Microsoft is the primary reason hardware is so expensive: Microsoft Windows XP, Microsoft Office, and all that are resource hogs that must be installed, administered, and run locally. Furthermore, it is Windows XP that is so tied to PC hardware. And it is Microsoft (and Apple) that push useless eye candy on the masses, requiring an arms race in hardware.
In an ideal world, we'd have $100 ARM-based, silent, solid state systems on everybody's desk, with lightweight software suites that get day-to-day work done, and most functions moved off to the web. Even as it is, properly configured Linux-based desktop environments can run fine on sub-$200 PCs or embedded PCs.
But, frankly, Microsoft wants it that way. Many people pay for Windows and Office only when they buy a new PC, so if that desktop machine from 1990 were still usable with today's software, Microsoft's revenue would suffer greatly. Furthermore, if PC hardware were down to $100, people would look even more seriously at whether they want to shell out 5-10 times as much for Microsoft software, software whose features they mostly don't use anyway.
'In a country where the average monthly salary is about $240, buying the latest album for $15 is a grotesque luxury, let alone spending $600 on Adobe Photoshop or a similar computer program.'
Copyright infringement is still the wrong answer because it doesn't solve the fundamental problem: that music and software are overpriced. In fact, it makes the problem worse because it creates a dependency on overpriced goods.
Russians could create their own music and software and sell it more cheaply on the world market, benefitting everybody. Or, they could contribute to OSS projects and make them better, again helping everybody. If Photoshop costs $600 and the monthly salary is $240, that should be a big incentive, and it doesn't take a rocket scientist to create good software (although Russia has a lot of those, too).
Let's reduce this to basic principals: are all public places prohibited from annoying all people who might enter them? I mean just consider that. It's ridiculous.
Yes, let's. The basic principle that the parent post was trying to establish is that businesses can do what they want. I pointed out that no such principle exists: if someone makes a good case, a business like an airport can well be prohibited from blaring a TV at people.
Obviously, nobody has yet made that case. But someone could. They could be because of ADA rules, or for many other reasons. People with ADD (attention deficit disorder), for example, are not just annoyed by television sets, they might come to harm because of them. Or we might start recognizing television for what it is: a dangerous drug that people should only be enjoying by themselves in the privacy of their own home, like smoking.
There is no mention in the article of a story running in 2002 of genetic therapies unexpectedly causing cancer,
Nor should there be; general estimates of the number of genes have nothing to do with mechanisms by which gene therapy might cause cancer. Nor is it unexpected that gene therapy can cause cancer; that has always been a known risk.
although if you now factor in the increased complexity of interactions, it is possible that such side-efects can be better understood and therefore prevented.
Anything is possible, I suppose. But common ways in which gene therapy could cause cancer are already understood. Doubtlessly, there are many more possibilities, but to identify them requires a specific understanding of those "interactions", something that is being worked on anyway.
Sun's JRE license does not allow you to redistribute, so for every install, you have to be sure there's a JRE or download it from the internet.
In principle, you can redistribute the JRE, but you probably don't want to: it's big, it's an intrusive install for the user, and it may not be compatible with the target system.
Java also makes every non-private-non-final method virtual, which AFAIK means less opportunity for inlining.
That, in particular, Sun has a good excuse for. A good JIT should be able to make up for that. Also, declaring a class or method "final" can make the JIT's job simpler.
But there are several other areas where the JIT can't compensate for language differences. For example, template classes used with primitive argument types are much more efficient in C#.
Also, the lack of unsigned types in Java can sometimes be painful.
On the other hand, the mono developers did have full access to all the research papers from sun that have been published on JIT compilation, garbage collection, dynamic compilation etc.
When Java came out, garbage collection already was a mature field (Java's GC is still not state-of-the-art). Dynamic compilation and dynamic optimization had also been around for quite a number of years. So, both the Java and Mono implementors had the benefit of hindsight. In fact, Sun had many of the world experts on these subjects in-house when they started Java.
I haven't seen any decisive language feature that gives C# any edge over Java. Most of it is syntactic sugar which is nice to have but not really that important if you have code completion & refactoring tools.
No, it is not just syntactic sugar: the JVM just lacks crucial features. Try defining a class in Java that is equivalent to C++ "complex", then implement a complex FFT with it (complex 2D FFT is even worse). Not only is the Java syntax painful, the performance sucks, too.
Yes they do. It's their damned airport. Don't like it? Try another airport. Can't find one to accomodate your needs? Don't fly.
Airports are not completely private businesses; usually, they have received and continue to receive lots of public support, both in money and in kind. They have received that with the understanding that they provide everybody with good service. They can't just decide to do things that keep away a significant fraction of their customers.
Even if they were completely private businesses, there are still lots of things they cannot do. For example, even if you are the owner of a restaurant, you still may not be permitted to smoke on your premises during business hours.
It's no surprise Sun's JDK seems to perform the best out of the other versions. They know Java the best since they created it,
The techniques necessary for compiling Java well have largely been around since long before Sun even released Java. Sun has no special, secret knowledge there, they have just been hacking on their implementation longer than anybody else. If anything, Sun's progress on Java has been pokey.
Sun Java is probably pretty close to what is theoretically possible, so they will largely be standing still, while other implementations will keep improving until they also hit the limit. In a couple of years, you can expect that all actively developed JVM and CLR implementations will have roughly the same performance on comparable code.
What these benchmarks suggest to me is that C#/.NET is fully competitive in terms of performance with the best Java implementations, and C#/Mono is good enough for most work. In fact, given the additional language features of C#, it may well be easier to write fast code for many compute-intensive problems using Mono than Java today.
The level of performance of Mono is even more impressive given how young the project is; at the same point in time in its evolution, Sun had barely managed to produce a JIT compiler.
You'd think that natively compiled and optimized code would be faster than an interpreter. I guess there's optimization work to be done in gcjlib.
The JIT inside a Java or CLR (C#) runtime has more information available to it for optimization. That means that, ultimately, it should be able to do better than gcj on long-running, compute-intensive jobs.
Well, it's the old "blame the user" attitude. When a group like ITAA makes such a statement, it's not surprising that so much information technology still is so crappy.
In different words, you shouldn't even have to read a user manual to use a modern desktop application. And you certainly shouldn't have to read a user manual in order to vote.
"Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before and it has always been due to human error."
However, the period roughly 10 years ago was the time the original poster was talking about - no real competition for OpenStep existed
Already in the late 80's, you could get several practical Smalltalk implementations (e.g., ParcPlace, Smalltalk/V, Instantiations), and they were widely used commercially. VisualAge came out around 1993/4.
If you wanted something natively compiled, around that time, you could also get high quality native compilers for a variety of object-oriented languages with garbage collection, including Modula-3, Eiffel, and Oberon. Simula-67 even went back as far as the 1960's. If NeXT had wanted a more dynamic object model than those languages already had, the required additions would have been small compared to the changes NeXT made to Objective-C.
I think NeXTStep just represented a slightly different compromise between using standards accepted by industry (C, UNIX) and what was commercially available state of the art. It was so close to what regular "Joe programmers" felt comfortable with that they "got it", but awed them with a few small bits that were thrown in (GUI builder, nice graphical design, Objective-C). So, I think NeXT was a brilliant marketing compromise (and a good implementation), but technologically, I think it has never been state of the art.
First of all, Objective-C is not Smalltalk--not even close. The C heritage of ObjC means you have to deal with pointers, type errors, casts, , compilers, header files, and all that.
Second, the ObjC object model is nothing unusual anymore: people can try it out much more easily using Python or Ruby. And with the bindings of those languages to grk, wx, and qt, people can do mainstream gui programming with languages and environments that are altogether a lot nicer without leaving linux or gnome.
The funny thing is that people who actually experienced the generation of GUIs before NeXT might say the same thing about NeXT. I mean, you have to deal with C, memory allocation, and pointers? You can't inspect and change the code of a running program? Your GUI runs in a separate process and needs to be programmed in a different language? You can have type errors in your running system without anything flagging an error? What kind of backwards system is that?
I agree to the degree that C++ has always been a bad choice for GUI development. But between ObjectiveC/Cocoa and Java/Eclipse or Python/Gtk+, I hardly see what advantages you think ObjectiveC/Cocoa brings. And, yes, I have tried out Apple's IDE.
What was even more amazing was using Smalltalk a decade before NeXT. NeXT was modeled on Smalltalk, but relative to Smalltalk, even NeXT seemed cumbersome.
And the guys who brought you Smalltalk are bringing you more neat stuff; see the story on OpenCroquet earlier.
I didn't call you anything. Macintosh advocates in general regularly post all sorts of Apple trivia to Linux and X11 discussions. And Macintosh advocates in general make all sorts of claims about Linux and X11 in Apple forums on Slashdot. And that's OK: that's what Slashdot is about.
I don't inject Apple topics into Linux discussions, and I don't appreciate others injecting Linux into Apple related discussions.
Then you're on the wrong site. This is "News for Nerds", not "Macintosh driver central", "Macintosh advocacy", or "News for Apple Fans"; there are enough other sites and mailing lists where Apple fans can retreat in isolation, without the inconvenience of having their world view challenged.
No such question was asked. No question was asked at all. It sounded a lot like complaining about a company's policies in the wrong venue.
Slashdot is, in fact, exactly the right site to discuss Apple policies and contrast them with open source because this is a site where lots of people from all sorts of different backgrounds come. If not here, where else? What other site is there where people knowledgeable in different platforms can discuss these issues? The only problem is that people like you are quick to brand anybody as a "Troll" when you don't want to hear what they have to say.
If you want to know what Infiniband brings to the table of Linux supercomputing, the answer may be nothing, for reasons cited. But for what it brings to OS X, the topic of the article, the section of/. and the OS of the drivers released, I'd say it brings something to the table.
I want to know what Macintosh+Infiniband brings to the table of supercomputing in general, relative to, say, Linux+Myrinet. We know that it may give you 20% lower latency in some caes if one post is to be believed, but little else, and even that didn't talk about cost/benefit. The matter seems to be settled in your mind, given your put-down of Linux-based cluster, but, sorry, you haven't made your case.
In case you still do not realize it the analysis of said data is the starting point for the majority of science on global warming, and used as scientific proof for it
The paper in question has indeed been cited over 300 times, so it is important, but it contains several results. We don't now which of those results (if any) citing papers rely on and if the conclusions of those papers would be affected if the McIntyre and McKitrick criticism is accurate. Muller fails to make that analysis in his paper, and he also is not an expert on statistical data analysis.
What we are left with is that both points of view have been published, people have had the opportunity to make up their minds, and it is pretty clear what the mainstream scientific opinion on the matter is, at least for now.
Squid farther north bacause of CO2 and global warming = junk science
No, that's not called "junk science", it's called a hypothesis (and a pretty plausible one). If you have an alternative hypothesis, please advance it. But simply ignoring the observation of species outside their normal range without explanation, that really is "junk science".
I haven't said that the cycles happen by themselves,
No, you didn't say that, you simply ignored the question of causation at all. We currently have one plausible hypothesis for why global warming is happening. When you come up with an alternative, please share it.
than a major flaw being discovered in some of the most used data in scientifically arguing for global warming
The article about PCA isn't about a flaw in the data, it's about a flaw in one particular statistical analysis.
Slashdot Mirror
Microsoft is the primary reason hardware is so expensive: Microsoft Windows XP, Microsoft Office, and all that are resource hogs that must be installed, administered, and run locally. Furthermore, it is Windows XP that is so tied to PC hardware. And it is Microsoft (and Apple) that push useless eye candy on the masses, requiring an arms race in hardware.
In an ideal world, we'd have $100 ARM-based, silent, solid state systems on everybody's desk, with lightweight software suites that get day-to-day work done, and most functions moved off to the web. Even as it is, properly configured Linux-based desktop environments can run fine on sub-$200 PCs or embedded PCs.
But, frankly, Microsoft wants it that way. Many people pay for Windows and Office only when they buy a new PC, so if that desktop machine from 1990 were still usable with today's software, Microsoft's revenue would suffer greatly. Furthermore, if PC hardware were down to $100, people would look even more seriously at whether they want to shell out 5-10 times as much for Microsoft software, software whose features they mostly don't use anyway.
'In a country where the average monthly salary is about $240, buying the latest album for $15 is a grotesque luxury, let alone spending $600 on Adobe Photoshop or a similar computer program.'
Copyright infringement is still the wrong answer because it doesn't solve the fundamental problem: that music and software are overpriced. In fact, it makes the problem worse because it creates a dependency on overpriced goods.
Russians could create their own music and software and sell it more cheaply on the world market, benefitting everybody. Or, they could contribute to OSS projects and make them better, again helping everybody. If Photoshop costs $600 and the monthly salary is $240, that should be a big incentive, and it doesn't take a rocket scientist to create good software (although Russia has a lot of those, too).
Let's reduce this to basic principals: are all public places prohibited from annoying all people who might enter them? I mean just consider that. It's ridiculous.
Yes, let's. The basic principle that the parent post was trying to establish is that businesses can do what they want. I pointed out that no such principle exists: if someone makes a good case, a business like an airport can well be prohibited from blaring a TV at people.
Obviously, nobody has yet made that case. But someone could. They could be because of ADA rules, or for many other reasons. People with ADD (attention deficit disorder), for example, are not just annoyed by television sets, they might come to harm because of them. Or we might start recognizing television for what it is: a dangerous drug that people should only be enjoying by themselves in the privacy of their own home, like smoking.
There is no mention in the article of a story running in 2002 of genetic therapies unexpectedly causing cancer,
Nor should there be; general estimates of the number of genes have nothing to do with mechanisms by which gene therapy might cause cancer. Nor is it unexpected that gene therapy can cause cancer; that has always been a known risk.
although if you now factor in the increased complexity of interactions, it is possible that such side-efects can be better understood and therefore prevented.
Anything is possible, I suppose. But common ways in which gene therapy could cause cancer are already understood. Doubtlessly, there are many more possibilities, but to identify them requires a specific understanding of those "interactions", something that is being worked on anyway.
Sun's JRE license does not allow you to redistribute, so for every install, you have to be sure there's a JRE or download it from the internet.
In principle, you can redistribute the JRE, but you probably don't want to: it's big, it's an intrusive install for the user, and it may not be compatible with the target system.
Java also makes every non-private-non-final method virtual, which AFAIK means less opportunity for inlining.
That, in particular, Sun has a good excuse for. A good JIT should be able to make up for that. Also, declaring a class or method "final" can make the JIT's job simpler.
But there are several other areas where the JIT can't compensate for language differences. For example, template classes used with primitive argument types are much more efficient in C#.
Also, the lack of unsigned types in Java can sometimes be painful.
Indeed.
On the other hand, the mono developers did have full access to all the research papers from sun that have been published on JIT compilation, garbage collection, dynamic compilation etc.
When Java came out, garbage collection already was a mature field (Java's GC is still not state-of-the-art). Dynamic compilation and dynamic optimization had also been around for quite a number of years. So, both the Java and Mono implementors had the benefit of hindsight. In fact, Sun had many of the world experts on these subjects in-house when they started Java.
I haven't seen any decisive language feature that gives C# any edge over Java. Most of it is syntactic sugar which is nice to have but not really that important if you have code completion & refactoring tools.
No, it is not just syntactic sugar: the JVM just lacks crucial features. Try defining a class in Java that is equivalent to C++ "complex", then implement a complex FFT with it (complex 2D FFT is even worse). Not only is the Java syntax painful, the performance sucks, too.
Yes they do. It's their damned airport. Don't like it? Try another airport. Can't find one to accomodate your needs? Don't fly.
Airports are not completely private businesses; usually, they have received and continue to receive lots of public support, both in money and in kind. They have received that with the understanding that they provide everybody with good service. They can't just decide to do things that keep away a significant fraction of their customers.
Even if they were completely private businesses, there are still lots of things they cannot do. For example, even if you are the owner of a restaurant, you still may not be permitted to smoke on your premises during business hours.
Many modders leave out shielding, and that plays havoc with amateur radio and other kinds of transmissions.
It's no surprise Sun's JDK seems to perform the best out of the other versions. They know Java the best since they created it,
The techniques necessary for compiling Java well have largely been around since long before Sun even released Java. Sun has no special, secret knowledge there, they have just been hacking on their implementation longer than anybody else. If anything, Sun's progress on Java has been pokey.
Sun Java is probably pretty close to what is theoretically possible, so they will largely be standing still, while other implementations will keep improving until they also hit the limit. In a couple of years, you can expect that all actively developed JVM and CLR implementations will have roughly the same performance on comparable code.
What these benchmarks suggest to me is that C#/.NET is fully competitive in terms of performance with the best Java implementations, and C#/Mono is good enough for most work. In fact, given the additional language features of C#, it may well be easier to write fast code for many compute-intensive problems using Mono than Java today.
The level of performance of Mono is even more impressive given how young the project is; at the same point in time in its evolution, Sun had barely managed to produce a JIT compiler.
You'd think that natively compiled and optimized code would be faster than an interpreter. I guess there's optimization work to be done in gcjlib.
The JIT inside a Java or CLR (C#) runtime has more information available to it for optimization. That means that, ultimately, it should be able to do better than gcj on long-running, compute-intensive jobs.
Well, it's the old "blame the user" attitude. When a group like ITAA makes such a statement, it's not surprising that so much information technology still is so crappy.
In different words, you shouldn't even have to read a user manual to use a modern desktop application. And you certainly shouldn't have to read a user manual in order to vote.
"Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before and it has always been due to human error."
However, the period roughly 10 years ago was the time the original poster was talking about - no real competition for OpenStep existed
Already in the late 80's, you could get several practical Smalltalk implementations (e.g., ParcPlace, Smalltalk/V, Instantiations), and they were widely used commercially. VisualAge came out around 1993/4.
If you wanted something natively compiled, around that time, you could also get high quality native compilers for a variety of object-oriented languages with garbage collection, including Modula-3, Eiffel, and Oberon. Simula-67 even went back as far as the 1960's. If NeXT had wanted a more dynamic object model than those languages already had, the required additions would have been small compared to the changes NeXT made to Objective-C.
I think NeXTStep just represented a slightly different compromise between using standards accepted by industry (C, UNIX) and what was commercially available state of the art. It was so close to what regular "Joe programmers" felt comfortable with that they "got it", but awed them with a few small bits that were thrown in (GUI builder, nice graphical design, Objective-C). So, I think NeXT was a brilliant marketing compromise (and a good implementation), but technologically, I think it has never been state of the art.
I didn't see any peer reviewed publications by Art Robinson; where are they?
And, yes, solar variation has an influence on temperatures, and climatologists are aware of it. See Solar variation accounts for less than half of global warming in 20th Century, UA geoscientist finds, for example. And here.
NeXT is based on the Mach kernel, not BSD. It happens to implement the BSD APIs via some BSD code.
First of all, Objective-C is not Smalltalk--not even close. The C heritage of ObjC means you have to deal with pointers, type errors, casts, , compilers, header files, and all that.
Second, the ObjC object model is nothing unusual anymore: people can try it out much more easily using Python or Ruby. And with the bindings of those languages to grk, wx, and qt, people can do mainstream gui programming with languages and environments that are altogether a lot nicer without leaving linux or gnome.
The funny thing is that people who actually experienced the generation of GUIs before NeXT might say the same thing about NeXT. I mean, you have to deal with C, memory allocation, and pointers? You can't inspect and change the code of a running program? Your GUI runs in a separate process and needs to be programmed in a different language? You can have type errors in your running system without anything flagging an error? What kind of backwards system is that?
I agree to the degree that C++ has always been a bad choice for GUI development. But between ObjectiveC/Cocoa and Java/Eclipse or Python/Gtk+, I hardly see what advantages you think ObjectiveC/Cocoa brings. And, yes, I have tried out Apple's IDE.
What was even more amazing was using Smalltalk a decade before NeXT. NeXT was modeled on Smalltalk, but relative to Smalltalk, even NeXT seemed cumbersome.
And the guys who brought you Smalltalk are bringing you more neat stuff; see the story on OpenCroquet earlier.
Jobs did a great job selecting, integrating, and marketing. But what did he or his companies actually invent?
Maybe it's because Linux developers have other choices: PyGnome, wxPython, Java+SWT, Eclipse, ...
Calling me an Apple advocate
/. and the OS of the drivers released, I'd say it brings something to the table.
I didn't call you anything. Macintosh advocates in general regularly post all sorts of Apple trivia to Linux and X11 discussions. And Macintosh advocates in general make all sorts of claims about Linux and X11 in Apple forums on Slashdot. And that's OK: that's what Slashdot is about.
I don't inject Apple topics into Linux discussions, and I don't appreciate others injecting Linux into Apple related discussions.
Then you're on the wrong site. This is "News for Nerds", not "Macintosh driver central", "Macintosh advocacy", or "News for Apple Fans"; there are enough other sites and mailing lists where Apple fans can retreat in isolation, without the inconvenience of having their world view challenged.
No such question was asked. No question was asked at all. It sounded a lot like complaining about a company's policies in the wrong venue.
Slashdot is, in fact, exactly the right site to discuss Apple policies and contrast them with open source because this is a site where lots of people from all sorts of different backgrounds come. If not here, where else? What other site is there where people knowledgeable in different platforms can discuss these issues? The only problem is that people like you are quick to brand anybody as a "Troll" when you don't want to hear what they have to say.
If you want to know what Infiniband brings to the table of Linux supercomputing, the answer may be nothing, for reasons cited. But for what it brings to OS X, the topic of the article, the section of
I want to know what Macintosh+Infiniband brings to the table of supercomputing in general, relative to, say, Linux+Myrinet. We know that it may give you 20% lower latency in some caes if one post is to be believed, but little else, and even that didn't talk about cost/benefit. The matter seems to be settled in your mind, given your put-down of Linux-based cluster, but, sorry, you haven't made your case.
In case you still do not realize it the analysis of said data is the starting point for the majority of science on global warming, and used as scientific proof for it
The paper in question has indeed been cited over 300 times, so it is important, but it contains several results. We don't now which of those results (if any) citing papers rely on and if the conclusions of those papers would be affected if the McIntyre and McKitrick criticism is accurate. Muller fails to make that analysis in his paper, and he also is not an expert on statistical data analysis.
What we are left with is that both points of view have been published, people have had the opportunity to make up their minds, and it is pretty clear what the mainstream scientific opinion on the matter is, at least for now.
Squid farther north bacause of CO2 and global warming = junk science
No, that's not called "junk science", it's called a hypothesis (and a pretty plausible one). If you have an alternative hypothesis, please advance it. But simply ignoring the observation of species outside their normal range without explanation, that really is "junk science".
I haven't said that the cycles happen by themselves,
No, you didn't say that, you simply ignored the question of causation at all. We currently have one plausible hypothesis for why global warming is happening. When you come up with an alternative, please share it.
than a major flaw being discovered in some of the most used data in scientifically arguing for global warming
The article about PCA isn't about a flaw in the data, it's about a flaw in one particular statistical analysis.