You really do not want to go there.... If PolicyKit does indeed match the claims of the patent, then all it means is that it violates the patent.
I don't know...
I'm sure the PK devs obviously "don't want to go there", but if this turns out to be another example of multiple independent 'inventions' of the same idea (a not uncommon event in the annals of scientific discovery), then to me that would point to a problem with the patent system just as much as finding prior art for a 'bad' patent would.
MO, of course, but what irks me about this system is that winner is not determined by any qualities of their implementation of the idea (what implementation?), or who had the idea first (no way to always determine that) but simply which one was rich enough to afford the patent attorney who got their application rammed through the USPTO first...
Law is the programming language for the system of society. The problem is,...
My gut tells me the problem with your analogy (no offense, it is a nice effort; at least you didn't use cars in it), is more profound:
The vast majority of users of a programming language view it as a *tool*, where normally (outside of obfuscation contests) one of its most valuable traits is considered to be *clarity*.
The vast majority of users of Law's legalese language view it as a *weapon*, whose ability to completely avoid clarity altogether is considered to be its most valuable trait. Indeed, it would be utterly useless to its practitioners, considering the purpose for which they use it, if it could *not* do this.
Unless a company can maintain atleast some sort of competitive advantage against the rivals, guess what'll happen to the coders.
If companies didn't have to worry about building up defensive patent portfolio's they could let their coders spend more time... well... coding. And maybe, just maybe, they could even pay their coders more, since they won't need to waste millions on patent lawyer fees.
Without patents, a software company's 'competitive advantage' will swing back to the *coders* themselves, what it arguably should have been all along, if not for the lawyers sidetracking everyone with their patent violations, copyright violations, trademark infringements, EULAs, DRM, etc, etc, but certainly the software patent madness is the worst of their monstrous creations. All of that stuff is about control: locking someone in, or locking someone out, its not about competition.
Its only the patent lawyers who want you to think that gaming the legal system qualifies as a valid (as in creative/useful to society at large) way of gaining 'competitive advantage'. After all, software patents are inherently anti-competitive by their very nature.
The rest of us would prefer this game a lot better if it were left up to the coders themselves, by letting the best *coder*, rather than the best lawyer, win...
In other words, *you* didn't have a problem avoiding GPL'd code, which was the GGP's point.
it belongs to downstream distributors who find that the GPL has snuck in via the actions of another package maintainer.
Except that the 'downstream distributor' and the 'package maintainer' in this case are one in the same, the Debian distro and a Debian package maintainer. Your example is not one of GPL'd code mixing with non-GPL'd code by 'accident', your example is merely one of someone at Debian recognizing that *their* distro had a particular problem here and then *avoiding* that outcome... which is precisely what you say is supposed to be hard to avoid.... wait, what?
And if the GPL'd code was only there due to 'the actions of another package maintainer' then how can you honestly describe it as 'sneaking in'? The GPL'd code didn't slither in on its own, it was put there by a human being, for pete's sake. Software, like any other inanimate object, doesn't 'sneak'.
If Debian's inclusion of the GPL'd sub-parts in that LGPL'd lib was unintentional, then its presumably already been fixed, if it was not, then that is their (Debian's) choice. In any event, you haven't shown how *you* were inconvenienced in any way by the GPL, or how *you* found it difficult to avoid GPL'd code getting into *your* project (since avoiding it was your intent), which was what the GGP was speaking to.
Your example only raises questions in my mind about what that Debian package maintainer was thinking of (did he know of the GPL'd sub-parts?), but it doesn't say anything to me about the GPL being 'hard to avoid' for non-GPL'd code, since in this case, it was avoided.
So, how is licensing under the GPL going to stop dishonest pricks from being dishonest pricks?
It won't, for the same reason that a law can't stop a law-breaker. A law (or a software license) is just words on paper.
Note however, that in the GP's example of the legacy of 'weakly' licensed open source code in MS's closed-source Windows, MS wasn't being dishonest. The license of the code *allowed* them to do what they did. They could not legally do the same with GPL code however.
But the point is that most humans regard plagiarism as a much more serious offense than mere copyright infringement.
My gut says that most humans don't even understand the legal distinctions between those 2 things...
But lying and claiming credit for something that somebody else did, is a pretty serious sin.
In most parts of the world, 'a serious sin' is not the same thing as 'a serious crime' (depends on the actual 'sin' of course). Indeed, in some places, for some 'sins', they aren't even *illegal*.
I understand the point you're after, and in a saner world, morality would matter more than it does, but alas we don't live in a sane world, not by a long shot. In our case, what only matters is the legality of the act, and what is the punishment if said act was not legal. People willing to break the law are often far beyond concern over sinful behavior: the only thing they fear is a 'big stick'.
Unless of course you're on/. arguing about the finer points of FOSS software licenses, in which case ridicule seems to be the first rhetorical weapon-of-mass-insult that folks around here reach for...
How can anybody sue a closed source software company for GPL violation?
Ever go bit-diving in a binary? You'd be amazed at what can show up there, and thats especially true if the violation was accidental rather than willful, since in those cases no one bothered or thought to go through and scrub any incriminating text strings (like author's names!) or obfuscate uncommon function/variable names which can also appear in the binary (but are never printed or otherwise shown to the program's user).
So proper copyright use in both cases gives more to the public good.
True, but the GP's point wasn't about both involving copyright. Note the part in the GGP's rant, that I've bolded:
Next thread, something about RIAA, same people demanding...
It is, from my experience at least, not the same people doing both as the GGP implies, because as the GP points out, anyone using *any* software license, not just the GPL, or for that matter, anyone who just knows enough about them to argue over their differences (as we often seem to do here), has already demonstrated more respect for copyright law than somebody who simply thinks that all music should be free for them to download...
but I have never used BSD nor am I familiar with any licensing associated with it.
You probably *have* used BSD-licensed code, you just don't know that you did. BSD-licensed code is 'more free' in that it can even find itself incorporated into proprietary products whose users never know about it. Some see this particular aspect of being 'more free' as a good thing, but others don't, hence the existence & popularity of the GPL.
But the codec to be used isn't in the spec since the main players couldn't agree on a default/baseline codec, so the actual usefulness of HTML5 is unfortunately still questionable, and without the ability to 'just work' for nearly everyone, everywhere, its chances of actually 'killing' Flash are, for the moment at least, nonexistent.
banishment from iPhone
Outside of the US the iphone is not significant (Apple is only a bit player in the global cell phone market). I doubt that they (nor anyone else) who is operating on a global scale is really that concerned about the iphone.
Adobe is fighting an uphill battle to stay in the mix.
I don't have a horse in this race, but I'd say its HTML5 that has the uphill battle, not Adobe. Flash is here now, its real & ubiquitous, whereas HTML5 is, for the moment, just a promise... and without a common/default codec, it will probably always just be a promise/wish/hope, rather than a reality.
Theo is in charge of a BSD-based kernel that is only concerned with security, while Linus is in charge of a kernel that has to accommodate a much wider audience (like people who want to run Wine), and, of course, since both of them also have largish egos, they've both managed to say some silly things about each other's kernel...
Basically, unless you're already a Linus or Theo fanboy, their 'bickering' is not that important.:)
The immediate parent post was modded flamebait, but the point is nevertheless valid.
No it isn't.
But a Linux distribution should be professional, it should be of the highest quality, regardless of the price tag.
If Canonical had the kind of money that an MS or Apple has to spend on QA testing before release, then you and the GP would have a point. Except they don't.
When it comes to Linux specifically and Open Source in general, Free as in beer does not mean cheap as in crappy.
No it doesn't, but the 'free' part of this equation does mean users of FOSS (or at least early adopters) have to help with the QA testing since the devs simply can't do it all themselves.
Seriously, do you have any idea how hard QA is? Or how much money MS/Apple spends on it?
(Sorry, mandatory compiling time joke for each post mentioning Gentoo).
For *each* post? Seriously, why bother?
Most actual Gentoo users know that the compile-time meme hasn't made any sense in about a decade or so (because many of them can't even remember the last time they did a full install/rebuild - never mind that modern hardware turns this job from one of days to just hours), even if it is occasionally funny, but if you're doing this *every* time Gentoo is mentioned, even in passing as it was done here, then... your apology is not accepted.:)
I knew there was a good reason why I friend'ed you.:)
Well said, especially about the KDE 3-to-4 transition, that stuff was some of the most pointless ranting that I've ever seen. None of the ranters ever bothered to even try to understand *why* there was a (need for) KDE4...
You're not the only one who has seen this, and wondered about it. You can add osnews.com to your list of places where it occurs as well. And in this case, its even more interesting: I mean, if you really hate Linux, then why would you even be reading anything posted to 'linux.slashdot.org'? Go figure.
Now, this probably only means the client (the underlying protocol will probably be handled by a binary-only library), but even if that's the case, it seems like there is still reason to celebrate
The source is not open, until I can build and use it on FreeBSD/amd64 or some other "exotic" platform like that...
Interestingly, the oft-criticized Java has always been more "open-sourced" (even before going GPL), than what the excited write-up is preparing to "celebrate"... Must all be about managing expectations...
Right, and the reason for Java's more 'openness' was because its equivalent of a protocol (its bytecode specification) was just as 'open' as the rest of it.
This is the Internet we're talking about, open protocols are *FAR* more important than open clients of closed protocols.
Have we already forgotten the history MS & SAMBA? Hello?
Except for Skype users, who are already locked-in to the protocol, there really *is* nothing to see here, so folks, lets just move along...
It is available actually (grub-9999, e.g. compiled from grub2's source repo), along with latest KDE4 that was mentioned earlier, they just aren't marked 'stable' in Gentoo's main repo.
I think that things not moving quickly to 'stable' on Gentoo has to a lot to do with, aside from the smaller user-base, my suspicion that most Gentoo users are running the Gentoo equivalent of 'unstable' by default.
Gentoo doesn't have fixed releases, its a continuously updated source-based distro, so 'stable' and 'unstable' have different meanings for it anyway, and once you add in 'overlays' (external repos, typically bleeding-edge stuff), which are commonly used in Gentoo, then comparisons to the popular binary-based, versioned distros become almost meaningless.
Anything GUI-related has NOTHING to do with hardware support.
True. Neither I nor anyone before me in this subthread (including you) mentioned 'GUI'. Thats *why* I said your last post didn't make any sense.
Xorg only USES the hardware support offered by the underlying OS.
The underlying OS doesn't go far enough in the case of video cards for that to be true though. Effectively, all the Xorg drivers get from the kernel is read/write access to the hardware's IO registers and memory (framebuffer), but the kernel itself doesn't provide all of the 'support' needed to actually drive the specific device. Thats why Xorg has *hardware-specific* drivers of its own.
You are not separating hardware drivers from software implementation.
What do you mean by 'software implementation' here? Implementation of *what*?
You aren't realizing that for graphics, at least, the 'hardware driver' is split between the kernel and Xorg. Seriously, just look at some of the Xorg driver code. There is a lot of hardware register accessing being done there: those Xorg drivers are *talking* directly to the *hardware*, thus, they *are* hardware drivers.
See the past debates with Linus about implementing graphics drivers entirely within the kernel. This has never been allowed, which is why it is *not* all in the kernel.
Anything GUI-related has NOTHING to do with hardware support.
That makes no sense whatsoever.
What's driving my monitor right now is a video card, e.g. hardware, and whats driving it is Xorg, not the kernel.
Wait, are you thinking of the proprietary kernel modules from ATI/NV? They are only part of the process, but in any event, I use the open ATI driver currently in development: the only thing happening in the kernel is the initial graphics-mode setting, once the resolution is set and I/O buffers to the card are initialized, Xorg does *everything* else.
The HOWTOs are overblown - in a proper distro you shouldn't need them in the first place
Gentoo isn't a 'proper distro' in the sense that its a source-based distro, completely unlike a binary-based one thats meant for the masses.
(I also disagree with the GP as you do - Gentoo is not meant for Linux newbies)
and even when you do, nothing stops you from using a Gentoo HOWTO on, say, Debian
Wait, now you're saying Debian isn't a 'proper distro' either? Careful...:)
I've never got my system broken by updates either (even though I used "unstable") - in part because apt is quite a bit smarter about dependency breakages than Portage ever was, and I haven't ever seen any other kind of break in Debian repositories.
Ironically, its was breakages in Debian that caused me to try Gentoo, but that was a long time ago.
Crucial difference between a source-based distro and a binary one: on a source-based distro if a compile breaks, it doesn't break your system, because nothing on your system has changed. And the corollary: if the compile succeeds then you are *guaranteed* that that package has all its crucial dependencies met.
Slashdot Mirror
You really do not want to go there. ... If PolicyKit does indeed match the claims of the patent, then all it means is that it violates the patent.
I don't know...
I'm sure the PK devs obviously "don't want to go there", but if this turns out to be another example of multiple independent 'inventions' of the same idea (a not uncommon event in the annals of scientific discovery), then to me that would point to a problem with the patent system just as much as finding prior art for a 'bad' patent would.
MO, of course, but what irks me about this system is that winner is not determined by any qualities of their implementation of the idea (what implementation?), or who had the idea first (no way to always determine that) but simply which one was rich enough to afford the patent attorney who got their application rammed through the USPTO first...
Just sayin'
I want to have whatever that earlier mod was smoking...
Seriously guys, agree or disagree, this post is nowhere *near* to 'Troll' status.
No, we just need to disable Microsoft's access to run /usr/sbin/grantpatent as root in /etc/sudoers. ;)
I have a simpler, quicker solution that doesn't even require messing with any conf files:
'rm -f /usr/sbin/grantpatent'
Or find whichever idiotic package that installed that crazy-ass binary, and nuke *it* from orbit...
Law is the programming language for the system of society. The problem is, ...
My gut tells me the problem with your analogy (no offense, it is a nice effort; at least you didn't use cars in it), is more profound:
The vast majority of users of a programming language view it as a *tool*, where normally (outside of obfuscation contests) one of its most valuable traits is considered to be *clarity*.
The vast majority of users of Law's legalese language view it as a *weapon*, whose ability to completely avoid clarity altogether is considered to be its most valuable trait. Indeed, it would be utterly useless to its practitioners, considering the purpose for which they use it, if it could *not* do this.
fluent legalese
Wait a sec, I think we have a new candidate for best oxymoron since 'Military Intelligence'.
I like it!
Unless a company can maintain atleast some sort of competitive advantage against the rivals, guess what'll happen to the coders.
If companies didn't have to worry about building up defensive patent portfolio's they could let their coders spend more time... well... coding. And maybe, just maybe, they could even pay their coders more, since they won't need to waste millions on patent lawyer fees.
Without patents, a software company's 'competitive advantage' will swing back to the *coders* themselves, what it arguably should have been all along, if not for the lawyers sidetracking everyone with their patent violations, copyright violations, trademark infringements, EULAs, DRM, etc, etc, but certainly the software patent madness is the worst of their monstrous creations. All of that stuff is about control: locking someone in, or locking someone out, its not about competition.
Its only the patent lawyers who want you to think that gaming the legal system qualifies as a valid (as in creative/useful to society at large) way of gaining 'competitive advantage'. After all, software patents are inherently anti-competitive by their very nature.
The rest of us would prefer this game a lot better if it were left up to the coders themselves, by letting the best *coder*, rather than the best lawyer, win...
The problem is not mine,
In other words, *you* didn't have a problem avoiding GPL'd code, which was the GGP's point.
it belongs to downstream distributors who find that the GPL has snuck in via the actions of another package maintainer.
Except that the 'downstream distributor' and the 'package maintainer' in this case are one in the same, the Debian distro and a Debian package maintainer. Your example is not one of GPL'd code mixing with non-GPL'd code by 'accident', your example is merely one of someone at Debian recognizing that *their* distro had a particular problem here and then *avoiding* that outcome... which is precisely what you say is supposed to be hard to avoid.... wait, what?
And if the GPL'd code was only there due to 'the actions of another package maintainer' then how can you honestly describe it as 'sneaking in'? The GPL'd code didn't slither in on its own, it was put there by a human being, for pete's sake. Software, like any other inanimate object, doesn't 'sneak'.
If Debian's inclusion of the GPL'd sub-parts in that LGPL'd lib was unintentional, then its presumably already been fixed, if it was not, then that is their (Debian's) choice. In any event, you haven't shown how *you* were inconvenienced in any way by the GPL, or how *you* found it difficult to avoid GPL'd code getting into *your* project (since avoiding it was your intent), which was what the GGP was speaking to.
Your example only raises questions in my mind about what that Debian package maintainer was thinking of (did he know of the GPL'd sub-parts?), but it doesn't say anything to me about the GPL being 'hard to avoid' for non-GPL'd code, since in this case, it was avoided.
So, how is licensing under the GPL going to stop dishonest pricks from being dishonest pricks?
It won't, for the same reason that a law can't stop a law-breaker. A law (or a software license) is just words on paper.
Note however, that in the GP's example of the legacy of 'weakly' licensed open source code in MS's closed-source Windows, MS wasn't being dishonest. The license of the code *allowed* them to do what they did. They could not legally do the same with GPL code however.
But the point is that most humans regard plagiarism as a much more serious offense than mere copyright infringement.
My gut says that most humans don't even understand the legal distinctions between those 2 things...
But lying and claiming credit for something that somebody else did, is a pretty serious sin.
In most parts of the world, 'a serious sin' is not the same thing as 'a serious crime' (depends on the actual 'sin' of course). Indeed, in some places, for some 'sins', they aren't even *illegal*.
I understand the point you're after, and in a saner world, morality would matter more than it does, but alas we don't live in a sane world, not by a long shot. In our case, what only matters is the legality of the act, and what is the punishment if said act was not legal. People willing to break the law are often far beyond concern over sinful behavior: the only thing they fear is a 'big stick'.
Ridiculing folks gets folks nowhere.
Unless of course you're on /. arguing about the finer points of FOSS software licenses, in which case ridicule seems to be the first rhetorical weapon-of-mass-insult that folks around here reach for...
*sigh*
How can anybody sue a closed source software company for GPL violation?
Ever go bit-diving in a binary? You'd be amazed at what can show up there, and thats especially true if the violation was accidental rather than willful, since in those cases no one bothered or thought to go through and scrub any incriminating text strings (like author's names!) or obfuscate uncommon function/variable names which can also appear in the binary (but are never printed or otherwise shown to the program's user).
So proper copyright use in both cases gives more to the public good.
True, but the GP's point wasn't about both involving copyright. Note the part in the GGP's rant, that I've bolded:
Next thread, something about RIAA, same people demanding ...
It is, from my experience at least, not the same people doing both as the GGP implies, because as the GP points out, anyone using *any* software license, not just the GPL, or for that matter, anyone who just knows enough about them to argue over their differences (as we often seem to do here), has already demonstrated more respect for copyright law than somebody who simply thinks that all music should be free for them to download...
but I have never used BSD nor am I familiar with any licensing associated with it.
You probably *have* used BSD-licensed code, you just don't know that you did. BSD-licensed code is 'more free' in that it can even find itself incorporated into proprietary products whose users never know about it. Some see this particular aspect of being 'more free' as a good thing, but others don't, hence the existence & popularity of the GPL.
Pick whichever you prefer; choice is good.
With video specs in HTML5
But the codec to be used isn't in the spec since the main players couldn't agree on a default/baseline codec, so the actual usefulness of HTML5 is unfortunately still questionable, and without the ability to 'just work' for nearly everyone, everywhere, its chances of actually 'killing' Flash are, for the moment at least, nonexistent.
banishment from iPhone
Outside of the US the iphone is not significant (Apple is only a bit player in the global cell phone market). I doubt that they (nor anyone else) who is operating on a global scale is really that concerned about the iphone.
Adobe is fighting an uphill battle to stay in the mix.
I don't have a horse in this race, but I'd say its HTML5 that has the uphill battle, not Adobe. Flash is here now, its real & ubiquitous, whereas HTML5 is, for the moment, just a promise... and without a common/default codec, it will probably always just be a promise/wish/hope, rather than a reality.
What's the beef between Linus and Theo?
Theo is in charge of a BSD-based kernel that is only concerned with security, while Linus is in charge of a kernel that has to accommodate a much wider audience (like people who want to run Wine), and, of course, since both of them also have largish egos, they've both managed to say some silly things about each other's kernel...
Basically, unless you're already a Linus or Theo fanboy, their 'bickering' is not that important. :)
The immediate parent post was modded flamebait, but the point is nevertheless valid.
No it isn't.
But a Linux distribution should be professional, it should be of the highest quality, regardless of the price tag.
If Canonical had the kind of money that an MS or Apple has to spend on QA testing before release, then you and the GP would have a point. Except they don't.
When it comes to Linux specifically and Open Source in general, Free as in beer does not mean cheap as in crappy.
No it doesn't, but the 'free' part of this equation does mean users of FOSS (or at least early adopters) have to help with the QA testing since the devs simply can't do it all themselves.
Seriously, do you have any idea how hard QA is? Or how much money MS/Apple spends on it?
The Mono/Moonlight default is a dangerous game. It's going to burn them in the end.
Does Ubuntu *require* Mono? I knew that Novell's distro did (no surprise), but Ubuntu too?
I thought it was still possible to install a 'limited' Gnome that didn't require Mono. Not true?
(Obviously I'm neither a Gnome user nor an Ubuntu user...)
(Sorry, mandatory compiling time joke for each post mentioning Gentoo).
For *each* post? Seriously, why bother?
Most actual Gentoo users know that the compile-time meme hasn't made any sense in about a decade or so (because many of them can't even remember the last time they did a full install/rebuild - never mind that modern hardware turns this job from one of days to just hours), even if it is occasionally funny, but if you're doing this *every* time Gentoo is mentioned, even in passing as it was done here, then... your apology is not accepted. :)
I knew there was a good reason why I friend'ed you. :)
Well said, especially about the KDE 3-to-4 transition, that stuff was some of the most pointless ranting that I've ever seen. None of the ranters ever bothered to even try to understand *why* there was a (need for) KDE4...
Depends who is saying "most."
Nevermind 'most', the more important question is who did you mean by 'we' in your OP?
When Linux vs. other OS's comes up on /. the typical answer is that Linux is ready
Really? All the jokes about Linux's 'Year on the Desktop' strongly suggest otherwise, nevermind the 900+ posts here about TFA...
Are you reading the same /. that I am?
Leap of logic, indeed.
You're not the only one who has seen this, and wondered about it. You can add osnews.com to your list of places where it occurs as well. And in this case, its even more interesting: I mean, if you really hate Linux, then why would you even be reading anything posted to 'linux.slashdot.org'? Go figure.
The source is not open, until I can build and use it on FreeBSD/amd64 or some other "exotic" platform like that...
Interestingly, the oft-criticized Java has always been more "open-sourced" (even before going GPL), than what the excited write-up is preparing to "celebrate"... Must all be about managing expectations...
Right, and the reason for Java's more 'openness' was because its equivalent of a protocol (its bytecode specification) was just as 'open' as the rest of it.
This is the Internet we're talking about, open protocols are *FAR* more important than open clients of closed protocols.
Have we already forgotten the history MS & SAMBA? Hello?
Except for Skype users, who are already locked-in to the protocol, there really *is* nothing to see here, so folks, lets just move along...
It's not out on Gentoo?
It is available actually (grub-9999, e.g. compiled from grub2's source repo), along with latest KDE4 that was mentioned earlier, they just aren't marked 'stable' in Gentoo's main repo.
I think that things not moving quickly to 'stable' on Gentoo has to a lot to do with, aside from the smaller user-base, my suspicion that most Gentoo users are running the Gentoo equivalent of 'unstable' by default.
Gentoo doesn't have fixed releases, its a continuously updated source-based distro, so 'stable' and 'unstable' have different meanings for it anyway, and once you add in 'overlays' (external repos, typically bleeding-edge stuff), which are commonly used in Gentoo, then comparisons to the popular binary-based, versioned distros become almost meaningless.
Anything GUI-related has NOTHING to do with hardware support.
True. Neither I nor anyone before me in this subthread (including you) mentioned 'GUI'. Thats *why* I said your last post didn't make any sense.
Xorg only USES the hardware support offered by the underlying OS.
The underlying OS doesn't go far enough in the case of video cards for that to be true though. Effectively, all the Xorg drivers get from the kernel is read/write access to the hardware's IO registers and memory (framebuffer), but the kernel itself doesn't provide all of the 'support' needed to actually drive the specific device. Thats why Xorg has *hardware-specific* drivers of its own.
You are not separating hardware drivers from software implementation.
What do you mean by 'software implementation' here? Implementation of *what*?
You aren't realizing that for graphics, at least, the 'hardware driver' is split between the kernel and Xorg. Seriously, just look at some of the Xorg driver code. There is a lot of hardware register accessing being done there: those Xorg drivers are *talking* directly to the *hardware*, thus, they *are* hardware drivers.
See the past debates with Linus about implementing graphics drivers entirely within the kernel. This has never been allowed, which is why it is *not* all in the kernel.
Anything GUI-related has NOTHING to do with hardware support.
That makes no sense whatsoever.
What's driving my monitor right now is a video card, e.g. hardware, and whats driving it is Xorg, not the kernel.
Wait, are you thinking of the proprietary kernel modules from ATI/NV? They are only part of the process, but in any event, I use the open ATI driver currently in development: the only thing happening in the kernel is the initial graphics-mode setting, once the resolution is set and I/O buffers to the card are initialized, Xorg does *everything* else.
Hardware support is not all in the kernel.
The HOWTOs are overblown - in a proper distro you shouldn't need them in the first place
Gentoo isn't a 'proper distro' in the sense that its a source-based distro, completely unlike a binary-based one thats meant for the masses.
(I also disagree with the GP as you do - Gentoo is not meant for Linux newbies)
and even when you do, nothing stops you from using a Gentoo HOWTO on, say, Debian
Wait, now you're saying Debian isn't a 'proper distro' either? Careful... :)
I've never got my system broken by updates either (even though I used "unstable") - in part because apt is quite a bit smarter about dependency breakages than Portage ever was, and I haven't ever seen any other kind of break in Debian repositories.
Ironically, its was breakages in Debian that caused me to try Gentoo, but that was a long time ago.
Crucial difference between a source-based distro and a binary one: on a source-based distro if a compile breaks, it doesn't break your system, because nothing on your system has changed. And the corollary: if the compile succeeds then you are *guaranteed* that that package has all its crucial dependencies met.