They have to ensure this "*never*"ness is 100% flawless on all of the millions of $80 system boards it is going to be implemented on. But even with the best efforts they won't achieve that and they aren't going to get the best effort from all of the mobo manufacturers. Ensuring this crap is dead perfect is costly for one and accidentally on purpose screwing it up will prove profitable as DVD player manufacturers already well know.
If using "cult" negatively, I'd say the best two criteria are these:
1. The group is the member's sole source of support and validation. Any other form of emotional or intellectual support especially from skeptical family members is strongly discouraged and/or punished.
2. The group demands unreasonable amounts of the member's time and/or money. "Unreasonable" being defined as an amount that makes it very difficult to fulfill obligations outside the group like paying bills, attending family functions, or even being able to afford decent food.
Clearly, one can belong to a religious group without demands of that magnitude being regularly expected.
Most of the minor distros are specialty items like Knoppix or toolkits like Trinity Rescue CD. Having more of such isn't going to hurt anything. Besides, projects like this are a good way of trying out radical ideas without breaking anything. And I suspect the answer to "not teaming up" is that it seems that many developers would rather be Chiefs than common braves.
It's really a shame for F/OSS that, time and time again, there is such a huge duplication of effort and half-assed half-finished projects lying around in the junkyard of the Open Source cemetery.
And once again someone falls prey to a common misconception: F/OSS is not a monolith. If these guys didn't have the option of having their own sandbox to play in then what makes you think they'd be compelled to play in someone else's? The way this will more than likely shakeout is that fifty or so people will use this for awhile. Maybe it'll be a bit more popular if the primary devs have more stature than I'm giving them credit for.
These guys will get to have their fun and most everybody else will use an established distro. And that isn't to say good won't come of it. If they have good ideas, the bigger distros might adopt them. If they have REALLY good ideas they may supplant Gentoo among that crowd of people. Bugfixes may also go to upstream projects.
I know this is weird idea to someone accustomed to being served what they think they want from proprietary software houses but this is nothing but an exercise of freedom. Others are free to use what they make or not. What would you propose? Some sort of law saying that henceforth no one may attempt to start a BSD or Linux distribution?
The F/OSS world operates on a form of street-cred. These guys will either get it or not. It won't cause any sort of actual problem either way.
I'm not really seeing your argument. The included Linux is more of a diagnostic and rescue tool. Most people who have one of these computers will never even see the Linux desktop on the things. Only techs and Linux gearheads will bother with it.
And we aren't saying "bundling is teh Evel!". We're saying that a monopoly abusing bundling to push out competition is. The Linux on these things will not be the front and center user experience for most people who use these boards. It isn't in the same ballpark as some of the things MS has done. It isn't even the same sport.
It's best to think of the Xandros as a bastardized Debian Etch. Chuck etch apt deb lines into the sources.list. You can safely do a "apt-get upgrade" but DON'T do an "apt-get dist-upgrade". You can install any Etch package or package built for Etch as long as it
1. Doesn't replace or conflict any package with "asus" or "xandros" in the name. 2. Doesn't conflict or replace kdelibs or libqt-mt. 3. Doesn't replace or conflict the kernel or supplied drivers.
Those caveats don't get in the way of much at all. My EEE has OpenOffice 2.4 from Sun's site (there is a guide you need to follow on this one so the Easy Mode icons still work), a recent Adobe reader, mplayer from Sid built on an Etch machine and installed, taglib1.5 so AlbumArt works correctly in "Music Manager (Amarok)", dterm for configuring serial devices, and a few other things that escape my mind. The biggie is being able to build Sid packages on this Etch buildhost I keep around. That is the keyhole that lets me install anything I want without losing the good job Asus did of making wireless, suspend, video, sound, and the Desktop Just Work. Once FireFox 3 goes out of beta, I'll chuck that on too.
The one improvement that is somewhat difficult is removing UnionFS. If you do things like replace OpenOffice, you want that because you otherwise lose space for every large package that is "removed" but invisibly occupying space.
This manual doesn't sound like Grade A ultraweirdness on the order of OT III. Instead we have a rather pedestrian model of conduct and procedure from a church. Just let this one die already.
You're partially missing my point. On a product that will make 5 or 6 bucks of profit per unit sold, MS causes pain to the manufacturer at any price point. As the "race to the bottom" continues and these things become commodities like Casio watches, a bit of superficial familiarity isn't going to matter that much. MS will be put in a position where they don't dictate the price or featureset. If they do try to dictate the price, Linux starts squeezing right back in because being forced off some of these devices for a year or three won't hurt it the way it would hurt a corporate product. Sure, people may want familiar XP from Mama Microsoft. But that isn't a powerful enough effect to let MS dictate terms and prices in a cheap commodity market.
But Microsoft can, and likely will, continue to cut OEM licensing costs.
Indeed they can but that kicks their current business model in the balls. MS is accustomed to receiving a 20-100 dollar premium on every PC sold. They can temporarily push Linux out by dropping the price to $5 or $10 but this tactic will not suffice to entirely drive out Linux. If we're talking about $100 PCs being sold in blister packs at Wal-Mart, even $5 or $10 represents a significant raw materials cost.
Linux based systems will have no artificial limitations mandated for inexpensive new PC categories but MS will insist on all kinds. They'll either cripple the "UMPC Editions" or mandate what can and can't be built into hardware eligible for those licenses. MS themselves can create the situation where it is the cheap Linux PC that does more than the not as cheap Windows PC.
The funny thing is that I used to be that way and then crossed a point about four years ago where I could generally zero in and fix the problem. That did a lot to endear Linux and the other free Unixes to me. It is generally possible to recover from screw ups, even really bad ones, as long as the hardware is still OK.
These days I have a USB hard drive that contains:
A list of all installed packages. A tarred up copy of the/etc directory Entire home directory. (which contains all installers for anything I didn't get from repositories as well)
Even if my hard drive were to completely eat it, those three things would get me my exact working environment back in an hour or two. Most of that time would be spent downloading the installer and packages. But, I have yet to have to resort to that. I've picked up enough of the way a Linux system hangs together that I can just fix the damn thing if it breaks usually.
If you do things like replacing the supplied Acrobat and OpenOffice then yes, freeing space allows for latitude for customization. Otherwise, those two changes will cost about 400MB of space and that isn't all that I have done. When you only have about 1.6 GB free, that is significant and indeed cuts "latitude" for making changes. As it is, I have a software load on the unit that I like and about 1.3 GB free of user space. If I hadn't removed Unionfs, that would have been more like 3 or 4 hundred meg free.
Yes. But where is bites you is that removing things provided on the read-only side of the unionfs join doesn't free up space. I have no use for Skype for instance so I removed it. Thing is, if I hadn't done away with the unionfs that wouldn't have freed up any space. Now, I just have a regular ext2 partition. Once I got things the way I wanted them, I booted from a USB keychain and sshed the partitions and the bootloader to another machine. I'm just as safe from horking it up and have much more latitude to personalize the machine.
Of course, it should also have an easily accessible terminal emulation so the gurus can hack on their config files;-)
Ctrl-Alt-T brings up an xterm and there are two tools called "Easy Mode Editor" and TweakEE that allow other things to be changed including changing the xterm to something nicer like Konsole.
I've considerably tweaked the one I'm using. OpenOffice 2.4 instead of the included 2.0. I built the latest MPlayer from Debian Unstable on an Etch* machine with all codecs and so-forth enabled and installed that as well as various nice things for the sysadmin on the go. Unlike most Macs and Windows machines, it recognizes my USB-to-RS232 dongle immediately and I have a dterm binary sitting on it for configuring switches and the like. It is excellent for little jobs like that since the unit is so "toolbag friendly".
I also ditched the XP-like Icewm them since it is waaay to hard on the limited screen real estate. I'm currently running a very plain but functional theme that saves 32 pixels or so of the limited screen height.
Asus doesn't tout it as such but it is really quite friendly to the experienced Linux user. It is even friendlier if you can endure what it takes to remove the unionfs they use to idiot proof it and go to a straight ext2 partition.
* As long as you don't touch the tweaked QT or kdelibs they are using, you can install any package built for Debian Etch that you like. I have an Etch buildhost I use for anything nifty I want but don't have.
I don't quite get your dig on the NeoOffice guys. You can download and use NeoOffice without paying them a dime and nothing about the software is crippled. It IS true that in exchange for a small fee they'll grant access to binaries of their development versions (source for any version is free if you can stand the build process.....) but that is just a way for motivated users to fund development. The NeoOffice devs also do a good job of working with their own contributors.
I've also watched the disrespect those guys have had to endure getting bugfixes back upstream into OpenOffice. No, they "get" Open Source.
I'll agree that in general Mac users tend to be demanding, whiny, disrespectful, and "stuck in a shareware culture" but why the slam on the NeoOffice guys?
It also means Mac users just make a pain in the ass of themselves to support. Say I develop an app. The Windows users like it. The Linux users like it and the BSD users like it. Hell, it even got ported to Amiga and they like it too. But then I find my inbox stuffed full of whinging from Mac users and maybe I don't even have a Mac. Some other developer was kind enough to fix it so XCode would build it so I chucked in my tree and why not?
Enough of that brand of bullshit, and I would happily say "fuck it" when it comes to Mac support. Hell, and I even like OS X. I'm not a Stevie jobs fanboi but I like it well enough. I don't if I'd like it well enough to give gifts to what seems to be it's typical user though.
Actually, it seems to me that API cruft is dealt with in UNIX like OSes thusly:
"system call foo is in maintenance mode. no new features will be added"
2 years later: "system call foo is deprecated. new applications should not use it. maintained old apps should be migrated to new call foo.so.2 whichs solves the following issues...."
4 years later: "We really mean it this time. In the next release of libbar, foo.h will be removed."
5 years later: flamewar on developers list continues.....
5 1/2 years later: 2 or 3 devs stalk off in disgust. foo removed from libbar.
6 years later: distros remove unmaintained old apps frobnosticator, flubbertron, and happyfunball because no one can be bothered to cure foo dependency. Pundits hold this will be death of Linux on Desktop.
I don't know about.NET but I see remnants of DOS (or is that CP/M) every time I open "My Computer" and every time I save a file to a local disk (and usually a network share) on Windows.
The drive letter thing sucked in 1981 and it never did stop sucking.
It's only more intuitive until something blows up. It's then you find that the fixes are at least as arcane as anything Linux may expect of you. I've tangled with horked domain security policies, AD issues, and Exchange issues. All were at least as hard if not harder to troubleshoot and fix as the equivalent Linux issues. In the case of a Linux server I had that wouldn't boot at all, I got it back up in less than 10 minutes with a Knoppix disc. Every Windows server I ever had that wouldn't boot, things got ugly. I fixed the last one of those with a livecd too (mirrored volume lacked a bootloader and needed one written to it. Could not convince any of the MS tools to do it for me......)
Given the choice between a little harder to configure but Just Does It's Job or something that is easy to configure but violently eats itself, I'll take the first.
Only in action dramas do people revolt over having their rights trampled. The reality is that as long as our Glorious Leaders can keep us in beer, TV, and junk food then they can do pretty much as they please. It's called bread and circuses and it kept the Roman Empire around hundreds of years past the point of their highest glory.
Now couple the trampling with people who can neither eat nor get the resulting sicknesses treated and you have something.
This is a little silly on the face of it. There is little doubt that Israel could obliterate them right back and that is BEFORE we chuck a MIRV or two in their general direction. Israel has reliable delivery systems and there is very little doubt they have nukes of their own. And more than one or two nukes. It's probably more like 30. Israel can annihilate the cities of any Middle Eastern state of their choosing and still have a stick to wave afterwards.
If Iran were so foolish as to attempt to "obliterate" Israel, Iran would cease to exist within hours of the attempt.
Middle Eastern leaders talk of destroying Israel because it plays well to the masses and the Iranian leadership are crazy like foxes in this regard. These leaders themselves live comfortable privileged lives and will not act like the young suicide bombers they employ as cannon fodder. The mad-dog Arab who will do anything is a propaganda tool meant to scare the shit out of the West. And it works.
While the good guys are building code and doing what you say, the elephant in the room is that MS is bribing politicians, pulling little tricks like having Bumphackistan join the ISO just to flog their abortion of a standard, saying publically they won't adhere to their own standard (which in fact they don't) and in general pulling every dirty trick in the book. A few people have the sheer unmitigated gall to turn the lights on these cockroaches and you call it whinging?
Um, no just building a superior product isn't going to cut it when your opponents will basically lie and cheat to get their way.
Yeah. That is basically what I was getting at. Running say KDE apps inside say XFCE is little different from running them on OS X. So yes the desktop shell is separate. When running KOffice on OS X what you have is an application using "KDE APIs". When running it on KDE, it is the office portion of a full "Desktop Environment". I'm saying is that a DE is a "Desktop Shell"+"extensively encompassing API" and furthermore the Desktop Shell is developed against those self-same APIs so that you can have a highly integrated desktop whose apps work together.
Happily the apps will work without the shell but when running them on a "foreign" environment you get complaints about all the services that start up in the background. That's because a set of apps that go together have to talk to each other and "foreign" apps somehow and they need to do this without reinventing the wheel every time you write a new one.
They have to ensure this "*never*"ness is 100% flawless on all of the millions of $80 system boards it is going to be implemented on. But even with the best efforts they won't achieve that and they aren't going to get the best effort from all of the mobo manufacturers. Ensuring this crap is dead perfect is costly for one and accidentally on purpose screwing it up will prove profitable as DVD player manufacturers already well know.
If using "cult" negatively, I'd say the best two criteria are these:
1. The group is the member's sole source of support and validation. Any other form of emotional or intellectual support especially from skeptical family members is strongly discouraged and/or punished.
2. The group demands unreasonable amounts of the member's time and/or money. "Unreasonable" being defined as an amount that makes it very difficult to fulfill obligations outside the group like paying bills, attending family functions, or even being able to afford decent food.
Clearly, one can belong to a religious group without demands of that magnitude being regularly expected.
Most of the minor distros are specialty items like Knoppix or toolkits like Trinity Rescue CD. Having more of such isn't going to hurt anything. Besides, projects like this are a good way of trying out radical ideas without breaking anything. And I suspect the answer to "not teaming up" is that it seems that many developers would rather be Chiefs than common braves.
It's really a shame for F/OSS that, time and time again, there is such a huge duplication of effort and half-assed half-finished projects lying around in the junkyard of the Open Source cemetery.
And once again someone falls prey to a common misconception: F/OSS is not a monolith. If these guys didn't have the option of having their own sandbox to play in then what makes you think they'd be compelled to play in someone else's? The way this will more than likely shakeout is that fifty or so people will use this for awhile. Maybe it'll be a bit more popular if the primary devs have more stature than I'm giving them credit for.
These guys will get to have their fun and most everybody else will use an established distro. And that isn't to say good won't come of it. If they have good ideas, the bigger distros might adopt them. If they have REALLY good ideas they may supplant Gentoo among that crowd of people. Bugfixes may also go to upstream projects.
I know this is weird idea to someone accustomed to being served what they think they want from proprietary software houses but this is nothing but an exercise of freedom. Others are free to use what they make or not. What would you propose? Some sort of law saying that henceforth no one may attempt to start a BSD or Linux distribution?
The F/OSS world operates on a form of street-cred. These guys will either get it or not. It won't cause any sort of actual problem either way.
There is no doubt their fanbois can't handle a few simple truths or plausible suppositions at least. Mod back up and slap these punks on metamod.
I'm not really seeing your argument. The included Linux is more of a diagnostic and rescue tool. Most people who have one of these computers will never even see the Linux desktop on the things. Only techs and Linux gearheads will bother with it.
And we aren't saying "bundling is teh Evel!". We're saying that a monopoly abusing bundling to push out competition is. The Linux on these things will not be the front and center user experience for most people who use these boards. It isn't in the same ballpark as some of the things MS has done. It isn't even the same sport.
It's best to think of the Xandros as a bastardized Debian Etch. Chuck etch apt deb lines into the sources.list. You can safely do a "apt-get upgrade" but DON'T do an "apt-get dist-upgrade". You can install any Etch package or package built for Etch as long as it
1. Doesn't replace or conflict any package with "asus" or "xandros" in the name.
2. Doesn't conflict or replace kdelibs or libqt-mt.
3. Doesn't replace or conflict the kernel or supplied drivers.
Those caveats don't get in the way of much at all. My EEE has OpenOffice 2.4 from Sun's site (there is a guide you need to follow on this one so the Easy Mode icons still work), a recent Adobe reader, mplayer from Sid built on an Etch machine and installed, taglib1.5 so AlbumArt works correctly in "Music Manager (Amarok)", dterm for configuring serial devices, and a few other things that escape my mind. The biggie is being able to build Sid packages on this Etch buildhost I keep around. That is the keyhole that lets me install anything I want without losing the good job Asus did of making wireless, suspend, video, sound, and the Desktop Just Work. Once FireFox 3 goes out of beta, I'll chuck that on too.
The one improvement that is somewhat difficult is removing UnionFS. If you do things like replace OpenOffice, you want that because you otherwise lose space for every large package that is "removed" but invisibly occupying space.
This manual doesn't sound like Grade A ultraweirdness on the order of OT III. Instead we have a rather pedestrian model of conduct and procedure from a church. Just let this one die already.
You're partially missing my point. On a product that will make 5 or 6 bucks of profit per unit sold, MS causes pain to the manufacturer at any price point. As the "race to the bottom" continues and these things become commodities like Casio watches, a bit of superficial familiarity isn't going to matter that much. MS will be put in a position where they don't dictate the price or featureset. If they do try to dictate the price, Linux starts squeezing right back in because being forced off some of these devices for a year or three won't hurt it the way it would hurt a corporate product. Sure, people may want familiar XP from Mama Microsoft. But that isn't a powerful enough effect to let MS dictate terms and prices in a cheap commodity market.
But Microsoft can, and likely will, continue to cut OEM licensing costs.
Indeed they can but that kicks their current business model in the balls. MS is accustomed to receiving a 20-100 dollar premium on every PC sold. They can temporarily push Linux out by dropping the price to $5 or $10 but this tactic will not suffice to entirely drive out Linux. If we're talking about $100 PCs being sold in blister packs at Wal-Mart, even $5 or $10 represents a significant raw materials cost.
Linux based systems will have no artificial limitations mandated for inexpensive new PC categories but MS will insist on all kinds. They'll either cripple the "UMPC Editions" or mandate what can and can't be built into hardware eligible for those licenses. MS themselves can create the situation where it is the cheap Linux PC that does more than the not as cheap Windows PC.
The funny thing is that I used to be that way and then crossed a point about four years ago where I could generally zero in and fix the problem. That did a lot to endear Linux and the other free Unixes to me. It is generally possible to recover from screw ups, even really bad ones, as long as the hardware is still OK.
/etc directory
These days I have a USB hard drive that contains:
A list of all installed packages.
A tarred up copy of the
Entire home directory. (which contains all installers for anything I didn't get from repositories as well)
Even if my hard drive were to completely eat it, those three things would get me my exact working environment back in an hour or two. Most of that time would be spent downloading the installer and packages. But, I have yet to have to resort to that. I've picked up enough of the way a Linux system hangs together that I can just fix the damn thing if it breaks usually.
If you do things like replacing the supplied Acrobat and OpenOffice then yes, freeing space allows for latitude for customization. Otherwise, those two changes will cost about 400MB of space and that isn't all that I have done. When you only have about 1.6 GB free, that is significant and indeed cuts "latitude" for making changes. As it is, I have a software load on the unit that I like and about 1.3 GB free of user space. If I hadn't removed Unionfs, that would have been more like 3 or 4 hundred meg free.
Yes. But where is bites you is that removing things provided on the read-only side of the unionfs join doesn't free up space. I have no use for Skype for instance so I removed it. Thing is, if I hadn't done away with the unionfs that wouldn't have freed up any space. Now, I just have a regular ext2 partition. Once I got things the way I wanted them, I booted from a USB keychain and sshed the partitions and the bootloader to another machine. I'm just as safe from horking it up and have much more latitude to personalize the machine.
Of course, it should also have an easily accessible terminal emulation so the gurus can hack on their config files ;-)
Ctrl-Alt-T brings up an xterm and there are two tools called "Easy Mode Editor" and TweakEE that allow other things to be changed including changing the xterm to something nicer like Konsole.
I've considerably tweaked the one I'm using. OpenOffice 2.4 instead of the included 2.0. I built the latest MPlayer from Debian Unstable on an Etch* machine with all codecs and so-forth enabled and installed that as well as various nice things for the sysadmin on the go. Unlike most Macs and Windows machines, it recognizes my USB-to-RS232 dongle immediately and I have a dterm binary sitting on it for configuring switches and the like. It is excellent for little jobs like that since the unit is so "toolbag friendly".
I also ditched the XP-like Icewm them since it is waaay to hard on the limited screen real estate. I'm currently running a very plain but functional theme that saves 32 pixels or so of the limited screen height.
Asus doesn't tout it as such but it is really quite friendly to the experienced Linux user. It is even friendlier if you can endure what it takes to remove the unionfs they use to idiot proof it and go to a straight ext2 partition.
* As long as you don't touch the tweaked QT or kdelibs they are using, you can install any package built for Debian Etch that you like. I have an Etch buildhost I use for anything nifty I want but don't have.
I don't quite get your dig on the NeoOffice guys. You can download and use NeoOffice without paying them a dime and nothing about the software is crippled. It IS true that in exchange for a small fee they'll grant access to binaries of their development versions (source for any version is free if you can stand the build process.....) but that is just a way for motivated users to fund development. The NeoOffice devs also do a good job of working with their own contributors.
I've also watched the disrespect those guys have had to endure getting bugfixes back upstream into OpenOffice. No, they "get" Open Source.
I'll agree that in general Mac users tend to be demanding, whiny, disrespectful, and "stuck in a shareware culture" but why the slam on the NeoOffice guys?
It also means Mac users just make a pain in the ass of themselves to support. Say I develop an app. The Windows users like it. The Linux users like it and the BSD users like it. Hell, it even got ported to Amiga and they like it too. But then I find my inbox stuffed full of whinging from Mac users and maybe I don't even have a Mac. Some other developer was kind enough to fix it so XCode would build it so I chucked in my tree and why not?
Enough of that brand of bullshit, and I would happily say "fuck it" when it comes to Mac support. Hell, and I even like OS X. I'm not a Stevie jobs fanboi but I like it well enough. I don't if I'd like it well enough to give gifts to what seems to be it's typical user though.
Did he fart in their general direction?
Actually, it seems to me that API cruft is dealt with in UNIX like OSes thusly:
"system call foo is in maintenance mode. no new features will be added"
2 years later: "system call foo is deprecated. new applications should not
use it. maintained old apps should be migrated to new call foo.so.2 whichs
solves the following issues...."
4 years later: "We really mean it this time. In the next release of libbar,
foo.h will be removed."
5 years later: flamewar on developers list continues.....
5 1/2 years later: 2 or 3 devs stalk off in disgust. foo removed from libbar.
6 years later: distros remove unmaintained old apps frobnosticator,
flubbertron, and happyfunball because no one can be bothered to cure foo
dependency. Pundits hold this will be death of Linux on Desktop.
I don't know about .NET but I see remnants of DOS (or is that CP/M) every time I open "My Computer" and every time I save a file to a local disk (and usually a network share) on Windows.
The drive letter thing sucked in 1981 and it never did stop sucking.
It's only more intuitive until something blows up. It's then you find that the fixes are at least as arcane as anything Linux may expect of you. I've tangled with horked domain security policies, AD issues, and Exchange issues. All were at least as hard if not harder to troubleshoot and fix as the equivalent Linux issues. In the case of a Linux server I had that wouldn't boot at all, I got it back up in less than 10 minutes with a Knoppix disc. Every Windows server I ever had that wouldn't boot, things got ugly. I fixed the last one of those with a livecd too (mirrored volume lacked a bootloader and needed one written to it. Could not convince any of the MS tools to do it for me......)
Given the choice between a little harder to configure but Just Does It's Job or something that is easy to configure but violently eats itself, I'll take the first.
http://www.justice.gov/atr/cases/ms_tuncom/major/mtc-00030631_ex4.htm
Only in action dramas do people revolt over having their rights trampled. The reality is that as long as our Glorious Leaders can keep us in beer, TV, and junk food then they can do pretty much as they please. It's called bread and circuses and it kept the Roman Empire around hundreds of years past the point of their highest glory.
Now couple the trampling with people who can neither eat nor get the resulting sicknesses treated and you have something.
This is a little silly on the face of it. There is little doubt that Israel could obliterate them right back and that is BEFORE we chuck a MIRV or two in their general direction. Israel has reliable delivery systems and there is very little doubt they have nukes of their own. And more than one or two nukes. It's probably more like 30. Israel can annihilate the cities of any Middle Eastern state of their choosing and still have a stick to wave afterwards.
One Defense Intelligence Agency estimate puts the number of Israeli nukes at 65 to 85 weapons.
http://www.fas.org/nuke/guide/israel/nuke/
http://en.wikipedia.org/wiki/Israel_and_weapons_of_mass_destruction
If Iran were so foolish as to attempt to "obliterate" Israel, Iran would cease to exist within hours of the attempt.
Middle Eastern leaders talk of destroying Israel because it plays well to the masses and the Iranian leadership are crazy like foxes in this regard. These leaders themselves live comfortable privileged lives and will not act like the young suicide bombers they employ as cannon fodder. The mad-dog Arab who will do anything is a propaganda tool meant to scare the shit out of the West. And it works.
While the good guys are building code and doing what you say, the elephant in the room is that MS is bribing politicians, pulling little tricks like having Bumphackistan join the ISO just to flog their abortion of a standard, saying publically they won't adhere to their own standard (which in fact they don't) and in general pulling every dirty trick in the book. A few people have the sheer unmitigated gall to turn the lights on these cockroaches and you call it whinging?
Um, no just building a superior product isn't going to cut it when your opponents will basically lie and cheat to get their way.
Yeah. That is basically what I was getting at. Running say KDE apps inside say XFCE is little different from running them on OS X. So yes the desktop shell is separate. When running KOffice on OS X what you have is an application using "KDE APIs". When running it on KDE, it is the office portion of a full "Desktop Environment". I'm saying is that a DE is a "Desktop Shell"+"extensively encompassing API" and furthermore the Desktop Shell is developed against those self-same APIs so that you can have a highly integrated desktop whose apps work together.
Happily the apps will work without the shell but when running them on a "foreign" environment you get complaints about all the services that start up in the background. That's because a set of apps that go together have to talk to each other and "foreign" apps somehow and they need to do this without reinventing the wheel every time you write a new one.