That software is software you install by blowing away what the Zaurus ships with. So, no, the Zaurus doesn't "use" that software, but it can be put on there by someone sufficiently dedicated. But, then, you can do the same with a dozen or so PocketPC handhelds.
It is hard to exaggerate the tremendous value this search function has. Amazon has taken a huge leap foward in providing access to information. Really.
What's the point of providing this kind of access if you still need to pay way too much for content that comes in cumbersome form (DRM, books)?
I hope the authors' guild wins this one. The harder they make it for people to access their content, the easier it will be for truly free content to take over the market.
Bittorrent works for Linux because what is being distributed is free. Apple does have the equivalent for OSX, namely pirated CD's. It just happens not to mesh well with their business and licensing model.
The Zaurus runs the Linux kernel, but it uses a proprietary window system (TrollTech's Qtopia, vaguely like Qt). That means most Linux GUI development tools and applications won't work on it and you have to pay big bucks to TrollTech if you want to develop anything commercial.
As for running Palm applications, sure, the Palm emulator for Linux should run just fine on those things (in fact, I think it has been ported).
We have standards for that sort of thing. Presumably, the LG optical drives are standard ATAPI drives, not "Windows drives". If Linux destroys them with standard CD-ROM drivers, then it's a problem with the drives.
In fact, it's hard to see how any CD-ROM driver should be able to destroy any CD-ROM drive unless the drive has some kind of serious design flaw.
I don't know for sure, but if these black boxes operate as aircraft boxes do, I believe they will only store a certain time's worth of information.
The problem is that if they aren't regulated, there is no telling what they record. Car manufacturers may well find it convenient to record your entire driving history (maybe even GPS location) for marketing and product development purposes. They aren't even obligated to tell you. They might sell that data to lots of other companies, together with information about what kind of consumer you are, etc.
As I was saying: then make it more accurate for everybody then and make them mandatory. But the current situation, where people can select what kind of blackbox they get, is not acceptable because it encourages dangerous drivers not to have blackboxes, which is exactly what we don't want to happen.
I am not sure about American law but in Canada (where this took place) if you hit a person from behind and because they braked YOU are responsible. You did not keep a safe distance back from the other driver.
I seriously doubt that Canadian law is so rigid. There are many ways in which you can drive perfectly safely and you still end up rear-ending another car. For example, a car could squeeze into the space in front of you, creating an unsafe distance between them and you, and then its driver slams on the brakes; there is no way for you to avoid running into them, and the fault is clearly with them, first by taking away your safety zone and then by slamming on their brakes. In fact, I suspect that kind of accident is quite common.
But the situation is not analogous. First of all, you don't have much of a choice in whether you leave tire marks--it may happen or it may not. But you do have a choice in whether you drive a vehicle with a blackbox. Furthermore, different black-boxes record different kinds of information. Some may collect a lot of your driving history, not just a little information around the accident.
If this information is going to be used, then blackboxes should be mandatory and standardized.
The point is that, right now, only some drivers have them, and the people who don't have them can self-select. That's not a good situation.
What do you do, for example, in an accident where one car has a blackbox recorder and it shows the car was speeding, while the other car doesn't have a recorder? It may have been speeding much more, but that may be much harder to prove.
Furthermore, the information in these recorders can be used for other purposes: a private investigator can potentially find out how far you have been driving to your extramarital affair and what roads you have probably taken, and so can Mr. Ashcroft and his goons.
If these things are such a good idea, then politicians should make them mandatory. Of course, that proposal is probably going to be an uphill battle. But to stick only the drivers with this who don't know any better not to have one just isn't fair.
It isn't the police's right to know that you had the gas to the floor when you rear-ended the woman in front of you, killing her and her two kids
Just because you were speeding doesn't mean you are guilty in that case. The woman may have slammed on her brakes in the middle of the highway for no good reason, and the fact that you were 15 mph over the speed limit (like everybody else, her included, although her cheap and unsafe 1979 Honda CVC may not show that) may make little difference.
But because you have a blackbox and she doesn't, everybody is going to jump to conclusions. Like you did.
I'm not saying it should be mandated that these be installed in every single vehicle manufactured,
They should be mandatory if they are going to be used at all. Otherwise, people who know they are going to drive like maniacs will select cars without blackboxes while mostly law-abiding drivers will get screwed if they cross the line some time (emergency, etc.) and happen to have an accident.
Either make these things mandatory for everybody or don't permit them to be used as evidence at all. But using them as evidence in only some cases is really unfair.
That has led me to start looking around for a new pair of SLRs, one film, one digital. Ideally, I'd like to be able to have the same set of lenses that could be mounted on both a film & a digital camera body,
That means you really need to get a digital SLR with a full-frame sensor. But those are hugely expensive.
I wouldn't bother with the digital SLRs yet--I think they are overpriced and underperforming. Just get a high-end digital viewfinder and use the LCD when you need precise framing.
you can't rely on the on-line photo sites
on
Digital 35mm SLRs?
·
· Score: 3, Insightful
"Professional" reviewers of photographic equipment are almost always far too positive. Among other things, they usually depend on getting free loaners to review, and if they trashed a camera in a review, they might not get more free loaners in the future. Some of the digital cameras I have had have been real duds, yet they all received reasonably good reviews.
And then you have the analog traditionalist nuts, the photographic equivalent of the people who claim that vinyl and tubes are higher quality than CDs. You can have a 48Mpixel camera and they'll still claim that some random 35mm film beats it.
And what does it matter anyway? Digital is just different from analog. If you have the money, give it a try and see whether you like it. If you don't have the money, don't even get started.
Europe won't do it. Europe's creative drive is spent, gone, kaput. Japan can't even find a launch site that lets them get birds into orbit on schedule, and is experiencing an inverted population pyramid with all the lack of dynamism that implies. So's Europe, for that matter.
Well, at least China, India, Japan, and Europe haven't abandoned education, science, or engineering. The US educational system is in shambles and if the US didn't import a large fraction of its scientists and engineers, it wouldn't have much of a technology industry. In fact, without the influx of WWII refugees, the US would probably never have become a scientific or technological powerhouse.
I'd put my money on those other nations, rather than the US, to drive science and technology forward in the future.
It doesn't worry me that they'll go. It worries me that they'll use it as a way to expand the prestige of their dictatorship and extend their illegitimate powers. That would be good for nobody except the dictators.
What, pray tell, is "illegitimate" about the Chinese government? You may not like them and I may not like them, but that doesn't make them illegitimate. And while they aren't a democracy, they certainly aren't a dictatorship either.
In terms of quality, these digital SLRs are comparable to, or even better than, film. And given the cost of film, they are cost effective, too.
The problem is: they are not 35mm cameras, the sensor is smaller than 35mm film. That means that you get a focal length multiplier and are more limited on the wide angle side.
If you want a drop-in replacement for a 35mm camera, you need a full-frame digital SLR. Or, alternatively, get a non-SLR digital camera.
Unless we do something to upset the apple cart, we will be stuck with a couple people in orbit doing nothing and going nowhere. I do not find this an acceptable state of affairs. The people now running the show are some of the biggest obstacles to progress, and they have to be shoved out of the way.
Don't worry--when the costs have come down enough, China, India, Europe, Japan, and other nations will do that, no matter how cushy the relationship between the US government and large US aerospace contractors may be. In fact, China already has begun. Isn't international competition great?
Start a project that aims to develop extremely efficient programs designed to run very well with slow procs like these. Hell if you can web browse on a C64, this can be done.
Yes, it's called RedHat 3.0.
Seriously, even a 700MHz C3 is a pretty fast machine. And it will actually run today's software quite well.
There also is a full complement of small, efficient programs as part of the Linux handheld projects.
IEEE is going downhill. I mean, what a fluff piece.
To the degree that "augmentation" is going to happen, it's going to happen for medical purposes here on earth: drug delivery, joint replacement, osteoporosis treatment, etc.
"You look at this blob on the ground and say, 'There goes our ancestor,' " he said. -- writes Penelope Debelle.
She ought to read Stephen Jay Gould's work before writing about science. Her statement, in fact, is very much wrong. The common supposition that there is a steady march from simple forms to more complex ones is well debunked by Gould.
She was quoting someone else.
And what do you think is wrong with that statement anyway? We have a complex body plan and we have evolutionary ancestors. It is perfectly alright to consider a 500 million year fossil our ancestor. And it is even perfectly alright to look at some animal today and think of it as being very similar to our ancestor because, for one reason or another, some animals living today are close in many ways to some of our ancestors.
And, no, I don't think every science writer needs to have read Gould. In fact, I think science writers are far to enamored with Gould, attracted more by his writing style than the substance of his writing.
The Livermore "community space suit" station was designed as an alterative to this (more usable volume AND artificial gravity), and it would have launched in ONE shot of a Titan.
Well, that's nice, but it hasn't happened yet. Until people actually demonstrate cheap technologies for lifting people into orbit and keeping them there, we have to go by current prices, and they are hugely expensive. I'm sure that if costs for such projects come down to something that's comparable to unmanned launches and the benefits become obvious, they will be carried out, if not by the US then by Europe, India, Japan, China, or the Russians.
Actually, manned space (planetary) exploration would be a whole lot cheaper if it was all planned as one-way trips. But that's something the American public just can't stomach, although I suspect you probably could get the volunteers.
But when you compare even the most sophisticated surface rovers with the capabilities of a human with a rock hammer, it's obvious that really serious investigations are going to require people on the scene.
No, that's far from obvious to me. Surface rovers don't have to be fully automated--they can be remotely controlled from earth on missions that take many years. It doesn't matter if they can go only a few feet per day or work at a snail's pace.
The OS is obviously Linux: it's free and far more mature and better supported for cluster computing than OS X.
and add on Gigabit ethernet,
The Opteron price includes dual Gigabit Ethernet.
now add in support for setting up the Opteron whereas the G5's required little preparation,
You're kidding, right? The G5's are a pain to install in comparison to the Opteron: they need to be put on shelves (lots and lots of shelves becuse they come in big, bulky boxes), they don't come with cluster software preinstalled, and VT apparently even had to fiddle with the hardware configuration before deploying them.
The Opterons, you slide into the rack, plug into the hub, and turn them on. Or you can order them mounted in a rack. They'll boot over the network and install (and update!) the OS automatically.
Look, the deal is done. You are arguing against fact. Opterons were looked at. For this deal with VT, Apple had the best bid and VT chose that solution. What are you arguing? The point is that the Apple G5's provided the best bang for the buck and the project coordinators decided on that solution.
The fact is that neither you nor anybody else has been able to actually go through the calculations and arguments demonstrating that the G5's are a good choice for compute clusters. In fact, based on everything that's known, they are not.
As far as VT is concerned, that leads us to conclude that either some die-hard Mac zealots at VT pushed through the deal against rational arguments, or that Apple somehow sweetened the deal with conditions we don't know about as a PR stunt. If it's the latter, it's working, since people like you now go around claiming that those machines are a good choice.
In any case, thanks for the discussion: it has shown me that there probably really is nothing to these claims that G5's make good cluster machines: no extra-cheap configuration or deal, no extraordinary performance for particular cluster computing tasks, etc. They are just big, shiny, expensive boxes that some people have an emotional attachment to.
So, when they screw up your records, it's not just your credit rating or phone service that's screwed up, it's your entire life. Forget about trying to get them to correct things--if you complain, you are obviously a trouble-maker and a terrorist. And think of all the wonderful "marketing tie-ins" that will come with this.
Give me a government run national ID card system any day over that. Institutions like the INS or IRS may be a pain to deal with, but they don't sell my data (well, not as much), and sooner or later, they have to put their records in order. Even Ashcroft and Ridge and their agencies are subject to more public controls than any private outfit.
Why do patient records, in particular those sent for transcription, have names attached to everything anyway? The doctor doesn't need to mention the patient by name in his dictation. Without the names, this would all just be a bunch of unimportant medical mumbo jumbo and posting it on the Internet would be no threat.
There were no special showcase deals here. A proposal was made, a number of companies bid on it, and Apple won the proposal based upon total performance/dollar.
You keep saying that, but that just doesn't stand up to scrutiny if VT paid regular educational pricing on the G5's.
Dual rack-mounted Opteron 240's start at around $2600 with 1G of RAM, and they are still faster (according to SPEC) than than the 2GHz dual G5s with only 512M of RAM and educational pricing of $2700.
And compared to the rack-mounted Opterons, the dual G5s make really inconvenient cluster machines with their bulky enclosures and desktop OS.
There is an easy solution to that: don't send humans into space for the time being. Eventually, we can build spacecraft that are large enough to generate "artificial gravity" by rotation. Until then, robots and teleoperators are far cheaper and more effective for space exploration and scientific missions into space.
Slashdot Mirror
That software is software you install by blowing away what the Zaurus ships with. So, no, the Zaurus doesn't "use" that software, but it can be put on there by someone sufficiently dedicated. But, then, you can do the same with a dozen or so PocketPC handhelds.
It is hard to exaggerate the tremendous value this search function has. Amazon has taken a huge leap foward in providing access to information. Really.
What's the point of providing this kind of access if you still need to pay way too much for content that comes in cumbersome form (DRM, books)?
I hope the authors' guild wins this one. The harder they make it for people to access their content, the easier it will be for truly free content to take over the market.
Bittorrent works for Linux because what is being distributed is free. Apple does have the equivalent for OSX, namely pirated CD's. It just happens not to mesh well with their business and licensing model.
The Zaurus runs the Linux kernel, but it uses a proprietary window system (TrollTech's Qtopia, vaguely like Qt). That means most Linux GUI development tools and applications won't work on it and you have to pay big bucks to TrollTech if you want to develop anything commercial.
As for running Palm applications, sure, the Palm emulator for Linux should run just fine on those things (in fact, I think it has been ported).
We have standards for that sort of thing. Presumably, the LG optical drives are standard ATAPI drives, not "Windows drives". If Linux destroys them with standard CD-ROM drivers, then it's a problem with the drives.
In fact, it's hard to see how any CD-ROM driver should be able to destroy any CD-ROM drive unless the drive has some kind of serious design flaw.
I don't know for sure, but if these black boxes operate as aircraft boxes do, I believe they will only store a certain time's worth of information.
The problem is that if they aren't regulated, there is no telling what they record. Car manufacturers may well find it convenient to record your entire driving history (maybe even GPS location) for marketing and product development purposes. They aren't even obligated to tell you. They might sell that data to lots of other companies, together with information about what kind of consumer you are, etc.
The black box simply makes things more acurate.
As I was saying: then make it more accurate for everybody then and make them mandatory. But the current situation, where people can select what kind of blackbox they get, is not acceptable because it encourages dangerous drivers not to have blackboxes, which is exactly what we don't want to happen.
I am not sure about American law but in Canada (where this took place) if you hit a person from behind and because they braked YOU are responsible. You did not keep a safe distance back from the other driver.
I seriously doubt that Canadian law is so rigid. There are many ways in which you can drive perfectly safely and you still end up rear-ending another car. For example, a car could squeeze into the space in front of you, creating an unsafe distance between them and you, and then its driver slams on the brakes; there is no way for you to avoid running into them, and the fault is clearly with them, first by taking away your safety zone and then by slamming on their brakes. In fact, I suspect that kind of accident is quite common.
But the situation is not analogous. First of all, you don't have much of a choice in whether you leave tire marks--it may happen or it may not. But you do have a choice in whether you drive a vehicle with a blackbox. Furthermore, different black-boxes record different kinds of information. Some may collect a lot of your driving history, not just a little information around the accident.
If this information is going to be used, then blackboxes should be mandatory and standardized.
The point is that, right now, only some drivers have them, and the people who don't have them can self-select. That's not a good situation.
What do you do, for example, in an accident where one car has a blackbox recorder and it shows the car was speeding, while the other car doesn't have a recorder? It may have been speeding much more, but that may be much harder to prove.
Furthermore, the information in these recorders can be used for other purposes: a private investigator can potentially find out how far you have been driving to your extramarital affair and what roads you have probably taken, and so can Mr. Ashcroft and his goons.
If these things are such a good idea, then politicians should make them mandatory. Of course, that proposal is probably going to be an uphill battle. But to stick only the drivers with this who don't know any better not to have one just isn't fair.
It isn't the police's right to know that you had the gas to the floor when you rear-ended the woman in front of you, killing her and her two kids
Just because you were speeding doesn't mean you are guilty in that case. The woman may have slammed on her brakes in the middle of the highway for no good reason, and the fact that you were 15 mph over the speed limit (like everybody else, her included, although her cheap and unsafe 1979 Honda CVC may not show that) may make little difference.
But because you have a blackbox and she doesn't, everybody is going to jump to conclusions. Like you did.
I'm not saying it should be mandated that these be installed in every single vehicle manufactured,
They should be mandatory if they are going to be used at all. Otherwise, people who know they are going to drive like maniacs will select cars without blackboxes while mostly law-abiding drivers will get screwed if they cross the line some time (emergency, etc.) and happen to have an accident.
Either make these things mandatory for everybody or don't permit them to be used as evidence at all. But using them as evidence in only some cases is really unfair.
That has led me to start looking around for a new pair of SLRs, one film, one digital. Ideally, I'd like to be able to have the same set of lenses that could be mounted on both a film & a digital camera body,
That means you really need to get a digital SLR with a full-frame sensor. But those are hugely expensive.
I wouldn't bother with the digital SLRs yet--I think they are overpriced and underperforming. Just get a high-end digital viewfinder and use the LCD when you need precise framing.
"Professional" reviewers of photographic equipment are almost always far too positive. Among other things, they usually depend on getting free loaners to review, and if they trashed a camera in a review, they might not get more free loaners in the future. Some of the digital cameras I have had have been real duds, yet they all received reasonably good reviews.
And then you have the analog traditionalist nuts, the photographic equivalent of the people who claim that vinyl and tubes are higher quality than CDs. You can have a 48Mpixel camera and they'll still claim that some random 35mm film beats it.
And what does it matter anyway? Digital is just different from analog. If you have the money, give it a try and see whether you like it. If you don't have the money, don't even get started.
Europe won't do it. Europe's creative drive is spent, gone, kaput. Japan can't even find a launch site that lets them get birds into orbit on schedule, and is experiencing an inverted population pyramid with all the lack of dynamism that implies. So's Europe, for that matter.
Well, at least China, India, Japan, and Europe haven't abandoned education, science, or engineering. The US educational system is in shambles and if the US didn't import a large fraction of its scientists and engineers, it wouldn't have much of a technology industry. In fact, without the influx of WWII refugees, the US would probably never have become a scientific or technological powerhouse.
I'd put my money on those other nations, rather than the US, to drive science and technology forward in the future.
It doesn't worry me that they'll go. It worries me that they'll use it as a way to expand the prestige of their dictatorship and extend their illegitimate powers. That would be good for nobody except the dictators.
What, pray tell, is "illegitimate" about the Chinese government? You may not like them and I may not like them, but that doesn't make them illegitimate. And while they aren't a democracy, they certainly aren't a dictatorship either.
In terms of quality, these digital SLRs are comparable to, or even better than, film. And given the cost of film, they are cost effective, too.
The problem is: they are not 35mm cameras, the sensor is smaller than 35mm film. That means that you get a focal length multiplier and are more limited on the wide angle side.
If you want a drop-in replacement for a 35mm camera, you need a full-frame digital SLR. Or, alternatively, get a non-SLR digital camera.
Unless we do something to upset the apple cart, we will be stuck with a couple people in orbit doing nothing and going nowhere. I do not find this an acceptable state of affairs. The people now running the show are some of the biggest obstacles to progress, and they have to be shoved out of the way.
Don't worry--when the costs have come down enough, China, India, Europe, Japan, and other nations will do that, no matter how cushy the relationship between the US government and large US aerospace contractors may be. In fact, China already has begun. Isn't international competition great?
Start a project that aims to develop extremely efficient programs designed
to run very well with slow procs like these. Hell if you can web browse on a C64, this can be done.
Yes, it's called RedHat 3.0.
Seriously, even a 700MHz C3 is a pretty fast machine. And it will actually run today's software quite well.
There also is a full complement of small, efficient programs as part of the Linux handheld projects.
IEEE is going downhill. I mean, what a fluff piece.
To the degree that "augmentation" is going to happen, it's going to happen for medical purposes here on earth: drug delivery, joint replacement, osteoporosis treatment, etc.
it should include a substantial collection of pre-ripped CDs. You know, the "all classical" MP3 collection or the "all Jazz" MP3 collection.
"You look at this blob on the ground and say, 'There goes our ancestor,' " he said. -- writes Penelope Debelle.
She ought to read Stephen Jay Gould's work before writing about science. Her statement, in fact, is very much wrong. The common supposition that there is a steady march from simple forms to more complex ones is well debunked by Gould.
She was quoting someone else.
And what do you think is wrong with that statement anyway? We have a complex body plan and we have evolutionary ancestors. It is perfectly alright to consider a 500 million year fossil our ancestor. And it is even perfectly alright to look at some animal today and think of it as being very similar to our ancestor because, for one reason or another, some animals living today are close in many ways to some of our ancestors.
And, no, I don't think every science writer needs to have read Gould. In fact, I think science writers are far to enamored with Gould, attracted more by his writing style than the substance of his writing.
The Livermore "community space suit" station was designed as an alterative to this (more usable volume AND artificial gravity), and it would have launched in ONE shot of a Titan.
Well, that's nice, but it hasn't happened yet. Until people actually demonstrate cheap technologies for lifting people into orbit and keeping them there, we have to go by current prices, and they are hugely expensive. I'm sure that if costs for such projects come down to something that's comparable to unmanned launches and the benefits become obvious, they will be carried out, if not by the US then by Europe, India, Japan, China, or the Russians.
Actually, manned space (planetary) exploration would be a whole lot cheaper if it was all planned as one-way trips. But that's something the American public just can't stomach, although I suspect you probably could get the volunteers.
But when you compare even the most sophisticated surface rovers with the capabilities of a human with a rock hammer, it's obvious that really serious investigations are going to require people on the scene.
No, that's far from obvious to me. Surface rovers don't have to be fully automated--they can be remotely controlled from earth on missions that take many years. It doesn't matter if they can go only a few feet per day or work at a snail's pace.
O.K., now add on an OS to the Opteron,
The OS is obviously Linux: it's free and far more mature and better supported for cluster computing than OS X.
and add on Gigabit ethernet,
The Opteron price includes dual Gigabit Ethernet.
now add in support for setting up the Opteron whereas the G5's required little preparation,
You're kidding, right? The G5's are a pain to install in comparison to the Opteron: they need to be put on shelves (lots and lots of shelves becuse they come in big, bulky boxes), they don't come with cluster software preinstalled, and VT apparently even had to fiddle with the hardware configuration before deploying them.
The Opterons, you slide into the rack, plug into the hub, and turn them on. Or you can order them mounted in a rack. They'll boot over the network and install (and update!) the OS automatically.
Look, the deal is done. You are arguing against fact. Opterons were looked at. For this deal with VT, Apple had the best bid and VT chose that solution. What are you arguing? The point is that the Apple G5's provided the best bang for the buck and the project coordinators decided on that solution.
The fact is that neither you nor anybody else has been able to actually go through the calculations and arguments demonstrating that the G5's are a good choice for compute clusters. In fact, based on everything that's known, they are not.
As far as VT is concerned, that leads us to conclude that either some die-hard Mac zealots at VT pushed through the deal against rational arguments, or that Apple somehow sweetened the deal with conditions we don't know about as a PR stunt. If it's the latter, it's working, since people like you now go around claiming that those machines are a good choice.
In any case, thanks for the discussion: it has shown me that there probably really is nothing to these claims that G5's make good cluster machines: no extra-cheap configuration or deal, no extraordinary performance for particular cluster computing tasks, etc. They are just big, shiny, expensive boxes that some people have an emotional attachment to.
So, when they screw up your records, it's not just your credit rating or phone service that's screwed up, it's your entire life. Forget about trying to get them to correct things--if you complain, you are obviously a trouble-maker and a terrorist. And think of all the wonderful "marketing tie-ins" that will come with this.
Give me a government run national ID card system any day over that. Institutions like the INS or IRS may be a pain to deal with, but they don't sell my data (well, not as much), and sooner or later, they have to put their records in order. Even Ashcroft and Ridge and their agencies are subject to more public controls than any private outfit.
Why do patient records, in particular those sent for transcription, have names attached to everything anyway? The doctor doesn't need to mention the patient by name in his dictation. Without the names, this would all just be a bunch of unimportant medical mumbo jumbo and posting it on the Internet would be no threat.
There were no special showcase deals here. A proposal was made, a number of companies bid on it, and Apple won the proposal based upon total performance/dollar.
You keep saying that, but that just doesn't stand up to scrutiny if VT paid regular educational pricing on the G5's.
Dual rack-mounted Opteron 240's start at around $2600 with 1G of RAM, and they are still faster (according to SPEC) than than the 2GHz dual G5s with only 512M of RAM and educational pricing of $2700.
And compared to the rack-mounted Opterons, the dual G5s make really inconvenient cluster machines with their bulky enclosures and desktop OS.
There is an easy solution to that: don't send humans into space for the time being. Eventually, we can build spacecraft that are large enough to generate "artificial gravity" by rotation. Until then, robots and teleoperators are far cheaper and more effective for space exploration and scientific missions into space.