MS would design and publish standards that are so nasty and obscure that even skilled coders would have a hard time making any sense of it
Microsoft already does. That has been their stated policy with things like SOAP and XML: "make it 'open', but make it so complicated that only we can implement it". Of course, it hasn't worked so far, but they are trying...
Not media, INFORMATION. Notes, meetings, phone numbers, addresses, ideas, sketches, references, URLS, passwords (encrypted natch). For me, this information is still B&W text and graphics. I want
There isn't much difference between high-res B&W and color, at least if Palm supported ClearType (which it doesn't, unfortunately): you might as well use the color screen.
And with the processing power you need to do encryption and PDF display, video rendering comes in for free.
Sorry, but these Clies are pretty much what you want; another good choice is the Sharp SL-C760. Either way, a decent PDA with a good screen will set you back about $600-$700. If you want something cheaper, smaller, and lighter than that, you'll have to wait a bit longer until technology catches up with your desires.
If the lawsuit is groundless, the judge can order SCO to pay IBM's legal fees, in which case SCO lawers get nothing because there would be nothing left to get.
Legal fees can also be used to advantage: if a number of Linux kernel contributors individually filed a countersuit against SCO for possible copyright violations (after all, there is good cause: SCO has admitted that there is code that is the same in SCO Unix and Linux) and dragged them through discovery in multiple states and overseas, the company would implode.
The preliminary injunction in Germany is a good starting point. We need more of that...
In my opinion the real important projects right now are ones that are removing the distinctions between OSs.
Here is another example of why this isn't working: KDE's virtual file system tries to achieve some of the things Plan 9 does. But KDE's virtual file system stuff only works for KDE applications, not anything else.
It gains cross-platform support at the cost of not interoperating with native applications, and at the cost of not interoperating with other cross-platform applications. The only way those approaches work is if everybody ends up using KDE, but then you have essentially eliminated the "cross platform" problem by eliminating the individual platforms and replaced it with a bloated layer of libraries surrounding an operating system core nobody uses. That isn't a good approach as far as I'm concerned.
Cross platform tools like Python, Chandler, Mono and Mozilla. Using standards-based DHTML as the UI. Why add another platform to the mix when the real goal is to become platform agnostic?
You can see the end-point of that with Java. Java is fairly reasonable on Windows and less than mediocre on Linux/X11. And even to achieve that mediocrity, it inserts a huge layer of APIs and code between the operating system and applications. Java applications are bloated, slow, and don't work properly.
Mono is decidedly not cross-platform: it gives you easy access to native libraries and native code/data and people are using them. Gtk# will probably be the most widely used Mono GUI toolkit. It is precisely because Mono is not cross platform that it has a much better chance to catch on with Linux developers (analogous to.NET on Windows).
It all sums up to the same issues that squeak smalltalk has: Everything about it is great, but no-one uses it for anything real.
Squeak is another poster child for why operating system independence sucks. Squeak looks nice, but you can't do anything real in it because it doesn't give you access to the underlying operating system: no desktop integration, very limited I/O, no reuse of operating system libraries.
The operating system matters. If the operating system API sucks, you can't fix it at a higher level. Each level of cross-platform abstraction you add loses some important functionality and performance. The end result is that something like Java or Squeak that thumbs its nose at the underlying OS will never take over; cross-platform is a specialty use.
What is it good for? Right now Plan 9 has no compelling applications and a dearth of the applications most people use daily. This might be fixed soon as people port things like OpenOffice to it, but don't hold your breath.
The easiest way of addressing that would be via a "Linux personality". Unclean as far as Plan 9 is concerned, but a small compromise if the rest is left intact. Works for BSD.
Overcoming the installed base. It took Linux nearly ten years to achieve name recognition, and it still is running a distant third on the desktop.
Linux is not a "distant third" on the desktop; its usage is at least comparable to Mac OS X, if it hasn't surpassed it already.
In any case, one of the big things that is holding Linux back in my opinion is the difficulties of configuring and installing the kernel. I don't know for certain that Plan 9 fixes that, but it looks like it might; it really can't be worse than Linux in that regard.
And if open source developers worried about installed base, they wouldn't have started working on Linux and GNU. Installed base basically doesn't matter if either you plan on competing, or you care about scratching your own itch.
It can be very entertaining for all ages as you can brower the net, check emails, weather, watch movies (on demand), listen to music (on demand) or even cut codes some 30,000 feet in the sky!
All while being curled up in a fetal position in the amazingly "space efficient" seats.
But don't worry, the butt-sensors in the seats will tell you if your leg has merely fallen asleep or whether you are dying from deep vein thrombosis. And with its Internet connectivity, they'll have the ambulance right at the airport when you arrive, without pilot intervention even. Isn't technology great?
This doesn't have to be particularly onerous. The "right to reply" could perhaps simply be implemented by pointing to a discussion group on some other server where people can sign in and respond or by putting in a link to the organization being criticized.
In good ol' American entrepreneurial fashion, this need will probably be met by simple, advertising supported "right to reply" sites that you can point to.
In any case, it's just a proposal. Unlike the US, where the Communications Decency Act actually needed to get struck down by judges, it isn't even a law. McCullagh is on some kind of libertarian crusade, and when he isn't out bashing Democrats, he likes to bash Europe even more. My suggestion would be to ignore him--he has nothing of substance to say (and he is welcome to respond to this posting by hitting the "Reply" button below).
However, the Linux kernel has some pretty significant advantages over the main micro-kernel competitor (BSD's Kernel). It supports more hardware (try getting an HP OfficeJet to work on *BSD). Furthermore, more programs will work with it.
Well, yes, obviously it supports more hardware. It also supports more file systems, more network protocols, and more other stuff.
So, if you really feel the way you do about the Linux kernel, I suggest you start a project with the goal of trying to take the code in the Linux kernel and modularize it.
I have thought about it. That would only make sense if a modular kernel could take advantage of on-going main kernel development. Otherwise, it would be better to start with a new kernel and merely write a driver compatibility layer and Linux personality.
Modularized kernels also have some performance problems. [...] I think the best solution is in a modular development process of the kernel, but a monolithic+microlithic end-user experience.
I agree, except that I think the "monolithic/modular" distinction should not be expressed in the code but only at load time. That is not easily possible with the current Linux architecture. But if communication within the kernel is based on message passing, then you can choose to load kernel components either into the same address space and context (no more overhead than what Linux has right now) or each into its own separate address spaces. I would guess that QNX is taking an approach like that.
(There is actually a package for Linux that lets you put kernel components into user space by forwarding requests as needed, but I believe it's too slow and not fully compatible. With a kernel architected for that purpose from the ground up, that should be much easier.)
The notion that you could make EDRS standardized, mandated, and tamper-proof is absurd. Sorry, it just is. Car companies span across the globe.
They seem to have no trouble adapting to requirements for headlights, restraint systems, exhaust systems, computer maintenance interfaces, and all that other stuff.
This isn't like erasing a message on your answering machine. It's a little chipset integrated with your other car electronics. To tamper with it, you'd have to know how to get to it, how to access the data, and how to replace the data. Unless you work for the auto industry or NTSB designing these things, you probably don't know hwo to do that. So isn't that tamper-proof enough?
Oh, goodie, the "I don't understand it and so it must be hard to tamper with" approach to security. It's as common as it is stupid and responsible for one security disaster after another.
Car electronics are made from standard components and standard software. It isn't very complicated to figure those things out. You can look up the chip numbers and get the data sheets on the web. People managed to reverse engineer something as proprietary and protected as the Xbox and CSS; do you really think an open automotive system presents any difficulties?
The truth of the matter is these things are designed to save lives, not spy on you.
Yes, and that's fine. Problems potentially arise when devices designed to save lives are being used by courts to spy on me, because they are being used for something they weren't designed for.
And it's not like they're designed by incompetant morons either. So the arguement of "what if your car goes airborn with your foot on the gas and it looks like you're going 90" goes out the window.
The people who design these boxes are currently designing them for statistical purposes, not as legal evidence to be used in lawsuits. They probably don't care if readings are wrong 5% of the time because it doesn't matter.
So stop being paranoid people, the little chips in your car aren't out to get you.
I'm not paranoid, nor am I against EDRs. All I'm saying is that if EDRs are also being used regularly as evidence in court cases, they should be designed for that use.
I appreciate QNX as an embedded platform, but I have yet to hear convincing arguments as to how QNX manages to overcome the address translation and additional costs reguarding interprocess communication, with respect to performance.
Well, first of all, a microkernel architecture doesn't require any address translation or additional overhead at all; there have been microkernels that run without any MMU at all. And QNX seems competitively fast.
But let's say, just for the sake of argument, there were overhead associated with it. I would rather have a reliable if slower 2.6 or 3.0 kernel now with the features I need than see the 2.4 kernel limp along from bug regression to bug regression.
Even QNX has faculties for 'lightweight processes' that have independant stacks and a common global data sandbox.
Which only goes to show what I was saying: a microkernel architecture does not require that every single little OS process runs in a separate address space. In fact, a good design would let you decide on the fly whether to isolate a process (and pay the overhead) or run the process in a global address space.
But how do we know you didn't tamper with your EDR to make it appear you are innocent? How do we know that your EDR doesn't cut out at 55mph, while the other driver's brand of EDR routinely overreports? Maybe you were going 55mph (and drifting into the wrong lane) because you were asleep. And perhaps the other driver wouldn't have speeded at all if he had known he had an EDR.
If we are going to have EDRs (and they may be a good idea), they should be standardized, mandated, and tamper-proof. This ad-hoc stuff is no good.
I think this kind of device makes sense if it is standardized. Without standardization, it is difficult to assess how data from different devices compares.
It is also important to have an open debate about this. It may make sense to have black boxes for everybody, but we probably need to increase speed limits if we are going to have 100% enforcement because current speed limits assume some flexibility. And we would need to make sure everybody has those devices and that they are working (perhaps they could be tested along with the smog check), otherwise people could get an unfair advantage in court by disabling their devices.
Guess I'm not taking a long enough view or something, but I can point out a couple of counter examples. Come down south and behold the Kudzu, been quite a while since it was introduced into the local ecology and it is still kicking the ass off of everything it encounters. Or go to south Louisiana and behold the nutria, a small rodent like animal that has wreaked major havok on the local ecology since it arived and shows no sign of dying off anytime soon.
Yes, you aren't taking a long-enough view (and you also missed the qualifier "often"). It takes a while for predators to evolve that can kill off such pests, but they will evolve because the target is just too juicy.
Monolithic kernels, despite their theoretical inferiority, will be faster and more efficient because more developers will be working on them, and will be able to resolve inefficiencies faster.
But that's only initially, when they are small and simple. Once they have become big and complex, they get bogged down.
The Linux kernel is at that stage: kernel builds are a complete mess, as is adding and integrating new functionality. Every Linux machine I ever install the kernel on requires a recompilation to make all the hardware accessible, and then it takes hours of fiddling to find just the right combination of options to make it all work. And that's the experience of pretty much every other Linux user I know. Even then, things don't work properly and regress in serious ways (e.g., the 2.4.20 kernel corrupts file systems mounted via USB2, something that is working in 2.4.19). You can also see the problems with Linux kernel development in the mess around Bitkeeper--if the kernel were architected better, you wouldn't need a very high-tech source code management system to deal with it.
The Linux kernel will collapse under its own weight sooner or later: it simply isn't architected to be extensible.
So, the problems with monolithic archictectures are far from "theoretical", they are concrete and we are living with them every day. And they don't get fixed--there aren't enough programmers around to fix them. And, in fact, mature monolithic projects end up having a small fraction of the programmers they could have if the same projects were designed for extensibility and modularity.
what's cleaner and more efficient, a single program or 10 different ones passing messages to each other?
10 different ones passing messages to each other are clearly cleaner and often more efficient as well. It's the UNIX way. I mean, do you run one command line interface that has all programs linked into it, or do you run a command line shell that invokes programs as separate processes?
The detailed Mach approach itself is broken--far too complex and messy. But you can view Plan 9 as a kind of "microkernel"; that would be a UNIX-style "microkernel". And, of course, QNX is successful as well. The original Amiga OS was a kind of microkernel and worked like a charm.
Note that you can't compare Linux to the original UNIX design. The original UNIX design was kept religiously simple: one file system, a few machine types, etc. Linux, on the other hand, has zillions of modules and features.
So far, it didn't work for any real systems.
It has worked in plenty of real systems. But kludgy monolithic kernels simply have an easier time to attract developers initially--that's why systems like Linux and Windows have managed to grab a lot of market share in the OS area.
Let me second that. I think that's the direction open source operating systems should go.
Microkernels have gotten a bad reputation because Mach/Hurd, for one reason or another didn't deliver. But that doesn't mean the approach itself is flawed.
Traditional monolithic kernels like Linux (and UNIX and NT/XP--and don't try pretending that NT/XP is a "microkernel") are appealing for budding operating system projects because it's easy to hack something together quickly. But those architectures don't hold up in the long run. You can see the same in ecology: fast growing, non-native plants often displace native plants quickly, but in the end, they die because they aren't well adapted to the long-term conditions.
Well, maybe if SCO wins, we can look on the bright side: it will finally get Linux out of its rut and create more opportunities for other kernels. Don't get me wrong: like everybody else, I'd much rather not change from the Linux kernel, but if I do have to change, I don't view it as all bad. (Of course, I don't think SCO has any legal grounds at all, but that is probably not related to whether they can win.)
Re:Should be -2, Moronic
on
42-Volt Autos
·
· Score: 1
BTW, I do NOT clasify all enviromentalists as enviro-waccos.
No, apparently just the ones who disagree with your particular views.
They are responcible for impeding the development of enviromentaly friendly farming, ranching, and manufactering because they don't like those who violate their stereotypes
And your point is? People disagree on lots of things for lots of reasons. You can't divide them into two classes, the "enviro-waccos" and the rest.
Apparently, you have also have succumbed to the binary thinking so common in US politics.
And since when does a basic understanding of electronics, that any high schooler should have, classify as electrical engineering?
Is high-school physics not physics? Is high-school chemistry not chemistry?
clueless or caseless
on
Settling SCOres
·
· Score: 3, Insightful
The fact that SCO presented this without dates or source control log entries strongly suggests that either they either are completely naive about how to establish that code was copied, or that they simply don't have a case and are just playing a huge bluff.
With massive mineral wealth, including minerals needed for high technology and the space programme, china had to invade them.
That's a good commercial justification for them doing it, but China isn't in an arms race with the US. They didn't invade Tibet in order to build more weapons so that they can fight the US.
The hard truth of the matter is "To remain alive, safe and reasonably well off, we must damn the rest of the world."
If you take such a "realpolitik" view of the world, then Americans really have little basis for moral outrage if France, Russia, and Germany give America a hard time in international bodies, and if Muslims make terrorist attacks on the US.
But I don't think US politicians are quite that cynical yet. US politicians still have a naive, misguided view that by taking control, they are helping the world.
The real problem with that view is not that it is deeply immoral (which it is), the real problem is that it won't work. 19th century empires could be built because nations were much less interdependent. The US is far too dependent on foreign resources, foreign money, foreign consumers, and foreign know-how to make this work, and most of those resources are not in the form of anything that can be conquered and looted. If the US were to attack China or Europe or Asia, it would be harming itself more than it would harm the rest of the world.
We have to get off the planet, and begin spreading the human empire to other stars.
So that people can repeat the same mistakes there, multiplying misery a thousand-fold? Before we conquer space, we have to conquer human nature; if we can't, we deserve to die out. If we spread the way we are, we don't have to worry about the parasites from Independence Day, we become the parasites from Independence Day.
In any case, at this stage of technology, the notion of colonizing other solar systems is out of the question. We don't even know yet whether there are any habitable planets. Maybe in a couple of hundred years. Until we can spread, we should assume that we have to deal with our problems at home.
(I think you have played a little too much Civilization, which basically advocates exactly your world view. Too bad that it doesn't just take a tech advance or two to build the colonizing space craft.)
Even if there is some element, somewhere, that's true, what would be their chances of winning against Big Blue? Practically nothing.
Don't be so sure. Juries are often technically incompetent, and they are easily swayed by the "this little guy got crushed by this big bad company" stories. SCO is betting that they can get across as the innocent little victim.
The outcome of this lawsuit probably depends more on whether IBM can convince the jury that SCO is an out-of-control legal parasite than whether SCO has an actual case (which they don't).
Re:Should be -2, Moronic
on
42-Volt Autos
·
· Score: 1
You prove that enviro waccos are ALL a bunch of ignorant twits. Go back to school, and come back after your balls have dropped.
No, he just proved that ONE "enviro wacco" is ignorant of electrical engineering.
You just proved that YOU are an ignorant twit who has problems with elementary logic.
I'm sorry, but I fail to see the contradiction
on
A Mighty Wind
·
· Score: 1
I find it perfectly understandable that people don't want these things in their back yard. I don't see a contradiction with being for clean energy. I mean, do proponents of "clean coal power" or "clean nuclear power" want coal or nuclear power plants in their back yard (you know, the people who run the ads about "Americans for sensible energy choices")? I don't think so. Likewise, the vocal members of Congress that foam at the mouth about the glories of private sector health insurance have cushy government health care coverage, in addition to usually being independently wealthy.
This has really nothing to do with the nature of the energy, but with the nature of political power in the US. In many other nations, the government can eventually just put its foot down and make this sort of thing happen, and the income and power disparities are not as big as they are in the US. In the US, if you have enough money, you get the influence to keep this from happening. That doesnt affect just power plants, it affects every political decision. It's a problem that needs to be addressed, but proponents of coal and nuclear power shouldn't get a monopoly on hypocrisy until then.
But here's the point: exploring an icy moon billions of kilometres away with a nuclear-powered spacecraft, which is the topic of the story, will probably not change the global balance of power.
No, but creating the infrastructure and public opinion change that would allow the US military to put fission and fusion reactors into space will affect the balance of power.
Firstly, public opinion in the US was never overwhelmingly against the war.
We weren't talking about US public opinion (which was apparently for the war), we were talking about international opinion.
Secondly, you're bitter because the US government, together with other governments, made a decision which you don't support
I'm not "bitter" at all. And my objection isn't against the war itself (which I'm pretty indifferent to) but the fact that the US went to war against an autonomous nation that had not attacked it and without the clear and unequivocal mandate of the international community.
developing strong trade relations with the US (a large market) is better than not trading with the US (or trading less). Have a look at Syria, Cuba, Iran, and other countries which the US has trade embargoes against to see (in extreme cases) what a negative effect not trading with the US can have
So, you agree then that Eastern European governments supported the US because the US threatened economic pressure. Great. We are getting somewhere. Now, how is economic blackmail by the US the basis for democratic government and self-determination? How does caving in to US pressure represent "majority opinion"?
You demonize the US, but it's better than most other countries out there would be if they had the same power as the US.
I don't "demonize" the US at all. I think the US exercises the power it has in a fairly benign manner most of the time.
But you assume that just because the US is mostly benign, it should be entrusted with a lot of power. I disagree. Benign and benevolent dictators are still dictators.
The US has to decide whether it wants to lead the world as a dictator and empire, or whether it wants to participate in a democratic governance of the world. Aparently, the US has chosen the "dictator" role for itself for now. I hope the US will reverse that decision because I don't think the rest of the world will tolerate it and I'd much rather not live through the conflicts that will invariably arise if the US persists.
Slashdot Mirror
MS would design and publish standards that are so nasty and obscure that even skilled coders would have a hard time making any sense of it
Microsoft already does. That has been their stated policy with things like SOAP and XML: "make it 'open', but make it so complicated that only we can implement it". Of course, it hasn't worked so far, but they are trying...
Put hooks wherever necessary to update the database when the filesystem changes.
You mean like changedfiles.org, or about a zillion projects before it?
Not media, INFORMATION. Notes, meetings, phone numbers, addresses, ideas, sketches, references, URLS, passwords (encrypted natch). For me, this information is still B&W text and graphics. I want
There isn't much difference between high-res B&W and color, at least if Palm supported ClearType (which it doesn't, unfortunately): you might as well use the color screen.
And with the processing power you need to do encryption and PDF display, video rendering comes in for free.
Sorry, but these Clies are pretty much what you want; another good choice is the Sharp SL-C760. Either way, a decent PDA with a good screen will set you back about $600-$700. If you want something cheaper, smaller, and lighter than that, you'll have to wait a bit longer until technology catches up with your desires.
If the lawsuit is groundless, the judge can order SCO to pay IBM's legal fees, in which case SCO lawers get nothing because there would be nothing left to get.
Legal fees can also be used to advantage: if a number of Linux kernel contributors individually filed a countersuit against SCO for possible copyright violations (after all, there is good cause: SCO has admitted that there is code that is the same in SCO Unix and Linux) and dragged them through discovery in multiple states and overseas, the company would implode.
The preliminary injunction in Germany is a good starting point. We need more of that...
In my opinion the real important projects right now are ones that are removing the distinctions between OSs.
Here is another example of why this isn't working: KDE's virtual file system tries to achieve some of the things Plan 9 does. But KDE's virtual file system stuff only works for KDE applications, not anything else.
It gains cross-platform support at the cost of not interoperating with native applications, and at the cost of not interoperating with other cross-platform applications. The only way those approaches work is if everybody ends up using KDE, but then you have essentially eliminated the "cross platform" problem by eliminating the individual platforms and replaced it with a bloated layer of libraries surrounding an operating system core nobody uses. That isn't a good approach as far as I'm concerned.
Cross platform tools like Python, Chandler, Mono and Mozilla. Using standards-based DHTML as the UI. Why add another platform to the mix when the real goal is to become platform agnostic?
.NET on Windows).
You can see the end-point of that with Java. Java is fairly reasonable on Windows and less than mediocre on Linux/X11. And even to achieve that mediocrity, it inserts a huge layer of APIs and code between the operating system and applications. Java applications are bloated, slow, and don't work properly.
Mono is decidedly not cross-platform: it gives you easy access to native libraries and native code/data and people are using them. Gtk# will probably be the most widely used Mono GUI toolkit. It is precisely because Mono is not cross platform that it has a much better chance to catch on with Linux developers (analogous to
It all sums up to the same issues that squeak smalltalk has: Everything about it is great, but no-one uses it for anything real.
Squeak is another poster child for why operating system independence sucks. Squeak looks nice, but you can't do anything real in it because it doesn't give you access to the underlying operating system: no desktop integration, very limited I/O, no reuse of operating system libraries.
The operating system matters. If the operating system API sucks, you can't fix it at a higher level. Each level of cross-platform abstraction you add loses some important functionality and performance. The end result is that something like Java or Squeak that thumbs its nose at the underlying OS will never take over; cross-platform is a specialty use.
What is it good for? Right now Plan 9 has no compelling applications and a dearth of the applications most people use daily. This might be fixed soon as people port things like OpenOffice to it, but don't hold your breath.
The easiest way of addressing that would be via a "Linux personality". Unclean as far as Plan 9 is concerned, but a small compromise if the rest is left intact. Works for BSD.
Overcoming the installed base. It took Linux nearly ten years to achieve name recognition, and it still is running a distant third on the desktop.
Linux is not a "distant third" on the desktop; its usage is at least comparable to Mac OS X, if it hasn't surpassed it already.
In any case, one of the big things that is holding Linux back in my opinion is the difficulties of configuring and installing the kernel. I don't know for certain that Plan 9 fixes that, but it looks like it might; it really can't be worse than Linux in that regard.
And if open source developers worried about installed base, they wouldn't have started working on Linux and GNU. Installed base basically doesn't matter if either you plan on competing, or you care about scratching your own itch.
It can be very entertaining for all ages as you can brower the net, check emails, weather, watch movies (on demand), listen to music (on demand) or even cut codes some 30,000 feet in the sky!
All while being curled up in a fetal position in the amazingly "space efficient" seats.
But don't worry, the butt-sensors in the seats will tell you if your leg has merely fallen asleep or whether you are dying from deep vein thrombosis. And with its Internet connectivity, they'll have the ambulance right at the airport when you arrive, without pilot intervention even. Isn't technology great?
This doesn't have to be particularly onerous. The "right to reply" could perhaps simply be implemented by pointing to a discussion group on some other server where people can sign in and respond or by putting in a link to the organization being criticized.
In good ol' American entrepreneurial fashion, this need will probably be met by simple, advertising supported "right to reply" sites that you can point to.
In any case, it's just a proposal. Unlike the US, where the Communications Decency Act actually needed to get struck down by judges, it isn't even a law. McCullagh is on some kind of libertarian crusade, and when he isn't out bashing Democrats, he likes to bash Europe even more. My suggestion would be to ignore him--he has nothing of substance to say (and he is welcome to respond to this posting by hitting the "Reply" button below).
I doubt it. We are just too damned efficient at killing off anything that looks like a predator. :)
Many of those "predators" may end up being viruses, bacteria, fungi, nematodes, and insects.
However, the Linux kernel has some pretty significant advantages over the main micro-kernel competitor (BSD's Kernel). It supports more hardware (try getting an HP OfficeJet to work on *BSD). Furthermore, more programs will work with it.
Well, yes, obviously it supports more hardware. It also supports more file systems, more network protocols, and more other stuff.
So, if you really feel the way you do about the Linux kernel, I suggest you start a project with the goal of trying to take the code in the Linux kernel and modularize it.
I have thought about it. That would only make sense if a modular kernel could take advantage of on-going main kernel development. Otherwise, it would be better to start with a new kernel and merely write a driver compatibility layer and Linux personality.
Modularized kernels also have some performance problems. [...] I think the best solution is in a modular development process of the kernel, but a monolithic+microlithic end-user experience.
I agree, except that I think the "monolithic/modular" distinction should not be expressed in the code but only at load time. That is not easily possible with the current Linux architecture. But if communication within the kernel is based on message passing, then you can choose to load kernel components either into the same address space and context (no more overhead than what Linux has right now) or each into its own separate address spaces. I would guess that QNX is taking an approach like that.
(There is actually a package for Linux that lets you put kernel components into user space by forwarding requests as needed, but I believe it's too slow and not fully compatible. With a kernel architected for that purpose from the ground up, that should be much easier.)
The notion that you could make EDRS standardized, mandated, and tamper-proof is absurd. Sorry, it just is. Car companies span across the globe.
They seem to have no trouble adapting to requirements for headlights, restraint systems, exhaust systems, computer maintenance interfaces, and all that other stuff.
This isn't like erasing a message on your answering machine. It's a little chipset integrated with your other car electronics. To tamper with it, you'd have to know how to get to it, how to access the data, and how to replace the data. Unless you work for the auto industry or NTSB designing these things, you probably don't know hwo to do that. So isn't that tamper-proof enough?
Oh, goodie, the "I don't understand it and so it must be hard to tamper with" approach to security. It's as common as it is stupid and responsible for one security disaster after another.
Car electronics are made from standard components and standard software. It isn't very complicated to figure those things out. You can look up the chip numbers and get the data sheets on the web. People managed to reverse engineer something as proprietary and protected as the Xbox and CSS; do you really think an open automotive system presents any difficulties?
The truth of the matter is these things are designed to save lives, not spy on you.
Yes, and that's fine. Problems potentially arise when devices designed to save lives are being used by courts to spy on me, because they are being used for something they weren't designed for.
And it's not like they're designed by incompetant morons either. So the arguement of "what if your car goes airborn with your foot on the gas and it looks like you're going 90" goes out the window.
The people who design these boxes are currently designing them for statistical purposes, not as legal evidence to be used in lawsuits. They probably don't care if readings are wrong 5% of the time because it doesn't matter.
So stop being paranoid people, the little chips in your car aren't out to get you.
I'm not paranoid, nor am I against EDRs. All I'm saying is that if EDRs are also being used regularly as evidence in court cases, they should be designed for that use.
I appreciate QNX as an embedded platform, but I have yet to hear convincing arguments as to how QNX manages to overcome the address translation and additional costs reguarding interprocess communication, with respect to performance.
Well, first of all, a microkernel architecture doesn't require any address translation or additional overhead at all; there have been microkernels that run without any MMU at all. And QNX seems competitively fast.
But let's say, just for the sake of argument, there were overhead associated with it. I would rather have a reliable if slower 2.6 or 3.0 kernel now with the features I need than see the 2.4 kernel limp along from bug regression to bug regression.
Even QNX has faculties for 'lightweight processes' that have independant stacks and a common global data sandbox.
Which only goes to show what I was saying: a microkernel architecture does not require that every single little OS process runs in a separate address space. In fact, a good design would let you decide on the fly whether to isolate a process (and pay the overhead) or run the process in a global address space.
But how do we know you didn't tamper with your EDR to make it appear you are innocent? How do we know that your EDR doesn't cut out at 55mph, while the other driver's brand of EDR routinely overreports? Maybe you were going 55mph (and drifting into the wrong lane) because you were asleep. And perhaps the other driver wouldn't have speeded at all if he had known he had an EDR.
If we are going to have EDRs (and they may be a good idea), they should be standardized, mandated, and tamper-proof. This ad-hoc stuff is no good.
I think this kind of device makes sense if it is standardized. Without standardization, it is difficult to assess how data from different devices compares.
It is also important to have an open debate about this. It may make sense to have black boxes for everybody, but we probably need to increase speed limits if we are going to have 100% enforcement because current speed limits assume some flexibility. And we would need to make sure everybody has those devices and that they are working (perhaps they could be tested along with the smog check), otherwise people could get an unfair advantage in court by disabling their devices.
Guess I'm not taking a long enough view or something, but I can point out a couple of counter examples. Come down south and behold the Kudzu, been quite a while since it was introduced into the local ecology and it is still kicking the ass off of everything it encounters. Or go to south Louisiana and behold the nutria, a small rodent like animal that has wreaked major havok on the local ecology since it arived and shows no sign of dying off anytime soon.
Yes, you aren't taking a long-enough view (and you also missed the qualifier "often"). It takes a while for predators to evolve that can kill off such pests, but they will evolve because the target is just too juicy.
Monolithic kernels, despite their theoretical inferiority, will be faster and more efficient because more developers will be working on them, and will be able to resolve inefficiencies faster.
But that's only initially, when they are small and simple. Once they have become big and complex, they get bogged down.
The Linux kernel is at that stage: kernel builds are a complete mess, as is adding and integrating new functionality. Every Linux machine I ever install the kernel on requires a recompilation to make all the hardware accessible, and then it takes hours of fiddling to find just the right combination of options to make it all work. And that's the experience of pretty much every other Linux user I know. Even then, things don't work properly and regress in serious ways (e.g., the 2.4.20 kernel corrupts file systems mounted via USB2, something that is working in 2.4.19). You can also see the problems with Linux kernel development in the mess around Bitkeeper--if the kernel were architected better, you wouldn't need a very high-tech source code management system to deal with it.
The Linux kernel will collapse under its own weight sooner or later: it simply isn't architected to be extensible.
So, the problems with monolithic archictectures are far from "theoretical", they are concrete and we are living with them every day. And they don't get fixed--there aren't enough programmers around to fix them. And, in fact, mature monolithic projects end up having a small fraction of the programmers they could have if the same projects were designed for extensibility and modularity.
what's cleaner and more efficient, a single program or 10 different ones passing messages to each other?
10 different ones passing messages to each other are clearly cleaner and often more efficient as well. It's the UNIX way. I mean, do you run one command line interface that has all programs linked into it, or do you run a command line shell that invokes programs as separate processes?
The detailed Mach approach itself is broken--far too complex and messy. But you can view Plan 9 as a kind of "microkernel"; that would be a UNIX-style "microkernel". And, of course, QNX is successful as well. The original Amiga OS was a kind of microkernel and worked like a charm.
Note that you can't compare Linux to the original UNIX design. The original UNIX design was kept religiously simple: one file system, a few machine types, etc. Linux, on the other hand, has zillions of modules and features.
So far, it didn't work for any real systems.
It has worked in plenty of real systems. But kludgy monolithic kernels simply have an easier time to attract developers initially--that's why systems like Linux and Windows have managed to grab a lot of market share in the OS area.
Let me second that. I think that's the direction open source operating systems should go.
Microkernels have gotten a bad reputation because Mach/Hurd, for one reason or another didn't deliver. But that doesn't mean the approach itself is flawed.
Traditional monolithic kernels like Linux (and UNIX and NT/XP--and don't try pretending that NT/XP is a "microkernel") are appealing for budding operating system projects because it's easy to hack something together quickly. But those architectures don't hold up in the long run. You can see the same in ecology: fast growing, non-native plants often displace native plants quickly, but in the end, they die because they aren't well adapted to the long-term conditions.
Well, maybe if SCO wins, we can look on the bright side: it will finally get Linux out of its rut and create more opportunities for other kernels. Don't get me wrong: like everybody else, I'd much rather not change from the Linux kernel, but if I do have to change, I don't view it as all bad. (Of course, I don't think SCO has any legal grounds at all, but that is probably not related to whether they can win.)
BTW, I do NOT clasify all enviromentalists as enviro-waccos.
No, apparently just the ones who disagree with your particular views.
They are responcible for impeding the development of enviromentaly friendly farming, ranching, and manufactering because they don't like those who violate their stereotypes
And your point is? People disagree on lots of things for lots of reasons. You can't divide them into two classes, the "enviro-waccos" and the rest.
Apparently, you have also have succumbed to the binary thinking so common in US politics.
And since when does a basic understanding of electronics, that any high schooler should have, classify as electrical engineering?
Is high-school physics not physics? Is high-school chemistry not chemistry?
The fact that SCO presented this without dates or source control log entries strongly suggests that either they either are completely naive about how to establish that code was copied, or that they simply don't have a case and are just playing a huge bluff.
With massive mineral wealth, including minerals needed for high technology and the space programme, china had to invade them.
That's a good commercial justification for them doing it, but China isn't in an arms race with the US. They didn't invade Tibet in order to build more weapons so that they can fight the US.
The hard truth of the matter is "To remain alive, safe and reasonably well off, we must damn the rest of the world."
If you take such a "realpolitik" view of the world, then Americans really have little basis for moral outrage if France, Russia, and Germany give America a hard time in international bodies, and if Muslims make terrorist attacks on the US.
But I don't think US politicians are quite that cynical yet. US politicians still have a naive, misguided view that by taking control, they are helping the world.
The real problem with that view is not that it is deeply immoral (which it is), the real problem is that it won't work. 19th century empires could be built because nations were much less interdependent. The US is far too dependent on foreign resources, foreign money, foreign consumers, and foreign know-how to make this work, and most of those resources are not in the form of anything that can be conquered and looted. If the US were to attack China or Europe or Asia, it would be harming itself more than it would harm the rest of the world.
We have to get off the planet, and begin spreading the human empire to other stars.
So that people can repeat the same mistakes there, multiplying misery a thousand-fold? Before we conquer space, we have to conquer human nature; if we can't, we deserve to die out. If we spread the way we are, we don't have to worry about the parasites from Independence Day, we become the parasites from Independence Day.
In any case, at this stage of technology, the notion of colonizing other solar systems is out of the question. We don't even know yet whether there are any habitable planets. Maybe in a couple of hundred years. Until we can spread, we should assume that we have to deal with our problems at home.
(I think you have played a little too much Civilization, which basically advocates exactly your world view. Too bad that it doesn't just take a tech advance or two to build the colonizing space craft.)
Even if there is some element, somewhere, that's true, what would be their chances of winning against Big Blue? Practically nothing.
Don't be so sure. Juries are often technically incompetent, and they are easily swayed by the "this little guy got crushed by this big bad company" stories. SCO is betting that they can get across as the innocent little victim.
The outcome of this lawsuit probably depends more on whether IBM can convince the jury that SCO is an out-of-control legal parasite than whether SCO has an actual case (which they don't).
You prove that enviro waccos are ALL a bunch of ignorant twits. Go back to school, and come back after your balls have dropped.
No, he just proved that ONE "enviro wacco" is ignorant of electrical engineering.
You just proved that YOU are an ignorant twit who has problems with elementary logic.
I find it perfectly understandable that people don't want these things in their back yard. I don't see a contradiction with being for clean energy. I mean, do proponents of "clean coal power" or "clean nuclear power" want coal or nuclear power plants in their back yard (you know, the people who run the ads about "Americans for sensible energy choices")? I don't think so. Likewise, the vocal members of Congress that foam at the mouth about the glories of private sector health insurance have cushy government health care coverage, in addition to usually being independently wealthy.
This has really nothing to do with the nature of the energy, but with the nature of political power in the US. In many other nations, the government can eventually just put its foot down and make this sort of thing happen, and the income and power disparities are not as big as they are in the US. In the US, if you have enough money, you get the influence to keep this from happening. That doesnt affect just power plants, it affects every political decision. It's a problem that needs to be addressed, but proponents of coal and nuclear power shouldn't get a monopoly on hypocrisy until then.
But here's the point: exploring an icy moon billions of kilometres away with a nuclear-powered spacecraft, which is the topic of the story, will probably not change the global balance of power.
No, but creating the infrastructure and public opinion change that would allow the US military to put fission and fusion reactors into space will affect the balance of power.
Firstly, public opinion in the US was never overwhelmingly against the war.
We weren't talking about US public opinion (which was apparently for the war), we were talking about international opinion.
Secondly, you're bitter because the US government, together with other governments, made a decision which you don't support
I'm not "bitter" at all. And my objection isn't against the war itself (which I'm pretty indifferent to) but the fact that the US went to war against an autonomous nation that had not attacked it and without the clear and unequivocal mandate of the international community.
developing strong trade relations with the US (a large market) is better than not trading with the US (or trading less). Have a look at Syria, Cuba, Iran, and other countries which the US has trade embargoes against to see (in extreme cases) what a negative effect not trading with the US can have
So, you agree then that Eastern European governments supported the US because the US threatened economic pressure. Great. We are getting somewhere. Now, how is economic blackmail by the US the basis for democratic government and self-determination? How does caving in to US pressure represent "majority opinion"?
You demonize the US, but it's better than most other countries out there would be if they had the same power as the US.
I don't "demonize" the US at all. I think the US exercises the power it has in a fairly benign manner most of the time.
But you assume that just because the US is mostly benign, it should be entrusted with a lot of power. I disagree. Benign and benevolent dictators are still dictators.
The US has to decide whether it wants to lead the world as a dictator and empire, or whether it wants to participate in a democratic governance of the world. Aparently, the US has chosen the "dictator" role for itself for now. I hope the US will reverse that decision because I don't think the rest of the world will tolerate it and I'd much rather not live through the conflicts that will invariably arise if the US persists.