Just throwing out a wild guess with no evidence behind it, but could it be because engineers are less able to hold simutainous contradictory beliefs?
It might also be because in their day to day jobs, engineers are actively pitted against the forces of evolution, which tend more to be destructive than creative in their sphere of experience. They use their hearts and minds to build stuff, and random undirected change does its best to knock it down. Perhaps that background makes one less likely to assume that random undirected change on its own could do better than what smarth humans are trained and employed to do - otherwise, why do engineers have a job?
What's missing is a reputation system that would tag those bad decision makers as toxic.
That's a great idea! Some kind of scarce unique token which decision makers could compete for should do it. We could call it.. "Money"
Oh, wait. Crap.
Isn't the whole point of free markets that they don't reward bad decisionmaking? And yet, apparently they do. The entire foundation of our planetary social organisation is based on a lie.
And as any schoolchild can tell you, first you copy what has been done, then you invent your own using the knowledge obtained...
Bah! Kids of today have no morals, pirating the alphabet. Hardworking Greek and Roman captains of slav - er, industry invented those letters and now these little punks think they can just copy them and upload to the Internet for free? And this "arithmetic" business? Pure mental looting. Why, there's fifteen hundred patents on 1 plus 1 alone. Where's John Galt when you need him?
(oh right, he's off tinkering with the static-electric powered invisibility engine he nicked from Doc Smith's Lensmen. but I'm sure he'll be along in a minute.)
I expect to be paying a reasonable subscription fee for my uses of Ubuntu in a few short years. Exactly what shape that would be has yet to be determined, but I know that the cost would be comparable to the cost of electricity to run my desktop. And that I would not be paying for access to the software-- that would always be free-- but I would be paying for convenience features like automatic updates, compatibility-assured packages, and so on.
Can I pay them extra to not "upgrade" my desktop with weird unusable UI experiments?
They've managed it since "No wireless. Less space than a Nomad. Lame". How many/.ers *remember* the Nomad?
I dunno, but I was reading ebooks on a Palm in 1999, and I'm listening to MP3 podcasts on my Blackberry today, and I still don't have an iPod. Your point?
I hope that story helps convince you that you should never have to trust applications, only the operating system kernel.
I'm sorry, you have described a perfectly sensible, common-sense, solution to a security problem which solves the issue at the correct level. This will never succeed in today's software marketplace because (check all that apply):
[x] DOS didn't do it that way [x] Unix doesn't do it that way [x] NT/Win32/Win64/.NET/Win8 doesn't do it that way [x] Angry Birds OS doesn't do it that way [ ] There is no possible way anyone could ever invent an operating system better than what we have now [ ] It's not physically/mathematically feasible
In addition, the following social issues may apply:
[x] It doesn't create enough "computer repair" jobs [x] Antivirus vendors don't like it [x] It doesn't give Google enough ad revenue [x] Apple are too busy trying to figure out how to remove even the power button from i-devices [x] This was a solved problem in the mainframe era, ergo, it must be reinvented from scratch every decade in the PC world
and this is what I think of you:
[ ] You are a stupid person and this is a stupid idea [ ] Sorry, but I just don't think it will fly [x] This is an awesome idea and you should be given a billion dollars to build it [x] This idea is too threatening to the status quo and you will probably be assassinated if you build it
I can't think of a job that C++ would be useless for
How about guaranteeing 100% provable lack of buffer overflows? As far as I can see C++ is pretty useless at that.
And no, "but real programmerz r smrt enuff to nevr make mitsakes!!!" is not an answer. If you can't algorithmically prove that your code is 100% free of 0day exploits, it's darn near 100% guaranteed that it contains them. And if you don't find them before you ship, the black hats will.
If you don't have to use a file manager to do something that a program can automatically do for you, how in the world is that a bad thing by any means?
Because file management is a data-centric use of a computer, while applications are task-centric. And what people want to do with computers is as much or more about data-centricity than task-centricity.
If you restrict people to only manipulating files through applications, then you prevent them from using those files in unexpected (to the application designer) ways. That's not a good thing.
The correct answer is to make file managers more powerful and more easy to use, not to hide them.
And it can be done, of course - as you said, they're just toolkits on top of X11 on top of Linux.
If only they were. But KDE and GNOME aren't just X11 toolkits and widget sets, they're not even just tightly integrated sets of window managers and file explorers and MIME type registries and sound servers and search engines and mail/calendar databases and instant messagers and event notification systems, they're also two fundamentally different component/object models, and sets of IPC daemons - KParts, Bonobo, D-BUS, Pl and friends. They have two different low-level implementation languages and object systems - C++ with the KDE signal/slot preprocessor macros vs C and GObject - and so on. They implement their own entire virtual filesystems in userspace code.
Sure, there's whole chunks of Linux which remain agnostic about what desktop framework is running on top of them, including X11. But a raw Linux and X11 isn't actually that useful to a modern user, unless all you want to do is edit text files in EMACS and browse the Web with Lynx. Generally, you want some kind of runtime support for mounting USB devices when you plug them in, navigating compound documents and ZIP archives as if they were folders, and registering and instantiating component frameworks from dynamically loadable libraries. And most of that stuff is all done at the "desktop framework" level.
One could argue that this kind of file-and-process management functions should be the job of an OS, but it's too late - the Microsoft/Apple approach of "shove it all together into the graphical desktop shell" has won, the open-source DEs have copied the big boys, and now there's a war.
The DE is just a layer on top of the OS that lets you interact with it.
I'm not so sure that that's strictly true - or, if true, that it's a useful distinction to draw.
The big difference between KDE and GNOME is not just that they are different desktop environments but that they are different software frameworks for applications running on them, including different object/component models, inter-process communication and sets of running daemons. Some of these differences are being smoothed over via the Freedesktop.orgs (or were prior to the whole GNOME3/Unity/Plasma debacle where the DEs started chasing "teh shiney" instead of functionality) but major incompatibilities remain.
Yes, you can load a KDE binary under GNOME and if you have the appropriate KDE libraries installed, it will instantiate at least part of the KDE runtime framework (but not the whole daemon set), and it will sorta-kinda run. And yes, if you want to make a hard distinction between "OS" and "framework", you can do so.
The point I would like to make is that there are approaches to software development - for instance, the original Smalltalk vision - where it simply isn't relevant to make a hard distinction between "language", "framework", "runtime", "component system", "library", "set of deamons", and "operating system". If you blur your eyes and squint a little from the perspective of a user, what you have in any of these is a single "system of software components intended to operate together", which most people not involved in kernel development would call an "operating system". It's a system, ie it's made of multiple separate parts, and it operates together, and you can't do stuff without it.
You can sort of break these things down into layers: the OS deals with the hardware and is composed of kernel and drivers and library loaders and a shell, a runtime or virtual machine implements a language and its object model, a component model mediates between multiple language runtimes or VMs, frameworks are libraries which abstract all the other stuff... but it's not always clear why any of these roles should logically belong in any of the levels other than mere historical accident. And we're not today running the kind of machines - 1960s batch-processing mainframes moving toward 1970s time-sharing terminal-server systems - on which these distinctions made sense. Perhaps we should rethink some of our definitions a little?
At the end of the day, the user is faced with various "systems" or "platforms" made out of other premade chunks of software, some of which work together, some of which are shipped together, some of which contain others -- and only a few of these work well with the others. It would be nice if, from the user's perspective, there could be less arguing about exactly whose fault it is that their system doesn't do what it should, and more agreement on how to work together to resolve things.
"Hostile and empty". There's merely the rest of the universe out there.
That's a little like a plankton saying "look, there's a whole another universe out there if we go 'up' from the top of the ocean! It's just like walking to the edge of the whelk shell, only bigger!"
Except the plankton is going to have a better time surviving in dry air than we currently do in microgravity and vacuum.
Yes there's a universe out there. No, it's not built on a scale compatible with human exploration. It's five years just to send a text message to the neighbour's cellphone, 50,000 to get to the drive-in. If you want to explore space, you'd better not have any pressing carbon-based biological business to attend to in this or the next dozen lifetimes.
It would be so much nicer if we could just do meta-programming in the language itself, so to speak.
Yes, it's a pity that the language invented in the 1950s to do just this, and which gave us the IF-THEN statement, garbage collection and object-oriented programming as almost casually discarded spinoffs, has managed to wall itself off from the rest of the programming community and stagnated.
I sometimes wonder what would have happened if someone had ported a Tiny LISP, rather than Tiny BASIC, to the Altair 8800 in 1975.
So your complaint is that people who created something with their and energy for you to use freely, is that they aren't doing enough work for you?
Yeah, if a person freely and generously expends their own time and effort to hit me with a rock, I'm gonna go ahead and say that although they're doing a lot of work, they're not actually doing work that is useful to me. Or even pleasant.
So while I wish them well in their crusade to make the world a better place by hitting passers-by with rocks, possibly I'll cross over to the other side of the street to avoid their generous public-spirited efforts.
Not only that. Every car from every manufacturer also ships with a random number of C4 charges built into the wall panels, each of which is wired to a garage door opener radio receiver and can be triggered remotely by anyone with a garage door opener and the correct frequency, and will explode and kill everyone inside. This is considered normal because it means you get your car cheaply and quickly, and car manufacturers have to compete or they lose the market.
Every month, the manufacturer releases a list of the C4 charges they've "found", and tell you to return to your local service garage to get this month's explosive death removed. You have to hurry or some idiot with a garage door opener will blow you up! Some car manufacturers make this easier by making the car automatically drive you to the closest garage the first time you start it up on Patch Day of each month. It's a pain but it keeps you safe, right? It's taken for granted that if a car blows up, it's the driver's fault. They should have taken it to the garage sooner!
Most manufacturers don't make any other changs to the engine or air conditioner when you take it in for Monthly C4 Removal. But Mozilla has decided to, whenever you take it in for an explosives check, randomly repaint the seats, adjust the air conditioner, change the engine to the front or back, and make your manual gearbox an automatic or vice versa. Because "drivers shouldn't keep driving the same old cars every day, that's just boring". So now Firefox drivers have to decide whether to take their car in for a service, which might make their aftermarket GPS unit, car alarm, and CD player break down, or their seat belt no longer fit them -- or drive around with a carful of C4 waiting to blow up.
Someone looking on might ask "hang on, why are car manufacturers allowed to ship their cars full of C4 in the first place?" That's a silly question! It's because there is no way, no possible way, not ever in the world, for anyone to make cars without C4! And there's certainly no way to check at the factory to see if C4 is inside. It's just physically and mathematically impossible!!! Nobody could ever do it, ever, in a million years.
"But wait, in that case how do the manufacturers find the C4 that they remove every month?"
That's simple! The locations of the C4 get reported to them by local gangs of car thieves, for money. So you're perfectly safe whenever you drive!
"Er. How come the car gangs are able to find the C4 while the manufacturers, who built the cars in the first place, can't?"
Easy! The car gangs use clever instruments called "fuzzers" which they run over the car bodies, and that tells them where the C4 is embedded in the bodywork.
"So why can't the car manufacturers use fuzzers themselves? Don't they have more time and money and information than the gangs, and can't they afford to buy whatever tools they want?"
Um. They... just can't. Nope. Sorry. It's to do with quantum. Now stop asking difficult questions and go take your car in for its monthly C4 removal check and upholstery repaint! And this month, just for you, we're making the gas filler nozzle triangular!
Slashdot Mirror
Poor people very rarely incite revolution if those in power are just as poor and have no ability to change things.
"Powerful" and "poor" are kind of opposites by definition, aren't they?
Just throwing out a wild guess with no evidence behind it, but could it be because engineers are less able to hold simutainous contradictory beliefs?
It might also be because in their day to day jobs, engineers are actively pitted against the forces of evolution, which tend more to be destructive than creative in their sphere of experience. They use their hearts and minds to build stuff, and random undirected change does its best to knock it down. Perhaps that background makes one less likely to assume that random undirected change on its own could do better than what smarth humans are trained and employed to do - otherwise, why do engineers have a job?
Anonymous as an agent for change?
That's a joke. They're nothing more than vandals.
Vandalism is change. Might not be progress, but it's change.
Sometimes, even a broken status quo is better than trying to fix things badly.
What's missing is a reputation system that would tag those bad decision makers as toxic.
That's a great idea! Some kind of scarce unique token which decision makers could compete for should do it. We could call it.. "Money"
Oh, wait. Crap.
Isn't the whole point of free markets that they don't reward bad decisionmaking? And yet, apparently they do. The entire foundation of our planetary social organisation is based on a lie.
We're screwed, aren't we?
And as any schoolchild can tell you, first you copy what has been done, then you invent your own using the knowledge obtained...
Bah! Kids of today have no morals, pirating the alphabet. Hardworking Greek and Roman captains of slav - er, industry invented those letters and now these little punks think they can just copy them and upload to the Internet for free? And this "arithmetic" business? Pure mental looting. Why, there's fifteen hundred patents on 1 plus 1 alone. Where's John Galt when you need him?
(oh right, he's off tinkering with the static-electric powered invisibility engine he nicked from Doc Smith's Lensmen. but I'm sure he'll be along in a minute.)
If memory serves correct, we've only invaded one mid-east country for no reason, and that was to remove a miserable tyrant.
But we replaced him with a happy self-confident tyrant, so it was totally worth it.
I've always wondered why they named a car "Killed In Action".
Fighting Atari would require the hobbiest
Fortunately you don't have to be the hobbiest, just hobbier than the next hobbyist.
I expect to be paying a reasonable subscription fee for my uses of Ubuntu in a few short years. Exactly what shape that would be has yet to be determined, but I know that the cost would be comparable to the cost of electricity to run my desktop. And that I would not be paying for access to the software-- that would always be free-- but I would be paying for convenience features like automatic updates, compatibility-assured packages, and so on.
Can I pay them extra to not "upgrade" my desktop with weird unusable UI experiments?
They've managed it since "No wireless. Less space than a Nomad. Lame". How many /.ers *remember* the Nomad?
I dunno, but I was reading ebooks on a Palm in 1999, and I'm listening to MP3 podcasts on my Blackberry today, and I still don't have an iPod. Your point?
They're at Exxon levels.
I eagerly await the Apple Valdez.
Hi, "hassle" rhymes with "castle"
In what accent? Not mine - here in NZ, "hassle" rhymes with "tassel" and "castle" rhymes with "parcel".
Of COURSE is must be unsafe if its native code. Unless you mean native code on a new architecture that doesn't support pointers to arbitrary memory.
Isn't that precisely what memory management units were invented for?
I hope that story helps convince you that you should never have to trust applications, only the operating system kernel.
I'm sorry, you have described a perfectly sensible, common-sense, solution to a security problem which solves the issue at the correct level. This will never succeed in today's software marketplace because (check all that apply):
[x] DOS didn't do it that way
[x] Unix doesn't do it that way
[x] NT/Win32/Win64/.NET/Win8 doesn't do it that way
[x] Angry Birds OS doesn't do it that way
[ ] There is no possible way anyone could ever invent an operating system better than what we have now
[ ] It's not physically/mathematically feasible
In addition, the following social issues may apply:
[x] It doesn't create enough "computer repair" jobs
[x] Antivirus vendors don't like it
[x] It doesn't give Google enough ad revenue
[x] Apple are too busy trying to figure out how to remove even the power button from i-devices
[x] This was a solved problem in the mainframe era, ergo, it must be reinvented from scratch every decade in the PC world
and this is what I think of you:
[ ] You are a stupid person and this is a stupid idea
[ ] Sorry, but I just don't think it will fly
[x] This is an awesome idea and you should be given a billion dollars to build it
[x] This idea is too threatening to the status quo and you will probably be assassinated if you build it
I can't think of a job that C++ would be useless for
How about guaranteeing 100% provable lack of buffer overflows? As far as I can see C++ is pretty useless at that.
And no, "but real programmerz r smrt enuff to nevr make mitsakes!!!" is not an answer. If you can't algorithmically prove that your code is 100% free of 0day exploits, it's darn near 100% guaranteed that it contains them. And if you don't find them before you ship, the black hats will.
If you don't have to use a file manager to do something that a program can automatically do for you, how in the world is that a bad thing by any means?
Because file management is a data-centric use of a computer, while applications are task-centric. And what people want to do with computers is as much or more about data-centricity than task-centricity.
If you restrict people to only manipulating files through applications, then you prevent them from using those files in unexpected (to the application designer) ways. That's not a good thing.
The correct answer is to make file managers more powerful and more easy to use, not to hide them.
John McCann is a very respectable man and he certainly knows what they're trying to do.
Sure, but is what GNOME's trying to do even remotely close to what GNOME's users want to do?
And it can be done, of course - as you said, they're just toolkits on top of X11 on top of Linux.
If only they were. But KDE and GNOME aren't just X11 toolkits and widget sets, they're not even just tightly integrated sets of window managers and file explorers and MIME type registries and sound servers and search engines and mail/calendar databases and instant messagers and event notification systems, they're also two fundamentally different component/object models, and sets of IPC daemons - KParts, Bonobo, D-BUS, Pl and friends. They have two different low-level implementation languages and object systems - C++ with the KDE signal/slot preprocessor macros vs C and GObject - and so on. They implement their own entire virtual filesystems in userspace code.
Sure, there's whole chunks of Linux which remain agnostic about what desktop framework is running on top of them, including X11. But a raw Linux and X11 isn't actually that useful to a modern user, unless all you want to do is edit text files in EMACS and browse the Web with Lynx. Generally, you want some kind of runtime support for mounting USB devices when you plug them in, navigating compound documents and ZIP archives as if they were folders, and registering and instantiating component frameworks from dynamically loadable libraries. And most of that stuff is all done at the "desktop framework" level.
One could argue that this kind of file-and-process management functions should be the job of an OS, but it's too late - the Microsoft/Apple approach of "shove it all together into the graphical desktop shell" has won, the open-source DEs have copied the big boys, and now there's a war.
The DE is just a layer on top of the OS that lets you interact with it.
I'm not so sure that that's strictly true - or, if true, that it's a useful distinction to draw.
The big difference between KDE and GNOME is not just that they are different desktop environments but that they are different software frameworks for applications running on them, including different object/component models, inter-process communication and sets of running daemons. Some of these differences are being smoothed over via the Freedesktop.orgs (or were prior to the whole GNOME3/Unity/Plasma debacle where the DEs started chasing "teh shiney" instead of functionality) but major incompatibilities remain.
Yes, you can load a KDE binary under GNOME and if you have the appropriate KDE libraries installed, it will instantiate at least part of the KDE runtime framework (but not the whole daemon set), and it will sorta-kinda run. And yes, if you want to make a hard distinction between "OS" and "framework", you can do so.
The point I would like to make is that there are approaches to software development - for instance, the original Smalltalk vision - where it simply isn't relevant to make a hard distinction between "language", "framework", "runtime", "component system", "library", "set of deamons", and "operating system". If you blur your eyes and squint a little from the perspective of a user, what you have in any of these is a single "system of software components intended to operate together", which most people not involved in kernel development would call an "operating system". It's a system, ie it's made of multiple separate parts, and it operates together, and you can't do stuff without it.
You can sort of break these things down into layers: the OS deals with the hardware and is composed of kernel and drivers and library loaders and a shell, a runtime or virtual machine implements a language and its object model, a component model mediates between multiple language runtimes or VMs, frameworks are libraries which abstract all the other stuff... but it's not always clear why any of these roles should logically belong in any of the levels other than mere historical accident. And we're not today running the kind of machines - 1960s batch-processing mainframes moving toward 1970s time-sharing terminal-server systems - on which these distinctions made sense. Perhaps we should rethink some of our definitions a little?
At the end of the day, the user is faced with various "systems" or "platforms" made out of other premade chunks of software, some of which work together, some of which are shipped together, some of which contain others -- and only a few of these work well with the others. It would be nice if, from the user's perspective, there could be less arguing about exactly whose fault it is that their system doesn't do what it should, and more agreement on how to work together to resolve things.
"Hostile and empty". There's merely the rest of the universe out there.
That's a little like a plankton saying "look, there's a whole another universe out there if we go 'up' from the top of the ocean! It's just like walking to the edge of the whelk shell, only bigger!"
Except the plankton is going to have a better time surviving in dry air than we currently do in microgravity and vacuum.
Yes there's a universe out there. No, it's not built on a scale compatible with human exploration. It's five years just to send a text message to the neighbour's cellphone, 50,000 to get to the drive-in. If you want to explore space, you'd better not have any pressing carbon-based biological business to attend to in this or the next dozen lifetimes.
It would be so much nicer if we could just do meta-programming in the language itself, so to speak.
Yes, it's a pity that the language invented in the 1950s to do just this, and which gave us the IF-THEN statement, garbage collection and object-oriented programming as almost casually discarded spinoffs, has managed to wall itself off from the rest of the programming community and stagnated.
I sometimes wonder what would have happened if someone had ported a Tiny LISP, rather than Tiny BASIC, to the Altair 8800 in 1975.
Oops, reply to the post above, not to this one.
So your complaint is that people who created something with their and energy for you to use freely, is that they aren't doing enough work for you?
Yeah, if a person freely and generously expends their own time and effort to hit me with a rock, I'm gonna go ahead and say that although they're doing a lot of work, they're not actually doing work that is useful to me. Or even pleasant.
So while I wish them well in their crusade to make the world a better place by hitting passers-by with rocks, possibly I'll cross over to the other side of the street to avoid their generous public-spirited efforts.
But that's just me.
Not only that. Every car from every manufacturer also ships with a random number of C4 charges built into the wall panels, each of which is wired to a garage door opener radio receiver and can be triggered remotely by anyone with a garage door opener and the correct frequency, and will explode and kill everyone inside. This is considered normal because it means you get your car cheaply and quickly, and car manufacturers have to compete or they lose the market.
Every month, the manufacturer releases a list of the C4 charges they've "found", and tell you to return to your local service garage to get this month's explosive death removed. You have to hurry or some idiot with a garage door opener will blow you up! Some car manufacturers make this easier by making the car automatically drive you to the closest garage the first time you start it up on Patch Day of each month. It's a pain but it keeps you safe, right? It's taken for granted that if a car blows up, it's the driver's fault. They should have taken it to the garage sooner!
Most manufacturers don't make any other changs to the engine or air conditioner when you take it in for Monthly C4 Removal. But Mozilla has decided to, whenever you take it in for an explosives check, randomly repaint the seats, adjust the air conditioner, change the engine to the front or back, and make your manual gearbox an automatic or vice versa. Because "drivers shouldn't keep driving the same old cars every day, that's just boring". So now Firefox drivers have to decide whether to take their car in for a service, which might make their aftermarket GPS unit, car alarm, and CD player break down, or their seat belt no longer fit them -- or drive around with a carful of C4 waiting to blow up.
Someone looking on might ask "hang on, why are car manufacturers allowed to ship their cars full of C4 in the first place?" That's a silly question! It's because there is no way, no possible way, not ever in the world, for anyone to make cars without C4! And there's certainly no way to check at the factory to see if C4 is inside. It's just physically and mathematically impossible!!! Nobody could ever do it, ever, in a million years.
"But wait, in that case how do the manufacturers find the C4 that they remove every month?"
That's simple! The locations of the C4 get reported to them by local gangs of car thieves, for money. So you're perfectly safe whenever you drive!
"Er. How come the car gangs are able to find the C4 while the manufacturers, who built the cars in the first place, can't?"
Easy! The car gangs use clever instruments called "fuzzers" which they run over the car bodies, and that tells them where the C4 is embedded in the bodywork.
"So why can't the car manufacturers use fuzzers themselves? Don't they have more time and money and information than the gangs, and can't they afford to buy whatever tools they want?"
Um. They... just can't. Nope. Sorry. It's to do with quantum. Now stop asking difficult questions and go take your car in for its monthly C4 removal check and upholstery repaint! And this month, just for you, we're making the gas filler nozzle triangular!
So if they are not listening to IT crowd, who do they listen to?
The king of the Potato People.