all fixed as soon as the shared copy was updated. Again, demonstrable fact.
So, the basic problem is that the authors of DLLs for Windows seem incapable of maintaining backwards compatibility between minor revisions, despite the authors of shared libraries for other OSs managing it, and despite the security problems that arise from having multiple/private versions of a library installed which is a workaround that is a direct result of the lack of backwards compatibility.
Making it easier to install multiple private copies of DLLs, or allowing apps to depend on specific "side-by-side" versions of "assemblies", does not fix this problem, it simply papers over it.
Please, if I've overstated anything, claimed anything that's untrue, or made a leap of logic that is unsupportable, point it out in detail instead of just wildly claiming that my whole argument is FUD.
What's the solution to the problem? Demand that your library authors to do a better job than the one they're currently doing and maintain backwards compatibility. Especially if you're paying for your software, and those authors can't maintain the same level of professionalism as hobbyists and other people who are simply giving their code away.
You think lack of *Windows-specific* technical knowledge implies "moron"? How amusing. I'll take a knowledge and understanding of basic principles that can be applied in many scenarios over a set of vendor-specific kludges and workarounds any day thanks.
Still, thanks for the insults. Helped me to realise all the more quickly you couldn't counter any of the actual points I was making. Ad Hominem really is the most transparent and least convincing logical fallacy.
Uh, having multiple copies of the same major version of a shared library *is* DLL hell. All shared libraries should exist in either a system-wide folder (e.g./usr/lib on Unix), or in a per-user folder (e.g. ~/lib). That's why they're called *shared* libraries. Aside from the system/user distinction, you should never have multiple copies of the same major version of a shared library in different folders.
If you've got multiple system-wide copies of the same shared library on one system, that's DLL hell. If a user has more than one copy of the same shared library among their privately-installed applications, that's DLL hell..NET versioning is a stupid hack that tries to prevent DLL hell appearing to be a problem. It does not prevent DLL hell, it just stops the worst of the symptoms.
Heck, read that article you linked to:
"In the most typical case, one application will install a new version of the shared component that is not backward compatible with the version already on the machine."
Why is it not backwards compatible?
"In practice, writing code that is forever backward compatible is extremely difficult, if not impossible."
Demonstrably false. Any code written for and compiled against glibc 2.0.0 (July 1996) will work fine when used on a system with glibc 2.9.0 (Nov 2008). Ditto code written for/compiled against zlib 1.0.1 (May 96) will work against 1.2.3 (July 2005). Ditto libpng 1.0.0 (March 98) and 1.2.35 (Feb 2009). Ditto Qt 3.0.0 (Oct 2001) and 3.3.8 (Jan 2007). Ditto... almost every other Unix/Linux library ever written.
"Strong binding means an application or component can bind to a specific version of another component"
Is stupid, because when a security bug in a library is found, you *need* to update the library (or, if you have multiple copies, *all* copies of the library) to fix all the apps that use it. And the new version needs to have an updated version number, so you know that you've got the bug-free version.
Claiming that being able to load different copies of the same DLL because they're in different directories (or because.NET has this "side-by-side" feature) is not DLL hell, is like claiming that your house isn't really burning down because you're wearing a fireproof suit.
Damn, your life must be one empty, boring repetitive loop of nothingness if you think that's a life history. I hope you get to do something interesting soon.
Ah, sorry. You confused me by spuriously talking about browser plugins, instead of the subject of the article and almost every other comment which is about browser extensions. I mistakenly assumed you were talking about extensions like everyone else, only calling them plugins.
I'm a Linux user, got introduced to it in around '96 and started using it a fair amount in '99. Never experienced library hell or dependency hell.
When I started, your distro would give you a bare system, and everything else was a download, "gzip -cd source.tar.gz | tar -xf -", "./configure" and "make install" away.
If you were missing a dependency, or had a version that was too old for the software you wanted to install, configure would stop and tell you which library was missing. At which point you simply downloaded the latest version of that required library, installed it first with the same procedure, and then went back to the configure step of the software you were trying to install. OK, it's not quite a point-and-click away, but it's not excruciatingly hard either.
Never experienced library hell either. Seems that library writers for Unix/Linux know how, and have the discipline, to keep their libraries binary backwards compatible between minor revisions. I've only had installing a new minor version of a library break stuff a handful of times in 10 years, and nearly every single time it was by accident and a new version was released within a couple of days fixing it. If lib writers need to break binary compatibility (say, every 5 years or so), they generally create a new major version of the library that can be installed side-by-side with the old version (e.g. libqt-mt.so.3, libQtCore.so.4) and the two never interfere.
Nowadays I use Debian, and the combination of Debian's repository and apt-get (first released in 1998/1999, now has equivalents in nearly all modern distros) has also meant that I've not experienced any form of library or dependency hell in 5 the 5 years I've been using that.
As I understand it, Windows/.NET are the only platforms to speak of which suffer from these problems.
Also, the GP does not talk about viruses, but "the virus protection industry", which is almost entirely Windows-based (apart from scanning tools that run on other systems to check for Windows viruses) and is basically a result of the insecurity of Windows and other MS products. (See also Apache vs. IIS market share/vulnerabilities/crack rates before you even think about bringing up the "market leader gets most of the attention" excuse)
OK, it's been a while since I've been around Windows (thank God) but I follow as much tech stuff as I can and really hadn't heard anything about this before.
Could you post links to 3 other bits of software which, as well as the "normal" software they install (whatever that is) also install Firefox plugins?
Give the programmer some credit for basic intelligence: if the program really didn't work at all, they would probably have noticed. Since they haven't noticed, it must be working for them. Therefore, either you are doing something differently from them, or your environment is different from theirs.
The whole thing is worth reading, really.
Now, go file a damn bug, with a screenshot, and help make KDE rock!
"You don't own any application you use, unless you wrote it. Even under the GPL, you are permitted to modify and copy it, but the original copyright belongs to the author of that product. If you just -owned- the software you got, but could copy it freely,"
I must respectfully point out that you're talking bollocks.
If you have a book, you own that book. You own that copy of the book. You own that copy of the copyrighted material which is printed therein.
No, you do not own the copyright to the book. No-one has ever suggested that by owning a copy of a book that you own the copyright to that book.
If you have a copy of a piece of software which you have legally obtained, you own that copy of that software. Like a book, that does not mean that you own the copyright, or that you can exercise the rights of the copyright holder. It just means you own that copy. You are not allowed to make copies yourself, except where permitted by fair use, other relevant statutes, or a license from the copyright holder.
But, like owning a copy of a book, you own your copy of a piece of software.
You are, of course, free to give up some of the privileges of such ownership. So if you voluntarily agree to a binding EULA contract in which you consent to never engage in some otherwise-legal activities, and possibly give up your ownership of the copy of the software you have, that is your choice.
Fortunately, Free Software does not ask you to enter into any such agreement.
But I thought we were talking about how they were going to hack people's PCs with consent of the courts/warrants. Or are the courts now advocating torture as well?
Slashdot Mirror
Why?
Why is font licensing any different from image licensing? The page directs you to (optionally) download font information. Your computer either does or does not. If it does, it uses the font information to render something on the page. As the server gave you this information when your computer asked for it, you legitimately have a copy. However, you are not allowed to redistribute this copy to a third party unless you have a license to do so, else you are in breach of copyright.
It's just a bunch more bits that you've downloaded off of a server. How are these bits any different from any other bits?
(Is there a missing href in the story?)
Porting is what humans do.
Google Web Toolkit contains a Java to Javascript compiler. It is the automated translation by a program of one computer language into another, just like translating C to assembly language, or assembly language to machine code, or Java to Java Bytecode, or Java Bytecode to machine code. All of the programs which do those translations are "compilers"[0]. A program which takes Java and spits out Javascript is no different. It's just another compiler, albeit with a very unusual target.
[0] Although not all are necessarily called compilers all the time. The most common type of Java Bytecode to machine language compiler, because of when it is run, is often referred to as a "JITter", which is simply short for "Just In Time Compiler". And assembly language to machine code compilers are often called "assemblers", but again that does not stop them from *being* a compiler.
"compilers translate a higher-level language into a lower-level one"
Not always.
Actually, I'm fuzzy on which of Java or Javascript would be considered higher- or lower-level. It's not clear-cut, and could probably be considered more of a "sideways" shift than "downwards".
That'd be fixed point, not floating point.
If Microsoft choose to not implement HTML5 + CSS3 for 10 years, then HTML5 + CSS3 is, to all intents and purposes, 10 years out.
There's no point in using the parts of CSS2 that IE6/7 doesn't support on your web page, because the number of people using those browsers is so big and un-ignorable that you have to find a way to do what you want using the bits that it does support. But once you've done that, what you've done generally looks correct in modern browsers too. Using the bits of CSS that IE doesn't support is therefore mostly redundant and will add to your maintenance headaches. Hence, as a developer, you're generally stuck with whatever IE chooses to support.
And given that releases are roughly every 3 months, it exceeds 10,000 commits per minor release.
Source
Being a square, boring, mold-conformer is your idea of "living it up"? Well, if that's what "living it up" is, I, for one, don't really see the appeal. If it's what you want then I hope you have fun with it. Just don't assume it's necessarily for everyone.
"So in other words, using your examples, a jew, woman, homosexual, or disabled person has more rights than a white male."
Where the fuck did the GP write that? He didn't explicitly say "if they refuse to sell to white males, that's illegal" but it's implied, just like "if they refuse to sell to black people, that's illegal" is also implied, even though he didn't explicitly say that either. Or are you calling him a racist?
Are you really so fucking dumb you can't tell the difference between a list containing a few random examples, and an authoritative complete enumeration?
We do all have equal rights. The point is that the "somebody" in "if they refuse to sell to somebody who doesn't want to prove that they came by it legally, then that's well within their rights" is irrespective of age, race, sex, sexual orientation, or any other physical characteristic you care to name. The "somebody" includes whites, blacks, jews, men, women, straights, gays, bis or whatever.
FFS!
1atm? WTF? I think you mean 10^5Pa.
Should be:
It is difficult to get a man to understand something, when his salary depends upon his not understanding it!
-- Upton Sinclair.
"You make it sound so simple, but you ignored the part where the company blacklists you and/or gives bad references, because you refused to accept the cut in pay."
I don't see how that would follow. If they give you option A or option B, and you select one of those options calmly and rationally, without throwing a hissy fit, whining, bad-mouthing people or generally burning bridges, why on earth would they blacklist you?
Would a company blacklist you or give you bad references if you'd just quit because you felt you weren't getting enough out of that job, or felt you could do something more rewarding elsewhere, or just realised you couldn't hack the commute? I mean, unless they're just a bunch of faceless asshats who are vindictive just for the heck of .... oh, yes, I see what you mean now. :-)
Why? You have the exact same deal. You can decide to simply ask your employer to pay you more for reduced hours, and terminate the current deal if they don't agree to it. Nothing wrong with that.
A contract of employment is between two parties, where both sides agree to give the other something, in return for the thing the other side is giving them. You're giving the company your time, effort, expertise, and ultimately the wealth you create, which they value, while they give you money which you can exchange for wealth of types you are unable to create for yourself (food, shelter, clothing, shiny toys), which you value.
If either side feels they are not getting enough of what they value in return for what they're giving, why shouldn't they try to renegotiate for what they perceive to be a better deal? And if no mutually agreeable deal can be come to, why should the parties continue their unfair (at least to one side) exchange?
Meh. I hardly ever use tabs anymore, and have gone back to using multiple windows for all my browsing sessions.
Of course, it helps that I have a vertical taskbar on the left-hand side of my monitor. It's about 200px wide and has plenty of horizontal space to display the first few words of each window title, even with 15-20 windows open. The "Taskbar thumbnails" effect (display scaled down window in giant tooltip, along with whole window title, when hovering the taskbar item) is just a nice cherry on top.
Given that I rarely need any windows to be the full width of my screen, but find myself scrolling vertically a fair amount, the extra vertical space is really useful.
I find it's much easier to go up against a lot of people. I haven't fought just one person for so long... I've been specializing in groups, fighting gangs for local charities... that kind of thing. You see, you use different moves when you're fighting half a dozen people than when you only have to worry about one. So just one person would give me so much trouble.
"You don't even understand the problem."
I guess not.
"The main problem wasn't backwards compatibility. It was that installers routinely (i.e. - always) overwrote critical DLLs with their own versions and that broke other applications."
Surely it wouldn't matter if installers overwrote critical DLLs, if the DLLs maintained backwards compatibility properly like good shared libraries should! (Like good shared libraries *do* on other OSs)
*sigh*
I think our difference is down to the mindset of the system we use. I (and many other Free Software types) expect shared libraries to maintain backwards compatibility because that's just what they do. Any library that doesn't generally gets patches so that it does in future, or people don't bother using that functionality, fork the library, or just write their own version.
You, and many other Windows types, seem to expect libraries to not maintain backwards compatibility, and assume that compatibility problems will need to be worked around. This probably has the effect of making library authors for Windows not care about or pay any attention to backwards compatibility, because they know that any such problems will be worked around anyway. And without caring or paying attention to it, the workarounds *are* then needed just for the system to work.
Each of us is then looking at what the other is doing and claiming it doesn't make sense in our environment. Well, that might be correct, but you aren't doing your thing in my environment, you're doing it in yours. And in that environment, I guess it does make sense. No matter how alien it looks from over here. :-)
"Hence the Ignorance + FUD"
OK, I still don't get this. Ignorance, yes. I've not used Windows in a while, and many parts of it didn't make that much sense when I did.
But, FUD? Fear, Uncertainty and Doubt? I just don't get how this applies to a technical critique, which is what I'm trying to write.
My goal is not to scare users into thinking that Windows might not be the system for them because of vague technical or commercial demons which might come back to haunt them. Rather, it is to point out concrete problems as I see them, and to highlight ways that those problems can be solved, and have been solved (better, as far as I can tell) in other environments.
How am I not accomplishing this?
How on earth were my points FUD? Are you sure you're using the right word?
The main cause of DLL hell is shared library authors who are unable, either through incompetence or lack of discipline, to maintain backwards compatibility between minor updates. (A minor cause is bad installers overwriting newer versions of libraries with older versions. But bad installers can overwrite or just delete other files as well, as well as mess up your system in countless other ways, so that's not really a problem with the DLLs per se.)
The only reason I am aware of (other than the system/user distinction that I already made) that you might want to install multiple copies, or private copies, of shared libraries is to work around this lack of backwards compatibility. If you are aware of another reason, please state it instead of just claiming FUD without anything to back that up.
The .NET article linked to does state that maintaining backwards compatibility "is extremely difficult, if not impossible." That's a direct quote, the crux of the article, and the problem it tries to describe a solution to.
However, while maintaining backwards compatibility might require a modicum of discipline, the claim that is made simply does not stand up under scrutiny. Nearly every single shared library written for every operating system other than Windows *does* manage to maintain backwards compatibility between minor revisions. The libraries I listed, as examples of some of the largest, oldest, most complex, and most widely ported shared libraries I could think of, are concrete evidence of that. Or are you claiming that factual counterexamples to over-stated claims are "FUD"?
Having private copies of shared libraries does cause security problems when security bugs are found in those libraries. Take a look at the fallout from DSA 122-1. The 8 packages which included private copies of the zlib library were still vulnerable to the zlib bug, even after the system-wide shared library had been patched and updated. In contrast, the thousands of packages that use the shared version (1715 packages depend on zlib according to apt on my system) were all fixed as soon as the shared copy was updated. Again, demonstrable fact.
So, the basic problem is that the authors of DLLs for Windows seem incapable of maintaining backwards compatibility between minor revisions, despite the authors of shared libraries for other OSs managing it, and despite the security problems that arise from having multiple/private versions of a library installed which is a workaround that is a direct result of the lack of backwards compatibility.
Making it easier to install multiple private copies of DLLs, or allowing apps to depend on specific "side-by-side" versions of "assemblies", does not fix this problem, it simply papers over it.
Please, if I've overstated anything, claimed anything that's untrue, or made a leap of logic that is unsupportable, point it out in detail instead of just wildly claiming that my whole argument is FUD.
What's the solution to the problem? Demand that your library authors to do a better job than the one they're currently doing and maintain backwards compatibility. Especially if you're paying for your software, and those authors can't maintain the same level of professionalism as hobbyists and other people who are simply giving their code away.
Heh.
You think lack of *Windows-specific* technical knowledge implies "moron"? How amusing. I'll take a knowledge and understanding of basic principles that can be applied in many scenarios over a set of vendor-specific kludges and workarounds any day thanks.
Still, thanks for the insults. Helped me to realise all the more quickly you couldn't counter any of the actual points I was making. Ad Hominem really is the most transparent and least convincing logical fallacy.
Uh, having multiple copies of the same major version of a shared library *is* DLL hell. All shared libraries should exist in either a system-wide folder (e.g. /usr/lib on Unix), or in a per-user folder (e.g. ~/lib). That's why they're called *shared* libraries. Aside from the system/user distinction, you should never have multiple copies of the same major version of a shared library in different folders.
If you've got multiple system-wide copies of the same shared library on one system, that's DLL hell. If a user has more than one copy of the same shared library among their privately-installed applications, that's DLL hell. .NET versioning is a stupid hack that tries to prevent DLL hell appearing to be a problem. It does not prevent DLL hell, it just stops the worst of the symptoms.
Heck, read that article you linked to:
"In the most typical case, one application will install a new version of the shared component that is not backward compatible with the version already on the machine."
Why is it not backwards compatible?
"In practice, writing code that is forever backward compatible is extremely difficult, if not impossible."
Demonstrably false. Any code written for and compiled against glibc 2.0.0 (July 1996) will work fine when used on a system with glibc 2.9.0 (Nov 2008). Ditto code written for/compiled against zlib 1.0.1 (May 96) will work against 1.2.3 (July 2005). Ditto libpng 1.0.0 (March 98) and 1.2.35 (Feb 2009). Ditto Qt 3.0.0 (Oct 2001) and 3.3.8 (Jan 2007). Ditto ... almost every other Unix/Linux library ever written.
"Strong binding means an application or component can bind to a specific version of another component"
Is stupid, because when a security bug in a library is found, you *need* to update the library (or, if you have multiple copies, *all* copies of the library) to fix all the apps that use it. And the new version needs to have an updated version number, so you know that you've got the bug-free version.
Claiming that being able to load different copies of the same DLL because they're in different directories (or because .NET has this "side-by-side" feature) is not DLL hell, is like claiming that your house isn't really burning down because you're wearing a fireproof suit.
Damn, your life must be one empty, boring repetitive loop of nothingness if you think that's a life history. I hope you get to do something interesting soon.
Ah, sorry. You confused me by spuriously talking about browser plugins, instead of the subject of the article and almost every other comment which is about browser extensions. I mistakenly assumed you were talking about extensions like everyone else, only calling them plugins.
My bad.
I'm a Linux user, got introduced to it in around '96 and started using it a fair amount in '99. Never experienced library hell or dependency hell.
When I started, your distro would give you a bare system, and everything else was a download, "gzip -cd source.tar.gz | tar -xf -", "./configure" and "make install" away.
If you were missing a dependency, or had a version that was too old for the software you wanted to install, configure would stop and tell you which library was missing. At which point you simply downloaded the latest version of that required library, installed it first with the same procedure, and then went back to the configure step of the software you were trying to install. OK, it's not quite a point-and-click away, but it's not excruciatingly hard either.
Never experienced library hell either. Seems that library writers for Unix/Linux know how, and have the discipline, to keep their libraries binary backwards compatible between minor revisions. I've only had installing a new minor version of a library break stuff a handful of times in 10 years, and nearly every single time it was by accident and a new version was released within a couple of days fixing it. If lib writers need to break binary compatibility (say, every 5 years or so), they generally create a new major version of the library that can be installed side-by-side with the old version (e.g. libqt-mt.so.3, libQtCore.so.4) and the two never interfere.
Nowadays I use Debian, and the combination of Debian's repository and apt-get (first released in 1998/1999, now has equivalents in nearly all modern distros) has also meant that I've not experienced any form of library or dependency hell in 5 the 5 years I've been using that.
As I understand it, Windows/.NET are the only platforms to speak of which suffer from these problems.
Also, the GP does not talk about viruses, but "the virus protection industry", which is almost entirely Windows-based (apart from scanning tools that run on other systems to check for Windows viruses) and is basically a result of the insecurity of Windows and other MS products. (See also Apache vs. IIS market share/vulnerabilities/crack rates before you even think about bringing up the "market leader gets most of the attention" excuse)
OK, it's been a while since I've been around Windows (thank God) but I follow as much tech stuff as I can and really hadn't heard anything about this before.
Could you post links to 3 other bits of software which, as well as the "normal" software they install (whatever that is) also install Firefox plugins?
I'm not a developer, and I'm running KDE 4.2 RC. I have a clock on my panel showing the date and time. I do not see this bug.
From How to report bugs effectively:
The whole thing is worth reading, really.
Now, go file a damn bug, with a screenshot, and help make KDE rock!
"You don't own any application you use, unless you wrote it. Even under the GPL, you are permitted to modify and copy it, but the original copyright belongs to the author of that product. If you just -owned- the software you got, but could copy it freely,"
I must respectfully point out that you're talking bollocks.
If you have a book, you own that book. You own that copy of the book. You own that copy of the copyrighted material which is printed therein.
No, you do not own the copyright to the book. No-one has ever suggested that by owning a copy of a book that you own the copyright to that book.
If you have a copy of a piece of software which you have legally obtained, you own that copy of that software. Like a book, that does not mean that you own the copyright, or that you can exercise the rights of the copyright holder. It just means you own that copy. You are not allowed to make copies yourself, except where permitted by fair use, other relevant statutes, or a license from the copyright holder.
But, like owning a copy of a book, you own your copy of a piece of software.
You are, of course, free to give up some of the privileges of such ownership. So if you voluntarily agree to a binding EULA contract in which you consent to never engage in some otherwise-legal activities, and possibly give up your ownership of the copy of the software you have, that is your choice.
Fortunately, Free Software does not ask you to enter into any such agreement.
The correct term for that is Rubber-hose cryptanalysis.
But I thought we were talking about how they were going to hack people's PCs with consent of the courts/warrants. Or are the courts now advocating torture as well?