So you are reliant upon your governmennt to tell you what is right and what is wrong? Personally I separate what is moral from what is legal.
No, you just made a bad analogy.
In any case, the UN decided that Iraq cannot legally possess WMD. So the analogy holds.
You wrote:
No it's not. Other countries do have WMD. So it's not absurd to accuse them [other countries] of it.
You statement wasn't specifically about Iraq, you made a general statement about "other countries". I just pointed out that it was indeed absurd for the US to accuse "other countries" merely because those countries had WMDs. Most countries can legitimately have WMDs, whether the US likes it or not.
Whether it is absurd for the US to accuse Iraq specifically is a separate issue.
The problem with your statement and your attitude is that you are implying it's OK for the US to have WMDs but nobody else. And most people in the world would probably disagree with you.
If the attacker already had access to the machine then a keylogger would be able to take care of grabbing the passphrase.
Or maybe the attacker didn't have access to the machine and hacked into the repository through some other means.
Security isn't black-or-white. Adding signatures adds another checkpoint. It doesn't make things 100% secure, but it makes it makes some attacks impossible and it makes other attacks a lot more work. That's a good thing.
Linus pointed out a benefit to using BK: even if the official BK repository were changed, he doesn't pull from it
That's a function of how he uses the repository, not what revision control system he is using. If he wanted to set up a one-way mirror for his repository, he could do that with CVS as well, and depending on what software he uses for mirroring, it would also complain about unexpected changes.
It's a clever backdoor, and might have gone unnoticed, if not for those those good automated checks in the BitKeeper-to-CVS gateway. Notice that the particular coding style is a common C gotcha (using "=", assignment, instead of "==", comparison).
I suppose it seems "clever" if you haven't seen it before. But there have been quite a number of backdoors like that over the years using the same "trick" of changing an "==" into a "=" (e.g., in the VAX BSD kernel and old versions of sendmail). In the past, they have just gotten fixed without much noise.
It's actually kind of disconcerting that the Linux developers aren't scanning for this regularly; a simple C style checker that finds assignments inside "if" would do.
Vector graphics has been in many window systems. Neither Apple, nor KDE, nor Gnome, nor Microsoft, nor Macromedia have "innovated" in this area.
Thankfully, KDE and Gnome just adopted the W3C standard, SVG. Apple and Microsoft, of course, both have to do their own, messy, proprietary thing, DisplayPDF in the case of Apple and God-knows-what in the case of Microsoft.
Yeah, and about 15 years earlier by another company. That's in addition to the zillions of vector graphics APIs and toolkits available for lots of other window systems (including X) even before that.
Of course, historically, vector graphics was what most high-end graphics was done in (using hardware). Then, a bunch of smart folks at Xerox, Stanford, SRI, MIT, and AT&T came up with bitmapped user interfaces. Of course, Apple copied that, too, and then their marketing department managed to get people to believe that they invented it all.
Maybe RedHat isn't ready for the desktop. But my mother thoroughly approves of her Debian/Gnome/Linux desktop. Yep, it's obscure and buggy at times, but oh-so-much-less than Macintosh or Windows. And at least Debian comes with lots of games preinstalled, while with Mac and Windows they are $10-$50 each, with obscure download and installation procedures.
So is it absurd for someone to "accuse" a convicted felon of possessing a handgun if he himself has a handgun?
If the convicted felon can legally carry a handgun, then, yes, it is absurd. If the convicted felon cannot legally carry a handgun, then it is a valid legal accusation.
In the latter case, the analogy to other nations is not valid. While US politicians condemn anybody they don't like as "criminals", internationally, almost all nations are sovereign and can develop whatever they like. The US is in a particularly bad position to criticize other nations because of the recent moves the US has made against arms control and unilaterally getting out of international treaties. And many of the US's activities overseas have been highly illegal, such as the toppling of the democratically elected government of Iran or attempts to assassinate foreign leaders the US happened not to like.
Historical climate data is pretty much irrelevant to the question of whether we can safely emit greenhouse gasses. Emission of greenhouse gasses would be unsafe even if we were currently heading into a new ice age. Its lack of safety doesn't derive from historical data or complex climate modeling, but is a consequence of elementary thermodynamics: if you keep pumping those gasses into the atmosphere, sooner or later you are going to get large climate changes. The only thing that people are still debating is whether it's already happening or whether it's going to happen a few decades from now.
Kyoto may or may not have been justified to the public using historical climate records, but that doesn't matter. Kyoto is a meaningless band-aid anyway.
Why supposedly intelligent people keep being so hostile towards what is pretty much the consensus opinion of climatologists is beyond me. There seems to be some irrationally libertarian world view going around that basically states "apres nous le deluge", in this case, quite literally, a world view that was as self-descructive during Louis XIV's reign as it is today. And what is beyond me, too, is that people who want to engage in absolutely unprecendent experiments with our global climate and health call themselves "conservatives"--maybe they are just talking about their power and bank accounts when they are talking about being "conservative".
The US accusing some other country of having WMD is really quite absurd.
No it's not. Other countries do have WMD. So it's not absurd to accuse them of it.
Sure it is. The term "accuse" doesn't just mean "state impartially", it carries a notion of ethical or legal disapproval that both the speaker and the audience share.
It is not absurd to state that other countries have WMDs, it is not even absurd for the US to consider that state of affairs undesirable. However, it is absurd for the US to imply that everybody should agree that there is something ethically or legally wrong with that.
For example, I may "accuse" you of driving a car and thereby destroying the environment, but chances are that you and the majority of other Americans would consider my "accusation" absurd because I drive a car, too, and because you don't see anything wrong with driving a car in the first place.
Well, it's an uncertain world, and India, Pakistan, and China have as much interest in defending themselves as the US. And the US has done nothing to reduce the need for WMDs in this world--quite to the contrary. That's makes it truly absurd for the US to "accuse" other nations for having WMDs.
A nation like the Vatican, however, does have the moral authority to accuse other nations of having WMDs because they don't. And such nations generally accuse the US as much as they accuse other nations.
Umm, they just diappeared? You do agree that there were WMD at one time, right? Or was that just part of a global conspiracy against poor innocent muslims?
They were there. The UN was tasked with finding them. Then the Bush administration came in, said the UN weapons inspectors were all a bunch of idiots, and that if they went in with the US military, they could resolve this in no time.
The net result of the Bush action? Lots of dead innocent civilians, far more hostility towards the US in the region, broken diplomatic relationships with Europe, and whatever WMDs there may have been left after the UN inspections have scattered all over the Middle East. Oh, and let's not forget hundreds of billions of your and my tax dollars wasted while education and health care are falling apart at home.
So, to be clear: almost everybody, including the UN, agreed that there were some WMDs left before the US invaded. But Bush has made it nearly impossible to track them down now and killed lots of people and wasted lots of money in the process.
Same here: I just buy stuff on DVD or tape, or rent it. I get to choose what I see, I don't have to put up with ads, and I broadcast or cable rarely carries any interesting shows anyway.
That was not because of the software, but because of dragging performance. The symbolics hardware development basically couldn't keep up with the rest of the world,
Performance was part of the problem. But performance itself was an indication that the Symbolics system had fundamental design flaws.
Yet from a purely software-technical PoV, the Symbolics Lisp OS was decades ahead of its time in many respects.
I disagree. I think the Smalltalk folks pioneered most of the symbolic, dynamic, and interactive ideas that have had lasting impact--Smalltalk really was decades ahead of its time. Smalltalk's programming environment had a consistent paradigm and design.
The MIT Lisp machines took bits and pieces of DEC-10 history (Emacs, text files), a command-line oriented programming language, and lots of money to build a somewhat unharmonious, hugely expensive, and slow combination. For years, the Lisp machines didn't even have overlapping windows or real window management. Work on the Lisp machines also generalized some things (CLOS, presentation types), but compared to the ideas that Smalltalk pioneered, those seem more like historical oddities in terms of their long-term impact.
Between the two, I think the Symbolics was perhaps the more practical system (but we are talking degrees here), but I think Smalltalk should get most of the glory for the lasting and truly important innovations in dynamic languages and programming environments.
IBM's C,C++ and Fortran compilers are available for the Mac - download them if you want.
Yes, and your point is what? If those compilers were useful and recommended for Macintosh development, Apple would be using them for their own SPEC benchmarks to make their machines look better.
Besides, even using IBM's figures, the Apples are still not competitive--the PPC970 at 2GHz at best pulls even with current Opterons, and both IBM's and Apple's offerings are more expensive than 1U dual-Opteron rack mounts.
http://macbuyersguide.com/editorials/editorial-p pc 970.htm
Again, the G5 isn't a disaster--it's state-of-the-art performance, but Apple's machines are still a bit worse than others in terms of bang-for-the-buck. And buying a few thousand desktop machines is still not the best way of building a cluster. VT has a useful machine, but it's not the most cost-effective they could have gotten.
And since a PPC970 is a PPC970 whether it is in a Mac or an IBM it does make sense to consider IBM's SPEC results.
I'm sorry, but you are completely naive. System performance, even on purely numerical tasks, can be influenced greatly by all sorts of things other than the CPU.
People who propose systems like MSH want to iterate over files and all that good stuff using object oriented scripting features. I'm saying: that turns out not to be very useful in practice. The fact that psh happens to be able to emulate sh behavior is completely besides the point.
It is 'conception' that is basically the problem - you don't know what it is like to use a real scripting language from the command line so its strengths are not apparent.
Sure I do: I have used psh at times, and I have used Smalltalk and Lisp, which are much better at this kind of integration and scripting than even Perl or MSH.
But that kind of design of interactive shells, something that integrates full programming, has always lost out in the real world.
In fact, UNIX has a much better answer: rather than trying to force everything into the same address space, it provides facilities (environment variables, etc.) that let software written in multiple different "little languages" co-exist as if they were all part of a single, unified scripting environment.
If you want to use Perl in a script or pipe, just say "... | perl -e '...' |...". And if you want to invoke other things from Perl, you can, obviously, use "open", "system", "`...`", and all those other facilities.
The UNIX approach is great; you should give it a try sometimes. MSH and psh, on the other hand, are Microsoft-thinking: obvious, gimmicky, and not all that good in the end.
If you want something like MSH, just use Python, Jython, Perl, or psh. You see, people have been trying to improve on bash/ksh for years. But none of those have caught on because the sh-family of shells is actually very well adapted to the day-to-day needs of its users.
The shell has been one of the most lacking areas under Windows. I don't know how many times I've dropped into Cygwin or, before that, wasted time writing little C apps just to do basic bulk renaming operations and the likes.
You don't even need all of Cygwin. Just use Perl for that sort of thing. It works great under Windows and even lets you access lots of native Windows functionality.
I don't know why more people don't actively pursue a modern language for the shell interface. sh script syntax is tortorous. So much easier and maintainable to write perl scripts. So why not use perl from the command line??
Yes, you don't know. But think for a moment: people have had Perl-like languages since the 1960's. Do you really think you or Microsoft are the first to think that using an object-oriented scripting language is a good idea?
The reason why people use sh syntax is because it is enormously effective. Try expressing something like:
find . -type f | xargs grep -il foo
in Perl or some other scripting language.
Of course, many people who complain about sh syntax really just don't know how to use it.
For interactive use by skilled users and many scripting tasks, bash/ksh is unbeatable. And for the kinds of scripts where Perl makes sense--you can simply use Perl.
This would basically trump anything msh could muster and also provide the entire universe of CPAN to the shell.
Yes, psh is a better version of what msh is trying to achieve. But, you know, even that's nowhere near good enough to dethrone bash/ksh.
While Microsoft folks will doubtlessly be wowed by the features and flexibility, it's important to keep in mind that generations of programmers have tried to improve on the UNIX shell. The Symbolics, for example, had a spectactularly powerful, general, and user-friendly command shell, including what Microsoft calls "Intellisense" (yes, that was about 20 years ago) and completion, the ability to manipulate objects with commands, full integration with the preferred programming language (Lisp rather than.NET), and the ability to have objects render themselves in the transcript window graphically, and interact with the mouse.
So, for example, you could type "sd" (which would complete to "Show Directory" by itself, as would other unambiguous abbreviations), you'd get a styled directory listing, and you could click on files to select them, open them, and do other things.
Did it win? No. What people need to understand is that not every improvement is the right thing for the real world. Ksh/bash has hit some sort of sweet spot. People don't use it because it's the technically best, they use it because it has the right set of features.
Opterons weren't out except as parts that would have to be assembled and that wasn't going to fly for their requirements.
Dual Opteron servers were shipping months before Apple's G5s, so that argument is bogus.
Can you imagine the risk of having AMD declare your assembly methods out of spec and refuse to replace any downed processors? This is a multi-million dollar cluster. They needed a chip and a chassis and they wanted it right then.
Yes, and that's why PC hardware makes so much more sense: Opteron rackmounts for compute clusters don't require any hardware installation at all--they come preassembled and you get one-stop service from whoever you bought it from.
Instead, VT chose machines that weren't designed for clusters and required manual installation of additional hardware by VT students. Can you imagine how much finger pointing there will be when those machines fail and Apple claims that it's due to the PCI cards installed at VT?
Note that a dual Opteron is probably going to cost you less than a dual G5.
What more do you need? Faster systems, cheaper total cost, and slick looking cases.
What about good bang for the buck, choice of vendors, fully assembled hardware (including, if you choose, low-latency networking cards), on-board built-in dual Gigabit, preloaded Linux, and 1U rack mounts?
The G5 is about the same speed as a high-end P4 or Opteron in standard benchmarks like SPEC (a little slower actually), and the G5 machines with list or academic price are both more expensive and more expensive to deploy (since they require manual hardware configuration, don't come in rack mounts, and take up lots of space). Ergo, the machine can't offer the best price/performance ratio. Replicating a worse price/performance ratio by 2000 machines doesn't make it any better.
Don't get me wrong: the VT Mac cluster does not sound like a disaster--they paid a bit more and they have to live with a number of maintenance and programming hassles. But it is a fast machine, and the boxes do look pretty.
As for "the biggest switcher", I mean, unless it's a complete disaster, what do you expect someone who just spent millions of other people's money to say? "Well, we could have done better but it'll do?" I don't think so. He's going to try to make his decision look as good as it possibly can.
And that's why they have been hacking furiously in assembly language for the last few months trying to beat Intel/AMD on at least one benchmark and make the cluster look good.
Well, the burden of "clear and convincing prior art searches" would be on the patent holder and would require dilligent, well-documented, convincing prior art searches. It might also require creating a more definitive on-line searchable database of prior art, perhaps maintained by the USPTO, to which people can submit information.
Note that many companies already do this as part of their patent application process; but there are bad apples who not only don't do this but explicitly try to avoid doing them.
Slashdot Mirror
Lots of phones, lots of phone headsets, and many PDAs have Bluetooth (more than 802.11).
And if you search for "MP3" and "Bluetooth" on Google, you'll find plenty of products that combine MP3 players and Bluetooth headsets.
Former registry content will now be distributed across directories into a new file type). Now a Flash-a-like product as well.
Oh, goodie, just like UNIX. And it only took them, what, 15 years to realize what a lousy idea the registry was?
You mean like wireless headsets and mobile Internet access (through your cell phone)?
But maybe you know of 802.11 headsets and 802.11 cell phones?
Even if they existed, BT gets far better battery life.
No, you just made a bad analogy.
In any case, the UN decided that Iraq cannot legally possess WMD. So the analogy holds.
You wrote:
You statement wasn't specifically about Iraq, you made a general statement about "other countries". I just pointed out that it was indeed absurd for the US to accuse "other countries" merely because those countries had WMDs. Most countries can legitimately have WMDs, whether the US likes it or not.
Whether it is absurd for the US to accuse Iraq specifically is a separate issue.
The problem with your statement and your attitude is that you are implying it's OK for the US to have WMDs but nobody else. And most people in the world would probably disagree with you.
If the attacker already had access to the machine then a keylogger would be able to take care of grabbing the passphrase.
Or maybe the attacker didn't have access to the machine and hacked into the repository through some other means.
Security isn't black-or-white. Adding signatures adds another checkpoint. It doesn't make things 100% secure, but it makes it makes some attacks impossible and it makes other attacks a lot more work. That's a good thing.
Linus pointed out a benefit to using BK: even if the official BK repository were changed, he doesn't pull from it
That's a function of how he uses the repository, not what revision control system he is using. If he wanted to set up a one-way mirror for his repository, he could do that with CVS as well, and depending on what software he uses for mirroring, it would also complain about unexpected changes.
It's a clever backdoor, and might have gone unnoticed, if not for those those good automated checks in the BitKeeper-to-CVS gateway. Notice that the particular coding style is a common C gotcha (using "=", assignment, instead of "==", comparison).
I suppose it seems "clever" if you haven't seen it before. But there have been quite a number of backdoors like that over the years using the same "trick" of changing an "==" into a "=" (e.g., in the VAX BSD kernel and old versions of sendmail). In the past, they have just gotten fixed without much noise.
It's actually kind of disconcerting that the Linux developers aren't scanning for this regularly; a simple C style checker that finds assignments inside "if" would do.
Vector graphics has been in many window systems. Neither Apple, nor KDE, nor Gnome, nor Microsoft, nor Macromedia have "innovated" in this area.
Thankfully, KDE and Gnome just adopted the W3C standard, SVG. Apple and Microsoft, of course, both have to do their own, messy, proprietary thing, DisplayPDF in the case of Apple and God-knows-what in the case of Microsoft.
Yeah, and about 15 years earlier by another company. That's in addition to the zillions of vector graphics APIs and toolkits available for lots of other window systems (including X) even before that.
Of course, historically, vector graphics was what most high-end graphics was done in (using hardware). Then, a bunch of smart folks at Xerox, Stanford, SRI, MIT, and AT&T came up with bitmapped user interfaces. Of course, Apple copied that, too, and then their marketing department managed to get people to believe that they invented it all.
Maybe RedHat isn't ready for the desktop. But my mother thoroughly approves of her Debian/Gnome/Linux desktop. Yep, it's obscure and buggy at times, but oh-so-much-less than Macintosh or Windows. And at least Debian comes with lots of games preinstalled, while with Mac and Windows they are $10-$50 each, with obscure download and installation procedures.
So is it absurd for someone to "accuse" a convicted felon of possessing a handgun if he himself has a handgun?
If the convicted felon can legally carry a handgun, then, yes, it is absurd. If the convicted felon cannot legally carry a handgun, then it is a valid legal accusation.
In the latter case, the analogy to other nations is not valid. While US politicians condemn anybody they don't like as "criminals", internationally, almost all nations are sovereign and can develop whatever they like. The US is in a particularly bad position to criticize other nations because of the recent moves the US has made against arms control and unilaterally getting out of international treaties. And many of the US's activities overseas have been highly illegal, such as the toppling of the democratically elected government of Iran or attempts to assassinate foreign leaders the US happened not to like.
Historical climate data is pretty much irrelevant to the question of whether we can safely emit greenhouse gasses. Emission of greenhouse gasses would be unsafe even if we were currently heading into a new ice age. Its lack of safety doesn't derive from historical data or complex climate modeling, but is a consequence of elementary thermodynamics: if you keep pumping those gasses into the atmosphere, sooner or later you are going to get large climate changes. The only thing that people are still debating is whether it's already happening or whether it's going to happen a few decades from now.
Kyoto may or may not have been justified to the public using historical climate records, but that doesn't matter. Kyoto is a meaningless band-aid anyway.
Why supposedly intelligent people keep being so hostile towards what is pretty much the consensus opinion of climatologists is beyond me. There seems to be some irrationally libertarian world view going around that basically states "apres nous le deluge", in this case, quite literally, a world view that was as self-descructive during Louis XIV's reign as it is today. And what is beyond me, too, is that people who want to engage in absolutely unprecendent experiments with our global climate and health call themselves "conservatives"--maybe they are just talking about their power and bank accounts when they are talking about being "conservative".
No it's not. Other countries do have WMD. So it's not absurd to accuse them of it.
Sure it is. The term "accuse" doesn't just mean "state impartially", it carries a notion of ethical or legal disapproval that both the speaker and the audience share.
It is not absurd to state that other countries have WMDs, it is not even absurd for the US to consider that state of affairs undesirable. However, it is absurd for the US to imply that everybody should agree that there is something ethically or legally wrong with that.
For example, I may "accuse" you of driving a car and thereby destroying the environment, but chances are that you and the majority of other Americans would consider my "accusation" absurd because I drive a car, too, and because you don't see anything wrong with driving a car in the first place.
Well, it's an uncertain world, and India, Pakistan, and China have as much interest in defending themselves as the US. And the US has done nothing to reduce the need for WMDs in this world--quite to the contrary. That's makes it truly absurd for the US to "accuse" other nations for having WMDs.
A nation like the Vatican, however, does have the moral authority to accuse other nations of having WMDs because they don't. And such nations generally accuse the US as much as they accuse other nations.
Umm, they just diappeared? You do agree that there were WMD at one time, right? Or was that just part of a global conspiracy against poor innocent muslims?
They were there. The UN was tasked with finding them. Then the Bush administration came in, said the UN weapons inspectors were all a bunch of idiots, and that if they went in with the US military, they could resolve this in no time.
The net result of the Bush action? Lots of dead innocent civilians, far more hostility towards the US in the region, broken diplomatic relationships with Europe, and whatever WMDs there may have been left after the UN inspections have scattered all over the Middle East. Oh, and let's not forget hundreds of billions of your and my tax dollars wasted while education and health care are falling apart at home.
So, to be clear: almost everybody, including the UN, agreed that there were some WMDs left before the US invaded. But Bush has made it nearly impossible to track them down now and killed lots of people and wasted lots of money in the process.
Same here: I just buy stuff on DVD or tape, or rent it. I get to choose what I see, I don't have to put up with ads, and I broadcast or cable rarely carries any interesting shows anyway.
That was not because of the software, but because of dragging performance. The symbolics hardware development basically couldn't keep up with the rest of the world,
Performance was part of the problem. But performance itself was an indication that the Symbolics system had fundamental design flaws.
Yet from a purely software-technical PoV, the Symbolics Lisp OS was decades ahead of its time in many respects.
I disagree. I think the Smalltalk folks pioneered most of the symbolic, dynamic, and interactive ideas that have had lasting impact--Smalltalk really was decades ahead of its time. Smalltalk's programming environment had a consistent paradigm and design.
The MIT Lisp machines took bits and pieces of DEC-10 history (Emacs, text files), a command-line oriented programming language, and lots of money to build a somewhat unharmonious, hugely expensive, and slow combination. For years, the Lisp machines didn't even have overlapping windows or real window management. Work on the Lisp machines also generalized some things (CLOS, presentation types), but compared to the ideas that Smalltalk pioneered, those seem more like historical oddities in terms of their long-term impact.
Between the two, I think the Symbolics was perhaps the more practical system (but we are talking degrees here), but I think Smalltalk should get most of the glory for the lasting and truly important innovations in dynamic languages and programming environments.
IBM's C,C++ and Fortran compilers are available for the Mac - download them if you want.
p pc 970.htm
Yes, and your point is what? If those compilers were useful and recommended for Macintosh development, Apple would be using them for their own SPEC benchmarks to make their machines look better.
Besides, even using IBM's figures, the Apples are still not competitive--the PPC970 at 2GHz at best pulls even with current Opterons, and both IBM's and Apple's offerings are more expensive than 1U dual-Opteron rack mounts.
http://macbuyersguide.com/editorials/editorial-
Again, the G5 isn't a disaster--it's state-of-the-art performance, but Apple's machines are still a bit worse than others in terms of bang-for-the-buck. And buying a few thousand desktop machines is still not the best way of building a cluster. VT has a useful machine, but it's not the most cost-effective they could have gotten.
And since a PPC970 is a PPC970 whether it is in a Mac or an IBM it does make sense to consider IBM's SPEC results.
I'm sorry, but you are completely naive. System performance, even on purely numerical tasks, can be influenced greatly by all sorts of things other than the CPU.
in psh [...] works just fine.
...". And if you want to invoke other things from Perl, you can, obviously, use "open", "system", "`...`", and all those other facilities.
People who propose systems like MSH want to iterate over files and all that good stuff using object oriented scripting features. I'm saying: that turns out not to be very useful in practice. The fact that psh happens to be able to emulate sh behavior is completely besides the point.
It is 'conception' that is basically the problem - you don't know what it is like to use a real scripting language from the command line so its strengths are not apparent.
Sure I do: I have used psh at times, and I have used Smalltalk and Lisp, which are much better at this kind of integration and scripting than even Perl or MSH.
But that kind of design of interactive shells, something that integrates full programming, has always lost out in the real world.
In fact, UNIX has a much better answer: rather than trying to force everything into the same address space, it provides facilities (environment variables, etc.) that let software written in multiple different "little languages" co-exist as if they were all part of a single, unified scripting environment.
If you want to use Perl in a script or pipe, just say "... | perl -e '...' |
The UNIX approach is great; you should give it a try sometimes. MSH and psh, on the other hand, are Microsoft-thinking: obvious, gimmicky, and not all that good in the end.
If you want something like MSH, just use Python, Jython, Perl, or psh. You see, people have been trying to improve on bash/ksh for years. But none of those have caught on because the sh-family of shells is actually very well adapted to the day-to-day needs of its users.
The shell has been one of the most lacking areas under Windows. I don't know how many times I've dropped into Cygwin or, before that, wasted time writing little C apps just to do basic bulk renaming operations and the likes.
You don't even need all of Cygwin. Just use Perl for that sort of thing. It works great under Windows and even lets you access lots of native Windows functionality.
Yes, you don't know. But think for a moment: people have had Perl-like languages since the 1960's. Do you really think you or Microsoft are the first to think that using an object-oriented scripting language is a good idea?
The reason why people use sh syntax is because it is enormously effective. Try expressing something like:in Perl or some other scripting language.
Of course, many people who complain about sh syntax really just don't know how to use it.
For interactive use by skilled users and many scripting tasks, bash/ksh is unbeatable. And for the kinds of scripts where Perl makes sense--you can simply use Perl.
This would basically trump anything msh could muster and also provide the entire universe of CPAN to the shell.
Yes, psh is a better version of what msh is trying to achieve. But, you know, even that's nowhere near good enough to dethrone bash/ksh.
While Microsoft folks will doubtlessly be wowed by the features and flexibility, it's important to keep in mind that generations of programmers have tried to improve on the UNIX shell. The Symbolics, for example, had a spectactularly powerful, general, and user-friendly command shell, including what Microsoft calls "Intellisense" (yes, that was about 20 years ago) and completion, the ability to manipulate objects with commands, full integration with the preferred programming language (Lisp rather than .NET), and the ability to have objects render themselves in the transcript window graphically, and interact with the mouse.
So, for example, you could type "sd" (which would complete to "Show Directory" by itself, as would other unambiguous abbreviations), you'd get a styled directory listing, and you could click on files to select them, open them, and do other things.
Did it win? No. What people need to understand is that not every improvement is the right thing for the real world. Ksh/bash has hit some sort of sweet spot. People don't use it because it's the technically best, they use it because it has the right set of features.
Opterons weren't out except as parts that would have to be assembled and that wasn't going to fly for their requirements.
Dual Opteron servers were shipping months before Apple's G5s, so that argument is bogus.
Can you imagine the risk of having AMD declare your assembly methods out of spec and refuse to replace any downed processors? This is a multi-million dollar cluster. They needed a chip and a chassis and they wanted it right then.
Yes, and that's why PC hardware makes so much more sense: Opteron rackmounts for compute clusters don't require any hardware installation at all--they come preassembled and you get one-stop service from whoever you bought it from.
Instead, VT chose machines that weren't designed for clusters and required manual installation of additional hardware by VT students. Can you imagine how much finger pointing there will be when those machines fail and Apple claims that it's due to the PCI cards installed at VT?
Well, he doesn't know what he is talking about. Here are Apple's own claims for SPEC performance:Here are the results for Itanium from SPEC.org:The 1.5GHz Itanium is almost 2.5 times faster than the G5.
Here are the SPEC results for the Opterons:Note that a dual Opteron is probably going to cost you less than a dual G5.
What more do you need? Faster systems, cheaper total cost, and slick looking cases.
What about good bang for the buck, choice of vendors, fully assembled hardware (including, if you choose, low-latency networking cards), on-board built-in dual Gigabit, preloaded Linux, and 1U rack mounts?
References:
http://www.apple.com/g5processor/
http://www.s
The G5 is about the same speed as a high-end P4 or Opteron in standard benchmarks like SPEC (a little slower actually), and the G5 machines with list or academic price are both more expensive and more expensive to deploy (since they require manual hardware configuration, don't come in rack mounts, and take up lots of space). Ergo, the machine can't offer the best price/performance ratio. Replicating a worse price/performance ratio by 2000 machines doesn't make it any better.
Don't get me wrong: the VT Mac cluster does not sound like a disaster--they paid a bit more and they have to live with a number of maintenance and programming hassles. But it is a fast machine, and the boxes do look pretty.
As for "the biggest switcher", I mean, unless it's a complete disaster, what do you expect someone who just spent millions of other people's money to say? "Well, we could have done better but it'll do?" I don't think so. He's going to try to make his decision look as good as it possibly can.
And that's why they have been hacking furiously in assembly language for the last few months trying to beat Intel/AMD on at least one benchmark and make the cluster look good.
Well, the burden of "clear and convincing prior art searches" would be on the patent holder and would require dilligent, well-documented, convincing prior art searches. It might also require creating a more definitive on-line searchable database of prior art, perhaps maintained by the USPTO, to which people can submit information.
Note that many companies already do this as part of their patent application process; but there are bad apples who not only don't do this but explicitly try to avoid doing them.