NASA is not a scientific organisation. It is an administration and says so in the title: National Aeronautics and Space Administration). NASA is filled with administrators who measure themselves and compare themselves according to how big and glossy their projects are. "I manage a $2 billion, 200-person project" beats "I manage a $5 million, 7 person project". You get up the ladder by leading bigger, shinier projects.
So which probe is going to win its managers the most karma: the $500M project staffed by PhDs and covered in gold foil or the $250k probe staffed by highschool kids and covered in tin foil?
The/. summary is an excellent example of such unscientific hype. Linking last winter to global warming is pure speculation that does nothing to promote rational discussion about global warming. A mild wointer might indeed be a result of global warming or it could be just a peak in some other climatic cycle that we don't fully understand.
Here in New Zealand, we have just had a very cool summer, following on from a very cool winter. Where's some of that global warming stuff? Could have used it at the beach!
To think that we (as a human race) have a very good understanding of long-term climatic processes is just arrogance. We have models which we are always refining, but they will always just be speculation. We look back mockingly at how ignorant some scientists were 40 years ago (eg. during the 1960s many/most geologists did not accept tectonic plate theory). It is silly to think that people forty years from now won't be doing the same about us. That should be particularly true of climatic modelling. There is no robust equation for climate. People essentially just sit down and tweak the models until they get the results they expect, then use them to generate best case and worst case analysis. That folks, is hardly science.
The shuttle is a 20-something year old POS. A car manufacturer would have recalled and cancelled anything this bad long ago. Screwing up this badly requires government input. Twenty years back (yes I remember the first shuttle launch), there was a promise of a brave new age with space trips being as simple as regular airline flights (ence the name Shuttle - something like a shuttle bus which just takes you for an easy casual ride from one place to another). Roll forward 20+ years and we're just stuck in a 1980s time warp. If airlines were as unsafe as the shuttle, every day there'd be 4 plane crashes at LAX before breakfast.
Did you go to school? Did you remember there being a bully called Joe or whatever? Joe gets pulled in for bullying. Perhaps he got publicly humilated: "Class, Joe has been very bad. He has something he wants to say to John". Mumble, mumble.... "Soooory John". Ten minutes later when the teacher isn't looking, Joe smacks John.
Remember folks... A volcano is an investment that lasts for generations. That's why you need easy to use software for your volcano so that your kids and unborn grandkids can use it. Microsoft Windows for Volcanos is so simple, sperm can use it.
You need to get your Volcano software from a company that will be there to send you patches and keep your volcano humming along nicely. You can't trust those pesky Linux starups that last only until all the share options are cashed up.
You also need to consider Total Cost of Ownership. That is, beyond the ticket price you need to consider the costs associated with keeping your volcano up to date over the centuries. Independent analysis shows that Microsoft Windows for Volcanos is the best software choice for your volcano.
Maybe they're doing some trials, using a bunch of different distros, like any sensible org would do if contemplating a switch. A bank with millions of global customers, as HSBC is, can't switch systems overnight and risk mayhem.
They should however have factored the multi-distro part out ot their TCO equation. Perhaps they did and the Microvell evil twins stirred it back in.
Of course I have not RTFA, that's cheating, but a global corp has to also consider seamless service of its remote offices too. You can buy a Windows box in any city in the world and expect it to perform to a certain level. Setting up a Linux box is not always that easy.
"The Bible contains all sorts of statements that we now know to be false."
I am an agnostic and by no means a religous person. However I believe that statement to be wrong.
The Bible contains all sorts of statements that we currently believe to be false.
There is a huge difference. Religion, modern science, whatever are all believe systems. They have axioms which we must believe, then build up structures on those. Every now and thenscience encounters something that does not fit (eg relativity) and we need to adjust our axioms and belief systems and update science.
There are also scientific heretics. They don't, typically, get physically burnt any more. They just get ridiculed by others, don't get published or get fired from their academic positions.
Why is it that westerners think that the developing world has to start way back in the 1960s and slowly work forwards towards the current time? That thinking is plain ignorance and does nobody any service. It is highly disrespectful of the industries that are blooming in developing countries. It also give the westerners a false sense of superiority/security: "Bah! Don't worry about them. Their technology is still stuck in 1963." One day you'll wake up and start whining because your job went to India/China/Rwanda.
Cross species generalisations are the basis of most of this kind of research, mainly because lab rats are cheap and sdon't complain when you kill them and copen them up to look for changes. But how much of it is really valid?
Here's an interesting anecdote: Scientists were running rats through a maze with a reward system to measure how fast they learn. Rats soon learn the optimum path and get rewarded. COnclusion: rats are smart and learn. Now substitute in a ferret. Ferret searches maze and finds the food (hunk of rabbit). Next time the ferret does not go down that path of the maze. Conclusion: ferrets are stupid and don't learn. But if you know a bit more about ferrets, then you'd know that they know it is a waste of time to go down the same hole twice. They've either eaten or scared the occupants, so don't waste time.
It is the revenue stream that controls TV. That revenue stream comes from either subscribers or advertisers. Both depend on viewership/marketshare, so you control the revenue stream.
Is MCDonalds made a TurdBurger and nobody bought it, they'd soon cut it from the menu. If TV viewership dropped by 50% the TV industry would soon change their tune.
Most people, however, just won't care and will take whatever is thrown at them.
You don't need TV. I have not had TV for approx 10 years (and yes, I do have a wife + kids).
If the USA has the capability (which I'm sure they do) and the funding (debatable) then they have this stuff sitting up there. The USA has never believed in "playing fair" and will grab and hold military dominance in all spheres that it can - with or without treaties.
There is no real problem with microkernels in terms of being able to build a system. In many ways they are easier to develop than monolythics because an OS crash does not take down the whole system like, say, a kernel oops does.
Linus muddied things when he made his microkernel/masturbation simily. Just because he said things in a way geeks can relate to does not make him right on this point.
There are many highly reliable and highly efficient microkernels out there. For example, QNX (which runs just fine on a 286), is a microkernel, as are Minix, OS9 and many other RTOSs. You don't get to be an RTOS if you're just bloat. I saw some pretty cool systems using QNX clusters running on 286s back in 1989 or so.
Hurd just goes no where because there is no motivation and drive. Hurd lacks drivers. Hurd lacks drive. Hurd lacks.
Intel is a weird company when it comes to the way they do business and I am suprised they are stepping into NAND flash space. The writing was on the wall since they are members of ONFI http://www.onfi.org/
Intel bough the StrongARM off Digital, then sold it, presumably to focus on "core business" of x86 etc. They've done similar moves with their 8051 and USB parts. It is hard to see what would attract them to NAND flash which has very low margins. NAND flash now costs less than 1 cent per MByte, about a fifth or so of what it cost a year back, and there seems to be no slowing.
Intel seems to work well with high margin devices (Pentium etc) and not so well with low margin parts (USB chipsets, PXAxxx etc).It is hard to see Intel keeping in the NAND business for very long.
The cards with internal controllers do something like you say and you can thead the SD or SmartMedia specs for details. They manage a "free pool" primarily as a way to address bad blocks, but this also provides a degree of wear levelling.
Putting a FAT partition onto such a device, or into a file via loop mounting, only gives you wear levelling. It does not buy you integrity. If you eject a FAT file system before mounting it then you are likely to damage the file system (potentially killing all the files in the partition). This might be correctable via a fschk.
Proper flash file systems are designed to be safe from bad unmounts. THese tend to be log structured (eg. YAFFS and JFFS2). Sure, you might lose the data that was in flight, but you should not lose other files. That's why most embedded systems don't use FAT for critical files and only use it where FAT-ness is important (eg. data transfer to a PC).
A GPS-like spread spectrum signal just looks a lot like noise unless you are able to "de-spread" it, for which you need to know the pseudo-random numbers employed. The Red October signal was easy to detect because it was repetitive and thus very easy to "decode" and the bursts were relatively high energy.
For long psuedorandom number sequences you could potentially reduce power by a factor of 1000 or so. This stuff could look like white noise, so if the signal levels can be held low enough then it will be very difficult for any observer to be able to detect that the signal is there. Whether the level is low enough for military is an exercise for the guys in white lab coats.
These claims will be made at the flash level (ie. ignoring what the block managers and file systems do).
Different file systems and block managers do different things to code with wear levelling etc. For some file systems (eg. FAT) wear levelling is very important. For some other file systems - particularly those designed to work with NAND flash - wear levelling is not important.
The MTBF only applies to failures at ther NAND level, not the software level.
In most cases the part that fails is the software, not the hardware. For example, FAT is a terrible way to store data you love. To get reliability you need to use a flash file system that is designed to cope with NAND.
Slashdot Mirror
So which probe is going to win its managers the most karma: the $500M project staffed by PhDs and covered in gold foil or the $250k probe staffed by highschool kids and covered in tin foil?
Last time I checked, /. was not a dating agency.
Here in New Zealand, we have just had a very cool summer, following on from a very cool winter. Where's some of that global warming stuff? Could have used it at the beach!
To think that we (as a human race) have a very good understanding of long-term climatic processes is just arrogance. We have models which we are always refining, but they will always just be speculation. We look back mockingly at how ignorant some scientists were 40 years ago (eg. during the 1960s many/most geologists did not accept tectonic plate theory). It is silly to think that people forty years from now won't be doing the same about us. That should be particularly true of climatic modelling. There is no robust equation for climate. People essentially just sit down and tweak the models until they get the results they expect, then use them to generate best case and worst case analysis. That folks, is hardly science.
The shuttle is a 20-something year old POS. A car manufacturer would have recalled and cancelled anything this bad long ago. Screwing up this badly requires government input. Twenty years back (yes I remember the first shuttle launch), there was a promise of a brave new age with space trips being as simple as regular airline flights (ence the name Shuttle - something like a shuttle bus which just takes you for an easy casual ride from one place to another). Roll forward 20+ years and we're just stuck in a 1980s time warp. If airlines were as unsafe as the shuttle, every day there'd be 4 plane crashes at LAX before breakfast.
Did you go to school? Did you remember there being a bully called Joe or whatever? Joe gets pulled in for bullying. Perhaps he got publicly humilated: "Class, Joe has been very bad. He has something he wants to say to John". Mumble, mumble.... "Soooory John". Ten minutes later when the teacher isn't looking, Joe smacks John.
You need to get your Volcano software from a company that will be there to send you patches and keep your volcano humming along nicely. You can't trust those pesky Linux starups that last only until all the share options are cashed up.
You also need to consider Total Cost of Ownership. That is, beyond the ticket price you need to consider the costs associated with keeping your volcano up to date over the centuries. Independent analysis shows that Microsoft Windows for Volcanos is the best software choice for your volcano.
They should however have factored the multi-distro part out ot their TCO equation. Perhaps they did and the Microvell evil twins stirred it back in.
Of course I have not RTFA, that's cheating, but a global corp has to also consider seamless service of its remote offices too. You can buy a Windows box in any city in the world and expect it to perform to a certain level. Setting up a Linux box is not always that easy.
More seriously, these apologies do nothing useful. They just build resentment and I bet he's thinking: "I'll get you fuquers! Payback shall be mine!"
I am an agnostic and by no means a religous person. However I believe that statement to be wrong.
The Bible contains all sorts of statements that we currently believe to be false.
There is a huge difference. Religion, modern science, whatever are all believe systems. They have axioms which we must believe, then build up structures on those. Every now and thenscience encounters something that does not fit (eg relativity) and we need to adjust our axioms and belief systems and update science.
There are also scientific heretics. They don't, typically, get physically burnt any more. They just get ridiculed by others, don't get published or get fired from their academic positions.
Why is it that westerners think that the developing world has to start way back in the 1960s and slowly work forwards towards the current time? That thinking is plain ignorance and does nobody any service. It is highly disrespectful of the industries that are blooming in developing countries. It also give the westerners a false sense of superiority/security: "Bah! Don't worry about them. Their technology is still stuck in 1963." One day you'll wake up and start whining because your job went to India/China/Rwanda.
you buy one first!
Given the countelss times MS and dressed up evil in cool clothing, why should we believe them this time?
Here's an interesting anecdote: Scientists were running rats through a maze with a reward system to measure how fast they learn. Rats soon learn the optimum path and get rewarded. COnclusion: rats are smart and learn. Now substitute in a ferret. Ferret searches maze and finds the food (hunk of rabbit). Next time the ferret does not go down that path of the maze. Conclusion: ferrets are stupid and don't learn. But if you know a bit more about ferrets, then you'd know that they know it is a waste of time to go down the same hole twice. They've either eaten or scared the occupants, so don't waste time.
It is the revenue stream that controls TV. That revenue stream comes from either subscribers or advertisers. Both depend on viewership/marketshare, so you control the revenue stream.
Is MCDonalds made a TurdBurger and nobody bought it, they'd soon cut it from the menu. If TV viewership dropped by 50% the TV industry would soon change their tune.
Most people, however, just won't care and will take whatever is thrown at them.
You don't need TV. I have not had TV for approx 10 years (and yes, I do have a wife + kids).
Trust me, you'll thank the government when you can still drive your SUVs after oil has run out!
Still WTF is a Zimbabwean poet doing coining Geeky Computer terms? Fuck off buster! I don't try making clever terminology about poetry.
If the USA has the capability (which I'm sure they do) and the funding (debatable) then they have this stuff sitting up there. The USA has never believed in "playing fair" and will grab and hold military dominance in all spheres that it can - with or without treaties.
No it does not run Linux.
They will likely use the threat of moving off-shore as a way to get some breaks from Texas.
Linus muddied things when he made his microkernel/masturbation simily. Just because he said things in a way geeks can relate to does not make him right on this point.
There are many highly reliable and highly efficient microkernels out there. For example, QNX (which runs just fine on a 286), is a microkernel, as are Minix, OS9 and many other RTOSs. You don't get to be an RTOS if you're just bloat. I saw some pretty cool systems using QNX clusters running on 286s back in 1989 or so.
Hurd just goes no where because there is no motivation and drive. Hurd lacks drivers. Hurd lacks drive. Hurd lacks.
Intel bough the StrongARM off Digital, then sold it, presumably to focus on "core business" of x86 etc. They've done similar moves with their 8051 and USB parts. It is hard to see what would attract them to NAND flash which has very low margins. NAND flash now costs less than 1 cent per MByte, about a fifth or so of what it cost a year back, and there seems to be no slowing.
Intel seems to work well with high margin devices (Pentium etc) and not so well with low margin parts (USB chipsets, PXAxxx etc).It is hard to see Intel keeping in the NAND business for very long.
Putting a FAT partition onto such a device, or into a file via loop mounting, only gives you wear levelling. It does not buy you integrity. If you eject a FAT file system before mounting it then you are likely to damage the file system (potentially killing all the files in the partition). This might be correctable via a fschk.
Proper flash file systems are designed to be safe from bad unmounts. THese tend to be log structured (eg. YAFFS and JFFS2). Sure, you might lose the data that was in flight, but you should not lose other files. That's why most embedded systems don't use FAT for critical files and only use it where FAT-ness is important (eg. data transfer to a PC).
For long psuedorandom number sequences you could potentially reduce power by a factor of 1000 or so. This stuff could look like white noise, so if the signal levels can be held low enough then it will be very difficult for any observer to be able to detect that the signal is there. Whether the level is low enough for military is an exercise for the guys in white lab coats.
Different file systems and block managers do different things to code with wear levelling etc. For some file systems (eg. FAT) wear levelling is very important. For some other file systems - particularly those designed to work with NAND flash - wear levelling is not important.
In most cases the part that fails is the software, not the hardware. For example, FAT is a terrible way to store data you love. To get reliability you need to use a flash file system that is designed to cope with NAND.