Comparing the complexity of the space shuttle to a space elevator is seriously disingenuous. The shuttle is perhaps the single most complex device created by mankind, and entails enormous maintenance efforts, as would your proposed scheme.
The elevator, well, is a cable, with a relatively modest support infrastructure. It is extremely simple, and economies of scale provide immense advantages. Manufacturing tolerances are just another specification, and it isn't like the cable needs to be defect free. (It is made from numerous strands, each from numerous nanotubes, neither of which need to be particularly long.)
Even if laser propulsion is super efficient, it is still basically a rocket, which must be powered from the ground. An enormous infrastructure is required to provide that power. The numbers I have seen for the Isp are on the order of 1000s, so it isn't as if it is a free ride.
The key point, is that a rocket still has to supply the entire delta-V to reach orbit and beyond. That kinetic energy, with the v^2 term, is huge just to get to orbit--to say nothing of other destinations. For an elevator, you only need to supply the change in potential energy--the momentum comes from the earth itself. (Plus a small correction for orbit insertion, but that is insignificant.) For a trip to Mars or Jupiter, you just get off the ribbon at a greater altitude. It's nearly free.
You are dismissing the space elevator out of hand, without even reading about it, and proposing another scheme. At the very least, I think you should provide more information. 200 tons every 10 minutes sounds great; at what cost? For reference, the SSMEs output about 28GW. Even assuming 100% efficiency, you are talking about utilizing the full capacity of dozens of nuclear power plants.
I have a degree in physics, maybe you should go check your book. The force of gravity varies as 1/r^2, so the vast majority of the ribbon is not at all deep in the potential well.
If that is difficult to understand, here is an example, straight from the book on page 167: you can have 24 50 ton climbers spaced along a 200 ton capacity ribbon. They even developed a formula for it: "Within an upper limit of 1/3 of the ribbons capacity, the maximum number of climbers is 6C/M, where C is the ribbon capacity, and M is the mass of the proposed climber."
As for your three points:
1. The ribbons are not expected to fail often. It will happen eventually, but it just won't be a big deal. 2. That is not an insurmountable engineering problem, and depending on where they break, you may not even lose the whole thing. 3. It is not enormous, nor expensive; the first 200 ton ribbon discussed masses 8,900 tons, and is expected to cost $5B. (For perspective, the Sears Tower weighs in at 222,500 tons, and the Golden Gate Bridge at 419,800 tons.)
It is "friggin obvious" that it will happen at some point, it is only a matter of when. The necessary materials may be difficult or impossible to produce with current methods, but once molecular nanotechnology arrives, they will be both manufacturable, and extremely cheap, without question.
For comparison, the first ribbon (20 ton capacity) is estimated at ~$6B. While I won't discourage you from advocating laser propulsion in the near term, the capacity and economics will never allow for large scale manned space operations. It is not and will never be a replacement for the space elevator,
You seem to be rather ignorant of the idea, and all of your concerns are addressed directly in Edwards' book, The Space Elevator.
A single climber on a single cable is the first step, as it is most cost effective to launch. Once it is up, the first priority is widening the ribbon, and producing more ribbons. Once they are in place, loss of any single ribbon would be not be very significant, as the ribbon itself is cheap, and deploying it is now cheap.
Next, the goal was always to run multiple climbers up the ribbon, in a single direction, as that makes best use of the ribbons capacity. For climbers spaced along the ribbon, the force of gravity is greatly reduced the further up you go, so you can fit considerably more on the ribbon as compared to a single heavy climber. Still, you have the option of clearing the ribbon and sending up very heavy items, it would just cost more.
Anything else you can think of has also likely been addressed in that book, from technology to economics.
This looks brilliant, but I am curious if the plug can handle much current, given the tiny rotating sleeve type connections. Also, how durable are those connections?
Raising awareness is fine, but the parent clearly has an axe to grind, and is also presenting information which is far out of date. Incidentally, the Limitations section of the wiki on ZFS has been shrinking considerably, and resizing is the only notable one left.
Vdev removal has been a long requested feature, but I don't recall seeing any promises, or evidence of anyone seriously working on it until relatively recently. It may have slipped a year or so, but the developers have been rather open about their continued efforts, and the difficulty of the problem. It is clearly a priority at Sun now.
I admit that it has been a difficult wait, but I have been more worried about Oracle yanking the plug on recent activities, than the ZFS team not delivering as promised. Seeing the dedup work integrated does put my mind at ease regarding that.
How alarmist and uninformed; borderline FUD. The reality is as follows...
First, you can't remove a vdev yet, but development is in progress, and support is expected very soon now. Same with crypto.
Second, mistakenly typing add instead of attach will result in a warning that the specified redundancy is different, and refuse to add it.
Third, yes, you can't expand the width of a RAID-Z. You can still grow it though, by replacing it with larger drives. Once the block pointer rewrite work is merged, removal will be possible, and expansion won't be far off either.
Forth, vdevs no longer autoexpand by default. If you want that behavior, you can to set the autoexpand property to yes.
Last, there was no such assumption, it is simply a matter of priorities. If it were an easier problem, it would have been done long ago, but I'm happy to be patient, knowing that it will be done right. Most everyone who has seriously used ZFS will understand that the advantages will hugely outweigh these minor nits, which are easily worked around.
Huh? I think it fits quite well, though it does call Rider's integrity into question, which may have been a bit much. Supposedly, Dr. Bussard had tried to point out his errors in the past, but he refused to admit to them. Being clearly on the tokamak side of the line, Rider's motivations in discrediting alternative forms of fusion are certainly called into question.
Rider painted a convenient picture of fusion, and made a naive and overly general argument, which has been used as a tool to discredit alternatives ever since.
This again. Todd Rider's paper is essentially a straw man; his criticisms to do not apply to the Polywell, and they may not apply to the DPF either. His assumptions are flawed, and the resulting claims are too general.
I can't find the specific post I was looking for, but here is a comment from Dr. Nebel, the lead researcher of the Polywell. Dr. Nebel has also co-authored research on the periodically oscillating plasma sphere (POPS), which provides direct experimental evidence for something which should be impossible given Rider's claims.
I'm surprised that the DPF work has elicited nothing more than fusion humor from slashdot. While it is facing some significant engineering challenges, it is one of the approaches which should be taken seriously. (Along with the Polywell, General Fusion's approach, and the FRC based approaches from Helion and Tri-Alpha.)
While tokamak based fusion may still be 50 years away, will never burn p+B11, and may never be economically viable, there are other promising alternatives. We may very well have working fusion reactors in a few years, some even based on the elusive p+B11 reaction.
Perhaps the single most important advantage of PRAM has not even been mentioned yet. PRAM does not require the stupid block erase semantics of Flash--you can read or write as much or as little as you want, at whatever alignment, with no impact on performance. This also means that an SSD will be very simple, require no caches at all, and still have blazing fast write performance, even for synchronous writes.
PRAM will still require ECC algorithms, wear leveling, and bad block remapping, but on the spectrum of controller complexity, it is a lot closer to DRAM than Flash. (Incidentally, the same can be said of performance.) Reads and writes would still be buffered for queuing purposes, but this is very different from a cache; it is simply to allow requests to be pipelined from the storage controller.
Compared with the very simple constant time operations with PRAM, Flash is a dog. The controller must cache writes while it reads, erases, and otherwise shuffles blocks around. Moreover, as the controller operates with volatile memory, it must do this very slowly and carefully, or a power failure could severely corrupt the disk. (There are Flash SSDs with an onboard super capacitor to work around this, but they are obscenely expensive.)
Due to their inherent nature, even the best Flash SSDs have severely asymmetric read/write performance. The fact that only one company (Intel) has managed to produce a decent controller also betrays the immense complexity required to eek out even moderately acceptable random write performance. In my opinion, so called "SSDs" made with Flash don't even deserve that moniker, as they are more like a fast hard disk. (They still have a sort of geometry which constrains performance, and aren't anywhere near as fast as DRAM.)
PRAM will fix that, offering performance similar to a DRAM SSD. There are many companies banking on Phase-change RAM to displace Flash memory, Intel included. The wikipedia page has a lot more info, but basically, PRAM is superior to Flash in every way, except that the data on a prewritten chip won't survive a trip through the wave soldering machine.
My earplugs are rated for 33dB of protection; bringing the sound level down from 151dB to 118dB will probably not make you much happier. I can't comment on the relative pain or ear bleeding, but both will cause hearing damage in short order.
To be fair, the absorption depends on frequency, and may be somewhat better at whatever frequencies they are using. Still, there is no way you will manage an ideal seal before you are on the ground with an aneurism. It takes me about 30 seconds per ear to get it right, with most of that time waiting for the foam to expand. That is way too slow.
From your comment, it is clear that you are ignorant, and didn't bother to click those links--they are nothing like Palm's input. While it does take some time to learn them, comparing the difficulty to that of kanji is seriously disingenuous. The miniature qwerty keyboard is not the pinnacle of text-based input!
It has nothing to do with "change just for the sake of change." It has to do with a willingness to learn something new, for the sake of achieving much faster text input. If you can type 45-50WPM on a miniature qwerty keyboard, I am impressed.
It is still wasted space and adds considerable thickness to the phone, not to mention making it more fragile, both to impact and liquids. Above all though, there are much betterinputmethods for a handheld device than a shrunken conventional keyboard!
Most of the people who are attached to physical keyboards are simply creatures of habit. Unfortunately, existing physical and virtual keyboards (as on the iPhone) are targeted at people who are averse to change, even if they are both far from optimal.
Can you get the N900 without a lousy physical qwerty keyboard? While the HTC Magic looks decent, even it wastes space on physical buttons.
What I really want is an iPhone with a less restrictive software environment, using an efficient virtual keyboard like ShapeWriter. A minimal slab of computing hardware which is as densely packed with battery and display area as is physically possible.
The size of a byte is arbitrary (yet binary) to begin with: 2^3 bits. SI prefixes are generally applied to fundamental units, and the byte is not the fundamental unit of storage. It makes no sense to use SI prefixes on units which are fundamentally of a different base, like minutes or dozens.
Storage is addressed in binary, wether memory or disk, tape or optical. The conventional binary units are simply more appropriate. Networks and busses are measured by frequency, so it is only natural that they use the SI units on bits.
Using the iB prefixes would be fine. However, the binary prefixes should be used for storage--especially since every other operating system already does so. This only makes the Mac have non-portable sizes; people transferring data from other operating systems to their Mac will now discover that the size is different. Sometimes it may not even fit, when it looks like it should.
Assuming that the compiler is working properly, probably so. However, it wouldn't be the first time that I have seen code which when removed breaks things horribly, even though it shouldn't. (It is also not rare to see code break at higher optimization levels...)
Agreed about the assertions. Does cobol even have those though?
Why not just trust the ToS bits in the IP headers, and place reasonable limits on the types? In any case, there should to be some way for an application to specify the type of service it requires. By only giving precedence to established protocols, it limits the opportunity for innovative and competing protocols on the Internet. QoS should be protocol agnostic for this reason alone.
Maybe you should do some research as well. TCP is self clocking, and the rate of acknowledgements influences the rate of the data. TCP was not designed with severely asymmetric connections in mind, and if the upstream is choked, your performance will be terrible. Even under good circumstances, you won't be able to make full use of your massive downstream channel without some tuning.
That aside, bittorrent is a lot more than a method to distribute illegal TV-rips: it also allows ordinary people to distribute perfectly legal content, which would otherwise require extremely deep pockets. The benefits of enabling individuals to do so far outweigh any possible lost revenue that the content industry might sustain. (Not that there is even a shred of evidence for such loss though. The actual fear is competition from user generated content; the content industry want to remain the gatekeepers of entertainment.)
The Internet is not a TV. A bidirectional connection is necessary to make the most of it.
Protocol discrimination is an equally important issue for network neutrality, as it has the same result. While prioritizing traffic by protocol in the name of QOS may appear to be fair on the surface, it is anything but, and will stunt the growth of innovative and competing services on the Internet.
Think about it; how will a competing protocol, or any other innovative new protocol emerge when it is so disadvantaged? The most popular existing protocols end up with a natural monopolies on the Internet.
No one should be discriminating based on protocol, and certainly not modifying the payload or disrupting connections. Packets should be flagged with the appropriate ToS bits, and traffic management should be done on that basis instead. IPv6 also provides additional fields in the header for these purposes.
You should not compromise on your definition of Net Neutrality. Ideally, all traffic would be encrypted and authenticated, as anything less is just inviting abuse. Rather than "traffic management," the Internet infrastructure should be kept modern so that it can handle the increasing load.
No, not at all. There is a vast difference in energy required to lift a payload up an elevator versus launching it by rocket. A space elevator is immensely cheaper, even with power beaming.
The elevator has two overwhelming advantages. One, you don't need to bring the fuel with you, and two, the earth supplies the necessary angular momentum. The entire delta-v comes for free; you only pay for potential energy, and do so on a very small fraction of the mass of a rocket.
Not really, since burying the radioactive "waste" is a huge waste; more than 99% of the energy has yet to be extracted from it. (Which is also why it is so dangerous and long lived.) This "waste" can be burned in fast reactors though, and there is enough to supply them for hundreds of years before any further mining is necessary.
All that needs to be done is build the reactors. General Electric even has a design ready for a commercial reactor, called the S-PRISM. This is modeled after the Integral Fast Reactor, a modern design which addresses all of the concerns about nuclear power.
Re:The best ESATA isn't really ESATA at all.
on
Best eSATA JBOD?
·
· Score: 2, Informative
Make sure to find a port multiplier with FBSS (FIS-based switching) support. Also make sure that your SATA controller supports this feature. Otherwise, there can only be one outstanding command for all attached disks, and performance will be abysmal.
This reminds me of something Dr. Bussard said during his google talk:
"countless billions of stars in the universe all doing nuclear fusion...and not a single one of them is shaped like a donut!â
There are other promising possibilities for fusion; maybe we should be funding those, instead of the Tokamaks which cost billions upon billions, and are now 100 years away. Furthermore, even if they do work, they will never be economically viable.
Dr. Bussard's Polywell is one such approach, which thankfully, continues to be funded by the navy. If funding weren't so minimal, perhaps he would have lived long enough to see commercial fusion reactors using this concept. Even so, it looks like we should finally know whether it works within the next 1.5-2 years. Commercial reactors would follow shortly thereafter.
Whatever things people may like to call "nanotechnology," there is really only one important distinction. Can we assemble atoms in any desired configuration? That is what is commonly termed molecular nanotechnology, and it is what most people originally meant.
Once this and fusion are out of the way, life will start to get very interesting; the foundation of our economic systems will become irrelevant as scarcity will cease to be a useful concept.
Slashdot Mirror
Comparing the complexity of the space shuttle to a space elevator is seriously disingenuous. The shuttle is perhaps the single most complex device created by mankind, and entails enormous maintenance efforts, as would your proposed scheme.
The elevator, well, is a cable, with a relatively modest support infrastructure. It is extremely simple, and economies of scale provide immense advantages. Manufacturing tolerances are just another specification, and it isn't like the cable needs to be defect free. (It is made from numerous strands, each from numerous nanotubes, neither of which need to be particularly long.)
Even if laser propulsion is super efficient, it is still basically a rocket, which must be powered from the ground. An enormous infrastructure is required to provide that power. The numbers I have seen for the Isp are on the order of 1000s, so it isn't as if it is a free ride.
The key point, is that a rocket still has to supply the entire delta-V to reach orbit and beyond. That kinetic energy, with the v^2 term, is huge just to get to orbit--to say nothing of other destinations. For an elevator, you only need to supply the change in potential energy--the momentum comes from the earth itself. (Plus a small correction for orbit insertion, but that is insignificant.) For a trip to Mars or Jupiter, you just get off the ribbon at a greater altitude. It's nearly free.
You are dismissing the space elevator out of hand, without even reading about it, and proposing another scheme. At the very least, I think you should provide more information. 200 tons every 10 minutes sounds great; at what cost? For reference, the SSMEs output about 28GW. Even assuming 100% efficiency, you are talking about utilizing the full capacity of dozens of nuclear power plants.
I have a degree in physics, maybe you should go check your book. The force of gravity varies as 1/r^2, so the vast majority of the ribbon is not at all deep in the potential well.
If that is difficult to understand, here is an example, straight from the book on page 167: you can have 24 50 ton climbers spaced along a 200 ton capacity ribbon. They even developed a formula for it: "Within an upper limit of 1/3 of the ribbons capacity, the maximum number of climbers is 6C/M, where C is the ribbon capacity, and M is the mass of the proposed climber."
As for your three points:
1. The ribbons are not expected to fail often. It will happen eventually, but it just won't be a big deal.
2. That is not an insurmountable engineering problem, and depending on where they break, you may not even lose the whole thing.
3. It is not enormous, nor expensive; the first 200 ton ribbon discussed masses 8,900 tons, and is expected to cost $5B. (For perspective, the Sears Tower weighs in at 222,500 tons, and the Golden Gate Bridge at 419,800 tons.)
It is "friggin obvious" that it will happen at some point, it is only a matter of when. The necessary materials may be difficult or impossible to produce with current methods, but once molecular nanotechnology arrives, they will be both manufacturable, and extremely cheap, without question.
For comparison, the first ribbon (20 ton capacity) is estimated at ~$6B. While I won't discourage you from advocating laser propulsion in the near term, the capacity and economics will never allow for large scale manned space operations. It is not and will never be a replacement for the space elevator,
You seem to be rather ignorant of the idea, and all of your concerns are addressed directly in Edwards' book, The Space Elevator.
A single climber on a single cable is the first step, as it is most cost effective to launch. Once it is up, the first priority is widening the ribbon, and producing more ribbons. Once they are in place, loss of any single ribbon would be not be very significant, as the ribbon itself is cheap, and deploying it is now cheap.
Next, the goal was always to run multiple climbers up the ribbon, in a single direction, as that makes best use of the ribbons capacity. For climbers spaced along the ribbon, the force of gravity is greatly reduced the further up you go, so you can fit considerably more on the ribbon as compared to a single heavy climber. Still, you have the option of clearing the ribbon and sending up very heavy items, it would just cost more.
Anything else you can think of has also likely been addressed in that book, from technology to economics.
Wouldn't that be fun? Yeah, those are certainly nice.
In addition, lets have another couple of phases to improve DC conversion and simplify motors.
This looks brilliant, but I am curious if the plug can handle much current, given the tiny rotating sleeve type connections. Also, how durable are those connections?
Raising awareness is fine, but the parent clearly has an axe to grind, and is also presenting information which is far out of date. Incidentally, the Limitations section of the wiki on ZFS has been shrinking considerably, and resizing is the only notable one left.
Vdev removal has been a long requested feature, but I don't recall seeing any promises, or evidence of anyone seriously working on it until relatively recently. It may have slipped a year or so, but the developers have been rather open about their continued efforts, and the difficulty of the problem. It is clearly a priority at Sun now.
I admit that it has been a difficult wait, but I have been more worried about Oracle yanking the plug on recent activities, than the ZFS team not delivering as promised. Seeing the dedup work integrated does put my mind at ease regarding that.
How alarmist and uninformed; borderline FUD. The reality is as follows...
First, you can't remove a vdev yet, but development is in progress, and support is expected very soon now. Same with crypto.
Second, mistakenly typing add instead of attach will result in a warning that the specified redundancy is different, and refuse to add it.
Third, yes, you can't expand the width of a RAID-Z. You can still grow it though, by replacing it with larger drives. Once the block pointer rewrite work is merged, removal will be possible, and expansion won't be far off either.
Forth, vdevs no longer autoexpand by default. If you want that behavior, you can to set the autoexpand property to yes.
Last, there was no such assumption, it is simply a matter of priorities. If it were an easier problem, it would have been done long ago, but I'm happy to be patient, knowing that it will be done right. Most everyone who has seriously used ZFS will understand that the advantages will hugely outweigh these minor nits, which are easily worked around.
Huh? I think it fits quite well, though it does call Rider's integrity into question, which may have been a bit much. Supposedly, Dr. Bussard had tried to point out his errors in the past, but he refused to admit to them. Being clearly on the tokamak side of the line, Rider's motivations in discrediting alternative forms of fusion are certainly called into question.
Rider painted a convenient picture of fusion, and made a naive and overly general argument, which has been used as a tool to discredit alternatives ever since.
This again. Todd Rider's paper is essentially a straw man; his criticisms to do not apply to the Polywell, and they may not apply to the DPF either. His assumptions are flawed, and the resulting claims are too general.
I can't find the specific post I was looking for, but here is a comment from Dr. Nebel, the lead researcher of the Polywell. Dr. Nebel has also co-authored research on the periodically oscillating plasma sphere (POPS), which provides direct experimental evidence for something which should be impossible given Rider's claims.
I'm surprised that the DPF work has elicited nothing more than fusion humor from slashdot. While it is facing some significant engineering challenges, it is one of the approaches which should be taken seriously. (Along with the Polywell, General Fusion's approach, and the FRC based approaches from Helion and Tri-Alpha.)
While tokamak based fusion may still be 50 years away, will never burn p+B11, and may never be economically viable, there are other promising alternatives. We may very well have working fusion reactors in a few years, some even based on the elusive p+B11 reaction.
Perhaps the single most important advantage of PRAM has not even been mentioned yet. PRAM does not require the stupid block erase semantics of Flash--you can read or write as much or as little as you want, at whatever alignment, with no impact on performance. This also means that an SSD will be very simple, require no caches at all, and still have blazing fast write performance, even for synchronous writes.
PRAM will still require ECC algorithms, wear leveling, and bad block remapping, but on the spectrum of controller complexity, it is a lot closer to DRAM than Flash. (Incidentally, the same can be said of performance.) Reads and writes would still be buffered for queuing purposes, but this is very different from a cache; it is simply to allow requests to be pipelined from the storage controller.
Compared with the very simple constant time operations with PRAM, Flash is a dog. The controller must cache writes while it reads, erases, and otherwise shuffles blocks around. Moreover, as the controller operates with volatile memory, it must do this very slowly and carefully, or a power failure could severely corrupt the disk. (There are Flash SSDs with an onboard super capacitor to work around this, but they are obscenely expensive.)
Due to their inherent nature, even the best Flash SSDs have severely asymmetric read/write performance. The fact that only one company (Intel) has managed to produce a decent controller also betrays the immense complexity required to eek out even moderately acceptable random write performance. In my opinion, so called "SSDs" made with Flash don't even deserve that moniker, as they are more like a fast hard disk. (They still have a sort of geometry which constrains performance, and aren't anywhere near as fast as DRAM.)
PRAM will fix that, offering performance similar to a DRAM SSD. There are many companies banking on Phase-change RAM to displace Flash memory, Intel included. The wikipedia page has a lot more info, but basically, PRAM is superior to Flash in every way, except that the data on a prewritten chip won't survive a trip through the wave soldering machine.
My earplugs are rated for 33dB of protection; bringing the sound level down from 151dB to 118dB will probably not make you much happier. I can't comment on the relative pain or ear bleeding, but both will cause hearing damage in short order.
To be fair, the absorption depends on frequency, and may be somewhat better at whatever frequencies they are using. Still, there is no way you will manage an ideal seal before you are on the ground with an aneurism. It takes me about 30 seconds per ear to get it right, with most of that time waiting for the foam to expand. That is way too slow.
From your comment, it is clear that you are ignorant, and didn't bother to click those links--they are nothing like Palm's input. While it does take some time to learn them, comparing the difficulty to that of kanji is seriously disingenuous. The miniature qwerty keyboard is not the pinnacle of text-based input!
It has nothing to do with "change just for the sake of change." It has to do with a willingness to learn something new, for the sake of achieving much faster text input. If you can type 45-50WPM on a miniature qwerty keyboard, I am impressed.
It is still wasted space and adds considerable thickness to the phone, not to mention making it more fragile, both to impact and liquids. Above all though, there are much better input methods for a handheld device than a shrunken conventional keyboard!
Most of the people who are attached to physical keyboards are simply creatures of habit. Unfortunately, existing physical and virtual keyboards (as on the iPhone) are targeted at people who are averse to change, even if they are both far from optimal.
Can you get the N900 without a lousy physical qwerty keyboard? While the HTC Magic looks decent, even it wastes space on physical buttons.
What I really want is an iPhone with a less restrictive software environment, using an efficient virtual keyboard like ShapeWriter. A minimal slab of computing hardware which is as densely packed with battery and display area as is physically possible.
The size of a byte is arbitrary (yet binary) to begin with: 2^3 bits. SI prefixes are generally applied to fundamental units, and the byte is not the fundamental unit of storage. It makes no sense to use SI prefixes on units which are fundamentally of a different base, like minutes or dozens.
Storage is addressed in binary, wether memory or disk, tape or optical. The conventional binary units are simply more appropriate. Networks and busses are measured by frequency, so it is only natural that they use the SI units on bits.
Using the iB prefixes would be fine. However, the binary prefixes should be used for storage--especially since every other operating system already does so. This only makes the Mac have non-portable sizes; people transferring data from other operating systems to their Mac will now discover that the size is different. Sometimes it may not even fit, when it looks like it should.
Assuming that the compiler is working properly, probably so. However, it wouldn't be the first time that I have seen code which when removed breaks things horribly, even though it shouldn't. (It is also not rare to see code break at higher optimization levels...)
Agreed about the assertions. Does cobol even have those though?
Why not just trust the ToS bits in the IP headers, and place reasonable limits on the types? In any case, there should to be some way for an application to specify the type of service it requires. By only giving precedence to established protocols, it limits the opportunity for innovative and competing protocols on the Internet. QoS should be protocol agnostic for this reason alone.
Maybe you should do some research as well. TCP is self clocking, and the rate of acknowledgements influences the rate of the data. TCP was not designed with severely asymmetric connections in mind, and if the upstream is choked, your performance will be terrible. Even under good circumstances, you won't be able to make full use of your massive downstream channel without some tuning.
That aside, bittorrent is a lot more than a method to distribute illegal TV-rips: it also allows ordinary people to distribute perfectly legal content, which would otherwise require extremely deep pockets. The benefits of enabling individuals to do so far outweigh any possible lost revenue that the content industry might sustain. (Not that there is even a shred of evidence for such loss though. The actual fear is competition from user generated content; the content industry want to remain the gatekeepers of entertainment.)
The Internet is not a TV. A bidirectional connection is necessary to make the most of it.
Protocol discrimination is an equally important issue for network neutrality, as it has the same result. While prioritizing traffic by protocol in the name of QOS may appear to be fair on the surface, it is anything but, and will stunt the growth of innovative and competing services on the Internet.
Think about it; how will a competing protocol, or any other innovative new protocol emerge when it is so disadvantaged? The most popular existing protocols end up with a natural monopolies on the Internet.
No one should be discriminating based on protocol, and certainly not modifying the payload or disrupting connections. Packets should be flagged with the appropriate ToS bits, and traffic management should be done on that basis instead. IPv6 also provides additional fields in the header for these purposes.
You should not compromise on your definition of Net Neutrality. Ideally, all traffic would be encrypted and authenticated, as anything less is just inviting abuse. Rather than "traffic management," the Internet infrastructure should be kept modern so that it can handle the increasing load.
No, not at all. There is a vast difference in energy required to lift a payload up an elevator versus launching it by rocket. A space elevator is immensely cheaper, even with power beaming.
The elevator has two overwhelming advantages. One, you don't need to bring the fuel with you, and two, the earth supplies the necessary angular momentum. The entire delta-v comes for free; you only pay for potential energy, and do so on a very small fraction of the mass of a rocket.
Not really, since burying the radioactive "waste" is a huge waste; more than 99% of the energy has yet to be extracted from it. (Which is also why it is so dangerous and long lived.) This "waste" can be burned in fast reactors though, and there is enough to supply them for hundreds of years before any further mining is necessary.
All that needs to be done is build the reactors. General Electric even has a design ready for a commercial reactor, called the S-PRISM. This is modeled after the Integral Fast Reactor, a modern design which addresses all of the concerns about nuclear power.
Not with an OPOC diesel: http://www.ecomotors.com/
Make sure to find a port multiplier with FBSS (FIS-based switching) support. Also make sure that your SATA controller supports this feature. Otherwise, there can only be one outstanding command for all attached disks, and performance will be abysmal.
This reminds me of something Dr. Bussard said during his google talk:
"countless billions of stars in the universe all doing nuclear fusion...and not a single one of them is shaped like a donut!â
There are other promising possibilities for fusion; maybe we should be funding those, instead of the Tokamaks which cost billions upon billions, and are now 100 years away. Furthermore, even if they do work, they will never be economically viable.
Dr. Bussard's Polywell is one such approach, which thankfully, continues to be funded by the navy. If funding weren't so minimal, perhaps he would have lived long enough to see commercial fusion reactors using this concept. Even so, it looks like we should finally know whether it works within the next 1.5-2 years. Commercial reactors would follow shortly thereafter.
Whatever things people may like to call "nanotechnology," there is really only one important distinction. Can we assemble atoms in any desired configuration? That is what is commonly termed molecular nanotechnology, and it is what most people originally meant.
Once this and fusion are out of the way, life will start to get very interesting; the foundation of our economic systems will become irrelevant as scarcity will cease to be a useful concept.