Well, I did not ever say that garbage collection rivals the speed of C-compiled programs. What I said was that it made sense lately to a lot of VM-engineers to allocate variables of all kinds on the stack anyway, so the difference is not there anymore. Heap allocation will always be slower, and I responded to the guy because he appeared to be saying that never, no matter what, shall "high-level" languages rival C in speed. Which I think is an inaccurate statement:-)
I'll start with the good things. First of all, I like your style of writing - clear, precise and on point (of your choosing). Second, you explain quite well on the scenery here.
Now, to the bad things. I can almost bet you either are not a day-to-day programmer, as opposed to casually writing simple bits of code in C perhaps, or you just do not know either a lot of computing history or latest developments in compilers and technologies in general. Maybe you write niche software and are not interested in these developments, I do not know, but I think it is a bit odd you give such a good and knowledgeable read, yet completely (in my humble opinion) miss the facts overall.
Machines are different too. There is RISC, there is ZISC, there is VLIW and the CISC/RISC hybrid that modern CPUs mostly are. These days we are also starting to think how we can utilize vector processors, which to gamers are quite familiar as their video cards. Everyone has one, either they know it or not, nowadays they install a 500 mFlops graphics card in PCs in use by hotel receptionists.
So, C was designed to go close to the metal yes, but since metal is different, C may shoot or miss depending on the architecture too.
What is far more important, given that today we still use mostly the same instruction set we used when C was invented, is the fact that you are absolutely mistaken if you think high-level languages will not approach C. You overestimate hand-optimization and underestimate modern compilers. It is illogical to assume that a person IN FRONT of the computer terminal will know and benefit from knowing how a program of his writing may be optimized. It is the computer itself, that, based on sufficiently well developed compiler, has the potential to optimize code. The mere fact that in practice it is not always so, is because the field is immature, but not to worry, rapid developments are made.
Also, things like static typing, static code analysis and other logical solutions absolutely negate any benefit C may have. Also, I am surprised you compare garbage collecting to C, given how programs developed with C still need occasionally, depending on their domain, allocate objects on the heap, and how most virtual machines allocate values on the stack under the hood, even those with garbage collector.
Anyways, to cut short here, and perhaps give you a chance to explain and ridicule me:-), I will just say I find your comparison of C to say LISP is grossly oversimplified, and does not work on me. It is in fact programming paradigms that have liberated compiler writers to write increasingly effective compilers. Spend some time reading on theory of computation on Wikipedia for instance, it has given me a whole new look on the state of the art. Bottomline is, teaching computers how to translate human typed grammar more efficiently into their program execution machine is getting much cheaper and much more fruitful than spending time or energy hand-writing C code, and I am not talking about the "compromise of man hours", I am saying both LISP and C programs being equally 'good', they can be equally fast, especially depending on the LISP compiler.
Thank you for your attention, I know how precious it is here on Internetlands.
I had long suspected, and later has come to prove it to myself, that in the end, a human being can really go long way being active with the brain and cognitive power, as long as that brain is kept well fed with experiences, and not least, challenges. Something this woman has said herself. I wish more people would stop complaining and fearing death (wishful thinking at best, of course) if they realize their brain and body will support them if they themselves lust for more life and are willing to say to themselves "i want to live more, i am not ready to stop being a thinker". That, in my opinion, is the main difference between 70 year old depressed or apathic amnesiacs and those 95 year olds who somehow use Internet, move around and just are being a great example on how to live a mature life.
To ensure that a new PC is malware-free, [Kaspersky] recommended that before users connect the machine to the Internet, they install security software, update it by retrieving the latest definition file on another computer, and transferring that update to the new system, then running a full antivirus scan.
No, what users should do is re-format and re-install.
Now, instead of (and after) producing and emitting abnormally large quantities of CO2 into our atmosphere, we add insult to injury by starting to deplete our oxygen rapidly, yes rapidly - after all we know everybody likes cars, no matter what they drive on, a 500 million or so cars that run on oxygen should do it.
Animals can't drive cars, and neither can plants. But is is our oxygen, right?
Congratulations, my fellow cancer cells! We seem to be spreading:-)
The satellites in our orbit resemble our software. We reinvent the wheel and create a new program for each small task every time we want something done, instead of spending time on some research and find out how to reuse what we have. Most capitalists call this competition, which is fine to an extent, I guess. It is the lack of balance in applying this strategy that is the problem. Competition or not, most existing sattelites/software can be scrapped as such and its task done by combination of other sattelites/software.
How about fusing the satellite functions into a smaller set of functions, like using 12 satellites with multiple functionality like very wide range spectrometers (sound, image, infra-red, ultra-violet, radio, etc) or a set of narrow purpose spectrometers so that instead of launching a new piece of hitech junk into space, the satellite operation time is outsources/leased out to those interested. Just like a cloud computer, we can have a satellite cloud. Ok, we can have several, since some countries are not on speaking terms, and/or value total autonomy, but we do not have to have 20 different variations of GPS-like supporting satellites out there now. I mean Google does not have to launch its own Earth-mapping sat just because it has the money to do so, it can instead invest on replacing lens/mirror/firmware on something already in orbit, although I may be eating my hat saying this, given how costly repairs in space are. Well anyways, if 10 bodies invest in a space-repair, it MAY be cheaper than to launch 10 sats.
Granted, satellite orbit space is getting crowded faster than namespace for software, but I think we can agree that also in software things are getting a bit out of hand.
Anyways, just an analogy.
But then again, maybe we are exaggerating this? Orbit space is pretty damn big, no?
Joke or not, some of us get plenty. All you need is to groom where and when grooming is due, use more of what universe has to offer besides silicon logic gates so as to stay social and interesting to anyone but other nerds, and last but not least, be horny from time to time:-) And no, admitting the last one alone does not help.
And I have a beard and its shagalicious.
P.S. Yes, it is most likely a joke, but it is so goddamn old one, I am not laughing anymore..
The funny thing about money is that the curve of happiness against fortune flats out pretty early - i.e. having more money does not bring the desired effect of being even happier. In other words, not having money is surely a motivation to start having it and live a richer life, but early on from there, having more money does not mean anything anymore with regards to happiness and satisfaction. And I am sure you have enough to live a decent life now already. All that will change is that you will be able to dine at restaurants perhaps twice as often, go to cinema twice as often, travel twice as often. In return you submit your company to buyers will, and just sold your dreams to someone most likely not so idealistic about them as you were.
Point is, if you make money with your company and are happy, continue so and show the finger to those monopolists. Don't let yourself by fooled, they are not offering to buy you because they wish for a better future for you and your friends. It is about fear and control, basic emotions prevalent in humans who think too much life is about survival.
How about if you had a text interface that had some underlining, strikethrough, bold, italic and also colors. Also a help command that was like manpages, formatting content so you can view it better. All with commandline still, which in my opinion is quite a streamlined interface for administrative tasks. You will not forget commands if you have good documentation. I don't think it would take much space, if properly compressed.
How about, yes exactly, if you could run scripts and also store these to ROM or even on external storage devices that BIOS can read natively, f.e. USB sticks?
You fail to take into account how fast things can change on the desktop arena. I say we have had enough of either BIOS, VGA and the text mode as such. For all it is worth, do it like Macs do - startup the minimalistic OFI/EFI with the video card in graphics mode, and boot up the OS from the disk blocks as fast as you can. If anybody wants to mess with their system before the OS loads up, they should press that Alt+Option+O+F or whatever that was, and type firmware commands into console. BIOS accomplishes neither task well - it gives experts stipid interfaces, while they could be using command line instead, and novices do not even know what they are doing in BIOS.
And no stupid 4-bit color Dell/Lenovo/HP/Asus/Acer logos with stupid BIOS text and even more stupid BIOS itself.
There is no need for two operating systems on one computer for the majority of us, BIOS being one. And it will save us 10 seconds of idle time at startup. Degrade the common subset of hardware interfaces so that the only thing the bootstrap procedure needs to do is get to the boot block of whatever device that contains the further loading code. No VGA BIOS and BIOS interface is needed for that in their entirety. Just a way to read the boot sector from a device. That does not need a vendor logo on the screen or the multitude of settings BIOS provides, before these are superceded with OS drivers anyway.
I am not entirely sure you got the point. The point is sudo and su do not elevate privileges of the user calling them who needs to do some administration tasks. Instead they impersonate another user, usually root. The problem is, this impersonation naturally (it is impersonation, right?) brings with itself personal settings of the user being impersonated, usually root, which messes up the human computer interaction. Because the user using the system is still the very same person, only with a strong need (and hopefully access rights) to do whatever root can. He/She wants the privileges root has, not to become root, because becoming root has the already described adverse effect of negating their personal preferences with regard to screen, input devices, home folder, command input history etc. Same person needing elevated privileges very much wants to have the same environment as before, with the same personal preferences he or she always had, they just want more power temporarily. Neither sudo nor su by their very nature and by nature of UNIX systems accomplish this. They just do the easy thing by substituting the user with another user. Which is where the 'su' comes from. Which is fine for the sake of a good name, but there still no way to stay yourself. My user is say, 'owner' but both 'owner' and 'root' are me.
I have taken a look at PolicyKit and after reading a bit, cannot seem to find any relevance to the subject. PolicyKit DOES address some of the other dinosaur issues, but there is not a word that gives hope that like I said, anybody sees the important difference between user impersonation and privilege elevation. PolicyKit does not seem to be addressing the problem, although it does appear to address some of the other important problems.
Besides, and this is an important one, the difference between user impersonation and privilege elevation does not and should not affect application code AT ALL. This is to address your quote "not necessarily integrated into all the applications yet". I think the responsibility is that of the kernel, and its process/thread and security facilities, not individual applications. If a certain Gtk application breaks because it cannot access a resource that for one reason or another can only be accessed by root, the only hope lies in the kernel, because kernel assigns the privileges, not applications.
Apparently, nobody understands the big difference between temporarily elevating own privileges versus becoming another user (with everything that comes with that).
The problems people usually encounter are 1) Why do my GTK applications look weird when I launch them with "administrative privileges", to which experts simply explain that settings for root are not settings for their other user account, hence different theme etc. Even though the answer is correct, the problem is much worse. Impersonating a different user brings about alien setting values, not to mention stupid things like "Cannot find display" error when launching GUI applications from exactly same terminal on exactly the same hardware with exactly the same screen in front.
As you can see the difference is quote noticeable and annoying.
Microsoft, despite its braindead strategies over the years and the general and utter failure that is Vista, got it right thinking in the right direction when they understood the difference between forcing users to run applications as Administrator (thus subscribing them to a whole another environment) and elevating privileges of the SAME user account they are used to. Of course UAC that implements their vision is full of holes, but the idea is noble.
I am sure somewhere in Linux world the same technique exists, but I am not aware of it. Nor does my Ubuntu use it as far as I can see. As soon as I sudo, I become root. I dont want to become root, I just want to have root's power for a while.
The truth is they would like to work on Minix, especially Tannenbaum, but everybody needs money, and some like to combine it with their hobby. They just got the money, and are happy to put their heads to work for it. As simple as that.
I don't know whether you are a programmer, but I am, and for all it is worth, the term "self healing software" means exactly what it is supposed to. When people ask for money, they have to specify what their ideas are all about, however vague they can. You cannot submit a document that reads "Please give money, thinking out a better operating system. Yours truly, A. S. Tannenbaum". There is no politics there. Scientists and researchers suck at politics anyway, and that is a good thing.
Also first the goal and more importantly, the method and general feel are chosen, so no matter how many developers jump on the wagon, it is still going in the same direction it was going, and obviously faster still. Did you ever try to mention "how about doing a microkernel out of Linux instead?" on any forum populated with Linux developers? You will not even be ridiculed, at best just ignored, at worst have gotten a sarcastic comment back at you. Point is, as a certain project picks up weight on its own skeleton, it gains mass and momentum, later on becoming synonymous with its own implementation, rather than interface, so to speak. From there options include a) further development of potentially flawed idea as part of the large and loyal-to-the-roots group b) forking the project using better ideas (same interface, new implementation) while being ridiculed and booed out of that said group, or c) starting anew, which does not anger nor please the group. The latter two gather another group, but after a while the situation is the same as with the first project - the interface is confused for implementation, and the implementation becomes religion. Which in latin means by the way "to unite together".
To put it simply, thousands of OSS developers banging on the wall carving out Linux do not like to look back and rethink, they are too busy. And even if not too busy, the general human psychology does not favour looking back in the middle of a process.
They say any security is as good as its weakest part. Well in thise case defending a $300B project with obviously something cheap enough to allow hackers to download TERABYTES of data before someone found out and before news hit the Internet, that security system must be that weakest link. Obviously no point to invest billions inventing secret machine to defend from your enemies, if the system storing information on this machine can't keep the door closed. I mean if I buy a luxury car I better make sure the garage door is well locked, and if hackers were lock thieves and vice versa, garage doors are being opened up every second.
Wow, that's a lot of speculation (both in volume of text, and in how many assumptions you stack together).
I guess if pulling completely out of Germany (in response to onerous regulation) exposed a company to even more draconian EU-wide regulation, they would simply seek to limit their operations in Germany (with a focus on legal compliance).
Wow, that's some speculation alright. The assumptions stacked together!
Ironically, monopolies like Microsoft, are the one kind of businesses which CANNOT AFFORD to not do business in a EU state. And it has nothing to do with money. Well it does, but not in the sense most of us would think, I guess. It has just as much to do with a related and at least as potent unit of measure - power.
Imagine Microsoft, out of their infinite wisdom (the stupidity of which is explained further), decides to pull out of Germany, because Germany made life harder than sweet (they do not settle for less, do they?) for them. Now I will show you why Microsoft would NEVER do that, no matter how much EU or a EU state pissed them off:
If they pull out of a EU-national market altogether, the said EU state will shortly after demand they open up whatever data formats they need to interoperate with. Obviously since Microsoft no longer offers its products in Germany, and Germany uses other solutions, and since our Information Society depends on information from one continent being usable on another, Germany will politely (first) ask Microsoft to cough up anything and everything they need to read those pesky files coming their way from everywhere around. The funny thing is, platform lock-in in the manner of deliberately not-disclosing data formats in favour of operating only with 'products', is Microsoft's primary reason for revenue. Admit it, this is Microsoft makes money - develop new product, create demand, make everybody produce data using this product, and forget about disclosing specification. Turns out, after half the planet has accumulated piles of data generated with your product, they HAVE to keep around a copy of it, and those of their friends with the copies of said data who do not have the product yet HAVE to buy it, there is no alternatives anyway. Now back to the discussion - put simply Microsoft will NEVER agree to a general policy of disclosing data formats they create, because it is silver bullet to them. Fine, what happens then (provided they have pulled out and refuse to cooperate with said EU state), is that the EU state brings it to the attention of U.S. government, from which point on the game stops to be about software and interoperability, and becomes politics. And you have to be stupid to expect anything else than cooperation between U.S.A. and EU in 2009, with most of the world facilities entirely dependent on software and semi-dependent on eachother intercontinentally. Of course if you are the deepthinker type, you may think there will be a third world war instead. But I think I know what is cheaper for any party involved - force Microsoft shamefully back into EU. World War in 2009 is not only expensive, but so expensive nobody will fund it. Point is, Microsoft WILL TAKE THE BLOW because otherwise they are confined to a life of open standards which is NOT their game. Ironically, giant as they are, they are confined to a very rigid set of rules. And ironically if they do pull out, we get open standards for free!
Like the person below said, however, monopolies do not pay fines, their customers do. But everyone else, who has not yet monopolized their market, DOES pay fines. Lesser things bankrupt companies. Most are just head above water most of their history.
Also, the percentage fines just seem logical. But then again, when EU decides on a fine, they do take into account the size of the business they are targetting.
It is just, in thise case, the sum is ridiculus. They may be a bit afraid, lazy, or have just had a one-night stand with Microsoft.
Even though it may end up being expensive legal mess, chances are the expensive legal mess would end up them paying the fine anyway. Plus their lawyers' expenses, plus their prosecution's lawyers' expenses.
On another note, 9M for Microsoft is not "enough" anyways. Then you may as well not bother fining at all. If EU fined them, they were pretty sure the law is on their side, lawyers effort of both sides included.
I think I will wait out for some more opinions. But thanks for sharing.
Why are we still issuing fines in currency? Obviously, it should be percents of currency. Like percents of wages, income, revenue or whatever likes. Why are we not doing that? It would make it more fair, as a fine of 10% on the income would teach equally good lesson to a company with $10B income as to a company with $10M income. Isn't the whole point of fining to punish and make sure it does not happen again? Because 9M Euros for Microsoft is hardly a lesson more like a routine job for their accountants. Its what they spend on buying PostIt notes for their employees.
Fining them with 10% of their income, would cost them about $6B, which would more likely persuade their suits to not try and play the price fixing game.
How about giving money to lawyers AND giving money to EU? They can't win this one. It is not like their lawyers are miracle workers, even though for the salary they are getting, they do get away with a lot of things that are sure to bury everyone else in the same situation.
Slashdot Mirror
Well, I did not ever say that garbage collection rivals the speed of C-compiled programs. What I said was that it made sense lately to a lot of VM-engineers to allocate variables of all kinds on the stack anyway, so the difference is not there anymore. Heap allocation will always be slower, and I responded to the guy because he appeared to be saying that never, no matter what, shall "high-level" languages rival C in speed. Which I think is an inaccurate statement :-)
I'll start with the good things. First of all, I like your style of writing - clear, precise and on point (of your choosing). Second, you explain quite well on the scenery here.
Now, to the bad things. I can almost bet you either are not a day-to-day programmer, as opposed to casually writing simple bits of code in C perhaps, or you just do not know either a lot of computing history or latest developments in compilers and technologies in general. Maybe you write niche software and are not interested in these developments, I do not know, but I think it is a bit odd you give such a good and knowledgeable read, yet completely (in my humble opinion) miss the facts overall.
Machines are different too. There is RISC, there is ZISC, there is VLIW and the CISC/RISC hybrid that modern CPUs mostly are. These days we are also starting to think how we can utilize vector processors, which to gamers are quite familiar as their video cards. Everyone has one, either they know it or not, nowadays they install a 500 mFlops graphics card in PCs in use by hotel receptionists.
So, C was designed to go close to the metal yes, but since metal is different, C may shoot or miss depending on the architecture too.
What is far more important, given that today we still use mostly the same instruction set we used when C was invented, is the fact that you are absolutely mistaken if you think high-level languages will not approach C. You overestimate hand-optimization and underestimate modern compilers. It is illogical to assume that a person IN FRONT of the computer terminal will know and benefit from knowing how a program of his writing may be optimized. It is the computer itself, that, based on sufficiently well developed compiler, has the potential to optimize code. The mere fact that in practice it is not always so, is because the field is immature, but not to worry, rapid developments are made.
Also, things like static typing, static code analysis and other logical solutions absolutely negate any benefit C may have. Also, I am surprised you compare garbage collecting to C, given how programs developed with C still need occasionally, depending on their domain, allocate objects on the heap, and how most virtual machines allocate values on the stack under the hood, even those with garbage collector.
Anyways, to cut short here, and perhaps give you a chance to explain and ridicule me :-), I will just say I find your comparison of C to say LISP is grossly oversimplified, and does not work on me. It is in fact programming paradigms that have liberated compiler writers to write increasingly effective compilers. Spend some time reading on theory of computation on Wikipedia for instance, it has given me a whole new look on the state of the art. Bottomline is, teaching computers how to translate human typed grammar more efficiently into their program execution machine is getting much cheaper and much more fruitful than spending time or energy hand-writing C code, and I am not talking about the "compromise of man hours", I am saying both LISP and C programs being equally 'good', they can be equally fast, especially depending on the LISP compiler.
Thank you for your attention, I know how precious it is here on Internetlands.
I had long suspected, and later has come to prove it to myself, that in the end, a human being can really go long way being active with the brain and cognitive power, as long as that brain is kept well fed with experiences, and not least, challenges. Something this woman has said herself. I wish more people would stop complaining and fearing death (wishful thinking at best, of course) if they realize their brain and body will support them if they themselves lust for more life and are willing to say to themselves "i want to live more, i am not ready to stop being a thinker". That, in my opinion, is the main difference between 70 year old depressed or apathic amnesiacs and those 95 year olds who somehow use Internet, move around and just are being a great example on how to live a mature life.
To ensure that a new PC is malware-free, [Kaspersky] recommended that before users connect the machine to the Internet, they install security software, update it by retrieving the latest definition file on another computer, and transferring that update to the new system, then running a full antivirus scan.
No, what users should do is re-format and re-install.
Now, instead of (and after) producing and emitting abnormally large quantities of CO2 into our atmosphere, we add insult to injury by starting to deplete our oxygen rapidly, yes rapidly - after all we know everybody likes cars, no matter what they drive on, a 500 million or so cars that run on oxygen should do it.
Animals can't drive cars, and neither can plants. But is is our oxygen, right?
Congratulations, my fellow cancer cells! We seem to be spreading :-)
The satellites in our orbit resemble our software. We reinvent the wheel and create a new program for each small task every time we want something done, instead of spending time on some research and find out how to reuse what we have. Most capitalists call this competition, which is fine to an extent, I guess. It is the lack of balance in applying this strategy that is the problem. Competition or not, most existing sattelites/software can be scrapped as such and its task done by combination of other sattelites/software.
How about fusing the satellite functions into a smaller set of functions, like using 12 satellites with multiple functionality like very wide range spectrometers (sound, image, infra-red, ultra-violet, radio, etc) or a set of narrow purpose spectrometers so that instead of launching a new piece of hitech junk into space, the satellite operation time is outsources/leased out to those interested. Just like a cloud computer, we can have a satellite cloud. Ok, we can have several, since some countries are not on speaking terms, and/or value total autonomy, but we do not have to have 20 different variations of GPS-like supporting satellites out there now. I mean Google does not have to launch its own Earth-mapping sat just because it has the money to do so, it can instead invest on replacing lens/mirror/firmware on something already in orbit, although I may be eating my hat saying this, given how costly repairs in space are. Well anyways, if 10 bodies invest in a space-repair, it MAY be cheaper than to launch 10 sats.
Granted, satellite orbit space is getting crowded faster than namespace for software, but I think we can agree that also in software things are getting a bit out of hand.
Anyways, just an analogy.
But then again, maybe we are exaggerating this? Orbit space is pretty damn big, no?
Joke or not, some of us get plenty. All you need is to groom where and when grooming is due, use more of what universe has to offer besides silicon logic gates so as to stay social and interesting to anyone but other nerds, and last but not least, be horny from time to time :-) And no, admitting the last one alone does not help.
And I have a beard and its shagalicious.
P.S. Yes, it is most likely a joke, but it is so goddamn old one, I am not laughing anymore..
The funny thing about money is that the curve of happiness against fortune flats out pretty early - i.e. having more money does not bring the desired effect of being even happier. In other words, not having money is surely a motivation to start having it and live a richer life, but early on from there, having more money does not mean anything anymore with regards to happiness and satisfaction. And I am sure you have enough to live a decent life now already. All that will change is that you will be able to dine at restaurants perhaps twice as often, go to cinema twice as often, travel twice as often. In return you submit your company to buyers will, and just sold your dreams to someone most likely not so idealistic about them as you were.
Point is, if you make money with your company and are happy, continue so and show the finger to those monopolists. Don't let yourself by fooled, they are not offering to buy you because they wish for a better future for you and your friends. It is about fear and control, basic emotions prevalent in humans who think too much life is about survival.
To Trash-Bin with the whole thing already!
Subject says it all.
How about if you had a text interface that had some underlining, strikethrough, bold, italic and also colors. Also a help command that was like manpages, formatting content so you can view it better. All with commandline still, which in my opinion is quite a streamlined interface for administrative tasks. You will not forget commands if you have good documentation. I don't think it would take much space, if properly compressed.
How about, yes exactly, if you could run scripts and also store these to ROM or even on external storage devices that BIOS can read natively, f.e. USB sticks?
You fail to take into account how fast things can change on the desktop arena. I say we have had enough of either BIOS, VGA and the text mode as such. For all it is worth, do it like Macs do - startup the minimalistic OFI/EFI with the video card in graphics mode, and boot up the OS from the disk blocks as fast as you can. If anybody wants to mess with their system before the OS loads up, they should press that Alt+Option+O+F or whatever that was, and type firmware commands into console. BIOS accomplishes neither task well - it gives experts stipid interfaces, while they could be using command line instead, and novices do not even know what they are doing in BIOS.
And no stupid 4-bit color Dell/Lenovo/HP/Asus/Acer logos with stupid BIOS text and even more stupid BIOS itself.
There is no need for two operating systems on one computer for the majority of us, BIOS being one. And it will save us 10 seconds of idle time at startup. Degrade the common subset of hardware interfaces so that the only thing the bootstrap procedure needs to do is get to the boot block of whatever device that contains the further loading code. No VGA BIOS and BIOS interface is needed for that in their entirety. Just a way to read the boot sector from a device. That does not need a vendor logo on the screen or the multitude of settings BIOS provides, before these are superceded with OS drivers anyway.
I am not entirely sure you got the point. The point is sudo and su do not elevate privileges of the user calling them who needs to do some administration tasks. Instead they impersonate another user, usually root. The problem is, this impersonation naturally (it is impersonation, right?) brings with itself personal settings of the user being impersonated, usually root, which messes up the human computer interaction. Because the user using the system is still the very same person, only with a strong need (and hopefully access rights) to do whatever root can. He/She wants the privileges root has, not to become root, because becoming root has the already described adverse effect of negating their personal preferences with regard to screen, input devices, home folder, command input history etc. Same person needing elevated privileges very much wants to have the same environment as before, with the same personal preferences he or she always had, they just want more power temporarily. Neither sudo nor su by their very nature and by nature of UNIX systems accomplish this. They just do the easy thing by substituting the user with another user. Which is where the 'su' comes from. Which is fine for the sake of a good name, but there still no way to stay yourself. My user is say, 'owner' but both 'owner' and 'root' are me.
I have taken a look at PolicyKit and after reading a bit, cannot seem to find any relevance to the subject. PolicyKit DOES address some of the other dinosaur issues, but there is not a word that gives hope that like I said, anybody sees the important difference between user impersonation and privilege elevation. PolicyKit does not seem to be addressing the problem, although it does appear to address some of the other important problems.
Besides, and this is an important one, the difference between user impersonation and privilege elevation does not and should not affect application code AT ALL. This is to address your quote "not necessarily integrated into all the applications yet". I think the responsibility is that of the kernel, and its process/thread and security facilities, not individual applications. If a certain Gtk application breaks because it cannot access a resource that for one reason or another can only be accessed by root, the only hope lies in the kernel, because kernel assigns the privileges, not applications.
Actually both of them suck, but sudo sucks less.
Apparently, nobody understands the big difference between temporarily elevating own privileges versus becoming another user (with everything that comes with that).
The problems people usually encounter are 1) Why do my GTK applications look weird when I launch them with "administrative privileges", to which experts simply explain that settings for root are not settings for their other user account, hence different theme etc. Even though the answer is correct, the problem is much worse. Impersonating a different user brings about alien setting values, not to mention stupid things like "Cannot find display" error when launching GUI applications from exactly same terminal on exactly the same hardware with exactly the same screen in front.
As you can see the difference is quote noticeable and annoying.
Microsoft, despite its braindead strategies over the years and the general and utter failure that is Vista, got it right thinking in the right direction when they understood the difference between forcing users to run applications as Administrator (thus subscribing them to a whole another environment) and elevating privileges of the SAME user account they are used to. Of course UAC that implements their vision is full of holes, but the idea is noble.
I am sure somewhere in Linux world the same technique exists, but I am not aware of it. Nor does my Ubuntu use it as far as I can see. As soon as I sudo, I become root. I dont want to become root, I just want to have root's power for a while.
No, that is the "realist" in you talking.
The truth is they would like to work on Minix, especially Tannenbaum, but everybody needs money, and some like to combine it with their hobby. They just got the money, and are happy to put their heads to work for it. As simple as that.
I don't know whether you are a programmer, but I am, and for all it is worth, the term "self healing software" means exactly what it is supposed to. When people ask for money, they have to specify what their ideas are all about, however vague they can. You cannot submit a document that reads "Please give money, thinking out a better operating system. Yours truly, A. S. Tannenbaum". There is no politics there. Scientists and researchers suck at politics anyway, and that is a good thing.
Too many cooks spoil the broth, they say.
Also first the goal and more importantly, the method and general feel are chosen, so no matter how many developers jump on the wagon, it is still going in the same direction it was going, and obviously faster still. Did you ever try to mention "how about doing a microkernel out of Linux instead?" on any forum populated with Linux developers? You will not even be ridiculed, at best just ignored, at worst have gotten a sarcastic comment back at you. Point is, as a certain project picks up weight on its own skeleton, it gains mass and momentum, later on becoming synonymous with its own implementation, rather than interface, so to speak. From there options include a) further development of potentially flawed idea as part of the large and loyal-to-the-roots group b) forking the project using better ideas (same interface, new implementation) while being ridiculed and booed out of that said group, or c) starting anew, which does not anger nor please the group. The latter two gather another group, but after a while the situation is the same as with the first project - the interface is confused for implementation, and the implementation becomes religion. Which in latin means by the way "to unite together".
To put it simply, thousands of OSS developers banging on the wall carving out Linux do not like to look back and rethink, they are too busy. And even if not too busy, the general human psychology does not favour looking back in the middle of a process.
They say any security is as good as its weakest part. Well in thise case defending a $300B project with obviously something cheap enough to allow hackers to download TERABYTES of data before someone found out and before news hit the Internet, that security system must be that weakest link. Obviously no point to invest billions inventing secret machine to defend from your enemies, if the system storing information on this machine can't keep the door closed. I mean if I buy a luxury car I better make sure the garage door is well locked, and if hackers were lock thieves and vice versa, garage doors are being opened up every second.
If NASA still has sense of humour left, our future as space-fairing species is secure, as far as I am concerned.
I dunno, what do you call people who write Word documents? Depending on the person and the observer, learning Microsoft Word is much more difficult.
Maybe call HTML folks 'taggers'? After all it is all about tagging, the rest is just text writing.
Wow, that's a lot of speculation (both in volume of text, and in how many assumptions you stack together).
I guess if pulling completely out of Germany (in response to onerous regulation) exposed a company to even more draconian EU-wide regulation, they would simply seek to limit their operations in Germany (with a focus on legal compliance).
Wow, that's some speculation alright. The assumptions stacked together!
Ironically, monopolies like Microsoft, are the one kind of businesses which CANNOT AFFORD to not do business in a EU state. And it has nothing to do with money. Well it does, but not in the sense most of us would think, I guess. It has just as much to do with a related and at least as potent unit of measure - power.
Imagine Microsoft, out of their infinite wisdom (the stupidity of which is explained further), decides to pull out of Germany, because Germany made life harder than sweet (they do not settle for less, do they?) for them. Now I will show you why Microsoft would NEVER do that, no matter how much EU or a EU state pissed them off:
If they pull out of a EU-national market altogether, the said EU state will shortly after demand they open up whatever data formats they need to interoperate with. Obviously since Microsoft no longer offers its products in Germany, and Germany uses other solutions, and since our Information Society depends on information from one continent being usable on another, Germany will politely (first) ask Microsoft to cough up anything and everything they need to read those pesky files coming their way from everywhere around. The funny thing is, platform lock-in in the manner of deliberately not-disclosing data formats in favour of operating only with 'products', is Microsoft's primary reason for revenue. Admit it, this is Microsoft makes money - develop new product, create demand, make everybody produce data using this product, and forget about disclosing specification. Turns out, after half the planet has accumulated piles of data generated with your product, they HAVE to keep around a copy of it, and those of their friends with the copies of said data who do not have the product yet HAVE to buy it, there is no alternatives anyway. Now back to the discussion - put simply Microsoft will NEVER agree to a general policy of disclosing data formats they create, because it is silver bullet to them. Fine, what happens then (provided they have pulled out and refuse to cooperate with said EU state), is that the EU state brings it to the attention of U.S. government, from which point on the game stops to be about software and interoperability, and becomes politics. And you have to be stupid to expect anything else than cooperation between U.S.A. and EU in 2009, with most of the world facilities entirely dependent on software and semi-dependent on eachother intercontinentally. Of course if you are the deepthinker type, you may think there will be a third world war instead. But I think I know what is cheaper for any party involved - force Microsoft shamefully back into EU. World War in 2009 is not only expensive, but so expensive nobody will fund it. Point is, Microsoft WILL TAKE THE BLOW because otherwise they are confined to a life of open standards which is NOT their game. Ironically, giant as they are, they are confined to a very rigid set of rules. And ironically if they do pull out, we get open standards for free!
Just my opinion (easy to forget).
That is sadly very true in this case.
Like the person below said, however, monopolies do not pay fines, their customers do. But everyone else, who has not yet monopolized their market, DOES pay fines. Lesser things bankrupt companies. Most are just head above water most of their history.
Also, the percentage fines just seem logical. But then again, when EU decides on a fine, they do take into account the size of the business they are targetting.
It is just, in thise case, the sum is ridiculus. They may be a bit afraid, lazy, or have just had a one-night stand with Microsoft.
Even though it may end up being expensive legal mess, chances are the expensive legal mess would end up them paying the fine anyway. Plus their lawyers' expenses, plus their prosecution's lawyers' expenses.
On another note, 9M for Microsoft is not "enough" anyways. Then you may as well not bother fining at all. If EU fined them, they were pretty sure the law is on their side, lawyers effort of both sides included.
I think I will wait out for some more opinions. But thanks for sharing.
I am wondering about one thing:
Why are we still issuing fines in currency? Obviously, it should be percents of currency. Like percents of wages, income, revenue or whatever likes. Why are we not doing that? It would make it more fair, as a fine of 10% on the income would teach equally good lesson to a company with $10B income as to a company with $10M income. Isn't the whole point of fining to punish and make sure it does not happen again? Because 9M Euros for Microsoft is hardly a lesson more like a routine job for their accountants. Its what they spend on buying PostIt notes for their employees.
Fining them with 10% of their income, would cost them about $6B, which would more likely persuade their suits to not try and play the price fixing game.
How about giving money to lawyers AND giving money to EU? They can't win this one. It is not like their lawyers are miracle workers, even though for the salary they are getting, they do get away with a lot of things that are sure to bury everyone else in the same situation.