The only serious concern left is an astronomical disaster, such as a meteor strike. It seems that the reasonable thing to do would be to focus resources on a defense system for that.
No matter what you do, you are never going to develop a defense system against resource depletion. While many people constantly say that there's plenty of X on Earth for humans to use (where X is anything consumable), they're crazy. Helium, for instance, is a decidedly depletable resource, and one that is being used up quite rapidly. It's doubtful that the Earth's helium supply will last much past the end of the 21st century - and yes, this is true, even with people using helium to lift balloons.
There are plenty of other resources that're being heavily depleted as well. Yes, there are more sources for them, but it will not be economically feasible to recover them. Which, of course, means that they might as well not exist.
Plus there are other disasters to be concerned about: a magnetic pole reversal, an Ice Age, a sudden rise of the sea levels, etc. - none of which you can reasonably protect against. Earth is fragile, and it will always be fragile. It's also not permanent. It will die. It has a finite lifespan in the neighborhood of a few hundred million years left before the oceans boil off. The reasonable thing to do is get the hell off the planet.
Once people migrated to North America, suddenly oceanbound travel started to explode. And likewise, ship technology increased dramatically. There's no reason to believe that the same thing wouldn't happen here.
There's also no reason to believe that colonization wouldn't provide the same benefits it did in the 1600s-1700s: a fresh view of the world from a different perspective.
There've been many people that have said that the reason the Internet boomed so well in the US was due to free local phone access, because the phone infrastructure in the US is so good. This is because the US is a large country with lots of open land - certain technological advances started here because it was best suited for them. There's no reason to believe that a Martian colony wouldn't be subject to the same pressures.
The point is that human beings do best in adversity - "necessity is the mother of invention." There are surely people working on radiation treatments, space health issue, space transport mechanisms, etc., but there's no real need for them now. If there IS a need, then those sectors of science will literally explode, and the secondary benefits will be very hard to imagine.
It's important to realize that one can -never- estimate the benefit of a colony to the home country, virtually by design - a colony is a new settlement, with new needs, and new ideas. And nothing - nothing - is more valuable to the human race than new ideas.
So maybe you're right. Maybe off-planet colonization isn't the answer to all of our problems. But it might be the answer to a whole, whole lot of them. You simply don't know until you go there, and find out.
The weird thing about this is that you can obviously reverse time, and say "how much does Jupiter have to contract to release as much energy as it does?" and "how big must it have been?" When you do that, if Jupiter's output is even roughly of the same order of magnitude as it is now, Jupiter must have been on the order of the size of -the Sun- as recently as the Jurassic.
That's just weird to think about - looking up in the sky at night and seeing a visibly large disk. Not as large as the Sun or the Moon, obviously - Jupiter's about what, 4-6 times farther away? - but still decidedly very, very big.
No, it isn't. The most abundant elements in the Universe are H, He, C, N, and O, and then "everything else". H (and deuterons) is what everything started out as. He is the end product of the pp chain, along with trace amounts of lithium that are so trace they don't matter. After that, the next chain is the triple-alpha process (3 He -> C,N,O) which is why C, N, O are all around.
We have rocky planets because this close to the Sun, H, He, C, N, and O are all gaseous/liquid (well, most of their compounds are). Besides, that's what you want. Solids can't move around very well, so they can't form new compounds easily.
In addition, silicon isn't chemically identical to carbon. It's too much of a metal - the electrons in its valence shell are too weakly bound, so it tends to ionize pretty easily. Silicon CAN form long chains, but the Si-Si bond is much weaker than the C-C bond, meaning that these long chains are much more fragile than carbon chains, which produce things like diamond and carbon nanotubes.
Molecules aren't fragile - electromagnetic fields are (comparatively). The problem isn't existence - it's formation. Same reason that silicon life, even simple, would never exist: carbon life would always beat it out, because any place you find silicon, you find carbon (carbon is made in an earlier point of stellar nucleosynthesis, and it's thousands of times more abundant than silicon).
Electromagnetic fields aren't long lived like molecules are - if you form a bit of amino acid, it isn't going to go away easily. The environment that allowed it to form has to be relatively 'benign' to it (since it allowed it to form) and thus it should survive long enough to encounter other amino acids, etc.
I'm not saying it's not possible. It's just not likely. And in the space of the universe, it probably does exist. But chemical based life is much easier - it's more stable - and therefore much more likely, and easier to find.
We use calcium as a skeleton, and corals use other things as skeletons as well. But when I said skeleton, I meant -chemical- skeleton, not physical skeleton: you need a strong-linked chain that you can build really huge molecules out of, and then tack wacko things like nucleic acids, phosphorus groups, or iron onto. Carbon's the only way to go. Silicon can't form long chains.
Carbon dioxide is not a dipole (at least, not to first order). The carbon is slightly positive, and the oxygens are slightly negative, but the oxygens are equally distributed from the center of the molecule. No dipole moment. Won't dissolve things anywhere near as powerfully as water does.
Water's often referred to as the "universal solvent" - it is very good at its job.
Eh? You're giving water too much credit, now. The stuff on which all our beloved complex molecules depend is carbon--water is just a useful solvent. In and of itself, water creates structures like ice crystals or cages (e.g. clathrates). Interesting for a number of reasons, sometimes very pretty, but not particularly 'complex' in the sense that you mean. For us, water is a very nice solvent because it is polar (you can dissolve ions in it) and amphoteric (it can act as a proton donor or acceptor depending on ambient conditions). Liquid ammonia (NH_3) would do almost as well, in principle.
Water is the solvent. It's what allows complex chemistries to exist. As I said, -maybe- ammonia, but ammonia lacks oxygen, and the reactiveness of oxygen is a major benefit to complex chemistries. The other thing to remember is that wacko liquids of more than about 2 elements just aren't going to happen. The phase space is too small.
I'm not quite sure what you're getting at here. HX--where X is anything but hydrogen--is always a dipole.
Sorry - missed a word - covalent dipole. HF, HCl, etc. are ionic, not covalent. Fluorine, chlorine, etc. are too electronegative - I think the higher-ups might be covalent, but they're Covalent dipoles are 'stable': in a liquid form, they remain dipoles, rather than just charges migrating around. Simply put, water doesn't dissociate easily.
Because there aren't any simple atoms that we haven't discovered yet. We know them all. Water takes two hydrogen atoms, and one oxygen atom. A liquid that we haven't discovered yet would be very complicated, and thus, not likely to form.
We might not 'know' that water+carbon is the only way to go, but being 'pretty incredibly damned sure' is good enough, sometimes.
And Carbon doesn't have to be the building blocks of life forms, it just so happens it has good properties for such on earth. Elsewhere in the universe, Carbon may not be as easily found in solid form...
Care to propose another one? Life doesn't care whether or not carbon, phosphorus, nickel, or molybdenum is liquid, solid, or encased in a purple gel. It'll -make- it liquid, it just needs to have it, and of all of the elements on the periodic table, carbon is unique. You think it's a coincidence that life is almost entirely made out of C, H, N, and O, and that all of those elements are within the first and second row of the periodic table?
The properties that carbon has aren't physical - they're chemical. It's simple physics. Nothing you can say or do will ever make any other atom like carbon, and even if some wacko silicon based life started forming in an area, carbon life would rapidly outstrip it, because it simply is better suited to the task. More combinations. Silicon life would never be complicated enough to beat out carbon-based, and you will -never- find a place in the universe that has silicon, but no carbon. Carbon's about the 3rd most abundant element in the universe (actually, CNO are roughly the same).
Life didn't base itself on water and carbon on Earth because of dumb luck. It based itself on water and carbon because it's the best way to do it. The -only- other plausible life combination might be carbon and ammonia, but it may be that there's a gotcha somewhere in the chemistry there. (Lacking large amounts of oxygen is definitely a downer, chemistry wise - yes, I know it's caustic and abrasive, but that's what you want in a life-based chemistry - something that generates a ton of reactions).
That's not to say that life cannot exist without water, but it certainly makes life much more plausible.
You're not giving water enough credit. Basically, the important thing is to qualify what is life: life is the creation of complex systems that can adapt and increase in complexity over time. That's a decent definition of life - it excludes fire, for one, which is always a difficult one. In order to satisfy that definition, you need a framework which allows you tons of complexity, which is what water gives you. Gotta love water.
Water is the simplest dipole that can form. You can't make a dipole out of HX, and if you want H2X, water's the easiest choice. Is it really any amazing wonder that nature, needing a dipole (which allows for complex arrangements), chose the simplest one? Hmm. Bout as surprising that the elements used in life happen to be the most common in the universe (barring helium).
-Maybe- ammonia. Maybe.
Life -needs- a dipole. Life also needs a 'backbone' - a framework. Carbon's your only choice for that.
A carbon-silicon combo might work.
Why in the world would life EVER use silicon, when carbon is so much more abundant than it, and will be no matter where you go in the universe, and carbon doesn't need silicon? All it does is weaken the structure.
No, but I could exceed the speed of light via a poor explanation really easily.
Want to get to Alpha Centauri in, say, a year? No problem - that's about gamma = 4, or about 0.97 c. Of course, someone not familiar with relativity would say that you're travelling at 4c, but that's because of an improper definition of distance - in truth, you only traveled 1 light year, not 4 - the distance Lorentz-contracts.
Same situation here. Standard stupid way of calculating speeds against the sky don't work because your estimation of what the distance travelled is is wrong. It's just geometry.
Here for the math. You're not tacking. It's assuming that a distance in one reference frame holds in another, highly relativistic frame.
There are plenty of places to attack standard current astrophysics: the measurement of the fine structure constant changing? Very weak. SN 1a distance determination? A little odd, considering we don't really understand all the classifications we have. But superluminal motion is just a reference frame mistake.
Definition, actually. Quasars that don't have their beams pointed at us (and, incidentally, they have 2-pi steradians to point at, not 4-pi: they have two beams, not one) aren't called quasars. They're called radio galaxies. Still the same object.
Quasars/blazars/BL Lacs/radio galaxies/Seyfert galaxies are all (now) beginning to be understood to be the same thing, just looked at from a different point of view.
Apparent superluminal motion really is just a completely normal point-of-view problem. It's a standard undergraduate problem in relativity.
Because the definition of a "particle" in this case is completely arbitrary. It's a charm and an antistrange quark in a bound configuration. You can 'imagine' it as the charm and the antistrange orbiting each other (though this isn't strictly true!) There's a 'ground state' for a D_s+, which is like 1970 MeV as I said above. This is an 'excited state'. In particle physics you call excited states new 'particles'.
We know we didn't 'invent' it because a c and an s_bar existed a long time before this guy. We just put them together in a weird way.
Might be a little too early in the morning. This guy is just a resonance of D_s+, which has a mass of 1968 MeV, and also is made up of c & s_bar. Naked charm just means the particle has a c and no c_bar, which is perfectly fine. Charmed particles have been around for a while now.
But what's really interesting is that all of the reasons for their decisions as well a a thourough discussion of other possible excemptions which were considered but not selected! Included among these are most of slashdot's favorites, such as DVDs, video games, reverse engineering, research, etc. Yes it's a government document and looks intimidating, but it's well worth a close read by everybody. As noted a new round of hearings is underway; I encourage everybody to read up on why many exemptions failed last time around and what needs to be done to present a better case next time.
Actually, DVDs, video games are protected under that clause.
"protected by access control mechanisms that fail to permit access because of malfunction, damage, or obsoleteness."
If a video game fails to boot because it is scratched, then it stands to reason that you can copy it to circumvent the copy protection that is damaged - it's a legitimate work that you're trying to access, and it's preventing you from doing so.
It also allows you to circumvent copy protection on a DVD if DVD players become obsolete. You could probably make a case for circumventing CD protection in a few years to rip to MP3, as well.
It's a bit of a stretch, but I think it'd stand. The first clause actually covers a lot more of what people would consider "fair use" than one would think.
Not true. USB needs memory. RS232 does not. It generates a byte stream. Hell, RS232 UARTs even typically have a scratch register you can use if the system you're on doesn't have one.
As for how this applies to PCs: there are many points in a PC's boot cycle that the above situations may apply. Before the processor is initialized, before memory is initialized, etc. USB takes too much work - from a hardware point of view - to get it up and running. Serial just works right out of the box.
Re:Bad Things about Legacy HW
on
Legacy-Free PCs
·
· Score: 1
Oh, and the whole "IRQ", device driver, etc. thing is just nuts. The IRQ implementation is a BIOS thing that's there for compatibility. You don't have to get rid of the COM ports to get rid of the IRQ implementation. Just fix the BIOS.
As for the device driver: RS232 isn't for permanent peripherals. It's for communication. Plug it in, communicate, and unplug it. If it's a persistent communication, then you have a device driver, sure. But I have *never* seen a device which accesses the serial port have problems unless the machine is set up weird (as in, multiplexing with IrDA).
You don't need to write a device driver for RS232. That's what kicks butt. Under linux, it's just getc and putc. Just that simple. No device setup, no nothing. I can write a program which talks to another device over a serial link in five minutes. USB? Takes a bit longer. Say it with me if I don't need what USB offers, why should I go through the hoops?
RS232 is here to stay. The PC industry will have to violently and vehemently yank the rest of the world along to get rid of a beautifully simple-to-implement standard.
Re:Bad Things about Legacy HW
on
Legacy-Free PCs
·
· Score: 1
Takes up Connector Space
Many, many ways around this. Have a "port extender" for laptops. Move all the legacy ports to a removable header on motherboards. Don't get rid of them on the motherboard.
They don't take up space on the motherboard. Not significantly. Not on laptops, Palm Pilots, or anything else. RS232 interfaces are implemented in silicon on most of these things! All you need is the connector, and for that, a 2x5 header takes up remarkably little space.
Power Supply issues - Each interface type that needs yet another voltage level is going to affect the power supply and/or motherboard. It's much nicer if you can avoid all that stuff.
RS232 (or EIA-232)'s voltage levels are there for a reason. They're very noise tolerant. They work in nasty environments. A TTL-to-RS232 converter is cheap, small (a DS235 - I think that's what it is...) and low power, and doesn't require extra voltages. They're neat that way.
I don't know whether or not a USB implementation is power-competitive with an RS232 implementation. I would guess not - you need very very little to interface to RS232, and it's very very tolerant of "poor" implementations ("I'd like 12V, but if you give me about 4, I'll be happy!"). USB isn't, so it won't be able to "skimp" as much as RS232 does.
For instance, that DS235 (It's a Dallas Semiconductor RS232 transceiver... can't remember if that's the right part number) uses the negative voltage of the communcating partner to generate the negative voltages, and sneaks power from there when it can. If the partner only goes to -4V, then it will too.
Oh, and the speed thing is just plain wrong - large cache UARTs that do 1Mbps+ have been available for a long time now. You don't have them on motherboards because you don't need them.
This is a common argument, and it is also wrong. Legacy components are addressed via a completely modern addressing scheme. Look it up - they're all attached to the LPC bus, which is a low-pin count bus that connects to all the legacy devices. It's not ISA, and it's not addressed like ISA.
The 16-bit thing is completely wrong, unfortunately - addressing 16-bit devices is as simple as declaring (unsigned short *) rather than (unsigned int *), and then a couple of address lines acting differently.
Now, if it's implemented differently in a BIOS, that's the BIOS's fault, not the device's.
Complain to the motherboard manufacturers. Get them to have an all-USB (et al) backplane with a few headers on the board to go out to a PCI cover slot that contains the legacy stuff. If you don't want it, don't use it.
Removing it from the board is what bothers me. Also removing it from laptops, and there your argument would hold water - thin laptops simply are thinner than a DB-9. But it's still annoying. Laptops are exactly when you -want- a simple serial port.
Though with those, you could just put the legacy components on an (unpowered) "device expander". That'd work.
Re:Why USB is better than UART
on
Legacy-Free PCs
·
· Score: 1
See points to counter all of his.
And last time I checked, the distance limitation is a killer. I've had a 40 foot long serial cable. I simply couldn't've used a USB implementation there.
(And don't suggest USB->serial converters. They're such hacks, they barely work right, and more importantly, it's so trivial to implement on a motherboard there's no reason to get rid of them for the benefit they offer.)
Re:Why USB is better than UART
on
Legacy-Free PCs
·
· Score: 1
OK, I could debunk about -all- of those statements, but I'll keep it down to a few.
Standard serial ports don't have a power supply with a well-specified current budget (you have to use wierd parasitic power supplies that don't always work on laptop serial ports).
+plug and unplug +"device types" +"hubs" +"isochronous transfer modes"
What the hell? Serial ports are for communication, not powering things. They shouldn't supply power. Yes, in certain cases, they used to, but that was bad design because USB didn't exist. Now it does. Device needs power -> use USB.
They're also not for interfacing with generic devices. They're for communication. You talk over them with RS232. That's it. That's all. That's all they do, that's all they SHOULD do. If you want more, use USB. But if all you want is to send characters back and forth slowly, you shouldn't need to go through the gigantic hoops that USB offers.
Serial ports require negative voltages (more workarounds with switched-capacitor inverters)... because you get better noise performance that way. RS232 works in more "hostile" environments than USB does.
Serial ports require an interrupt per byte and are connected on the legacy ISA bus - each I/O cycle takes nearly a microsecond (thousands of cycles on a modern PC!). A USB controller is a bus-mastering PCI device with a scheduler driven by table data structures.
This is simply wrong. This is the implementation of a serial port from years gone by. Modern 16750's support DMA transfer to/from their buffers and are easily interfaceable to modern systems. Typically serial ports now go through the LPC bus rather than the ISA bus, and those overheads are completely wrong. The IRQ requirement is a BIOS limitation.... problems with DB-9 connectors
I'm sorry, but your statements are really quite bad. The retaining screws are there for a reason - to ensure complete mating. While USB is well designed, it's not as rugged a connector as a DB9 is.
DB-9s do have recessed pins. Grab a male serial port. The shell is in front of the pins. Ground will mate before any of the pins mate. The whole "USB has better ESD because of its connector!" is just bull. As for the hot-insertion and removal, RS232 doesn't HAVE power and ground pins. It has communication pins.
So that's the basic response. USB is an improvement upon RS232, but it is not a replacement for it. If you don't need USB, you shouldn't have to have it. UARTs are just -way- too easy to interface to. When I say "interface", I mean "hardware", not "software". If your BIOS is stupid, complain to the motherboard manufacturer, not the UART manufacturer. I've designed interfaces to UARTs. It's really really easy. You can even get purely serial UARTS so the total pin count is really low.
The day UARTs get end-of-lifed is the day I sigh in agony. I'm already annoyed that DRAM is going away. Not all of us want the whiz-bang superfast crap - we just want to make hardware that works that we don't have to spend a long time worrying about stray capacitances and such.
So, to sum up:
Serial ports: good Devices which use the serial port in bizarre ways (grabbing power, using them for something other than simple communication): bad USB: good
No. Because an idiot can integrate a UART. It's known hardware. You know what it needs. You know how to integrate it. In fact, in a lot of cases, it gets integrated into silicon inside ASICs, and all people know is "hook this up to a connector."
Serial ports, parallel ports, legacy ports, etc. - all cheap, and easy. Not as general purpose as USB, true - but not everything needs what USB offers.
Jeez, do you know how many ISA->IDE adapters I have lying around? Dozens. Every computer had them through 486s. And as for only being able to access the first 512M only, wouldn't the hacks like EZ-BIOS etc. work there as well?
Re:Unfortunately...
on
Legacy-Free PCs
·
· Score: 5, Insightful
OK. Ignore the x86 architecture bit completely, and just look at the add-ons you're addressing - ISA slots, floppy drives, RS232 ports, parallel ports, PS/2 keyboard/mouse adapters.
For instance, RS232 ports: What exactly is wrong with an RS232 port? Why is it "worse" than a USB port? There's no difficulty in actually using an RS232 port - UARTs are cheap, they're brain-dead easy to interface to, and they support rather modern interface methods (DMA, etc.). They are, however, low speed - but of course, for low speed operations that's all you need. You will never need high-speed data transfer to your keyboard or mouse - they're inherently low data transfer devices, since humans are slow.
Same goes for ISA slots and parallel ports. They don't hold back the state of the art. They're add-ons. If you don't use them, they don't do anything. It's just a memory space that doesn't get accessed. If you're complaining about their implementation on current PCs (the fact that they sit in I/O space, take up IRQs, etc.) then you're complaining about the BIOS, not the peripherals. I really didn't see the point of replacing the PS/2 keyboard and mouse. They're just serial devices - they interface via the same method that UARTs, etc. get addressed, which is ridiculously easy to interface to.
There is nothing fundamentally wrong with legacy components. Interfacing to a UART is trivial. Much more trivial than with USB, in fact. There's no reason a "clean" design of a PC couldn't have a serial port, ISA slot, ATA hard drives, and everything else.
Even the x86 architecture thing is 'not that bad'. Take the x87 architecture - everyone complains about the FXCH instruction, because it IS stupid, but on the P3 and Athlon (but not the P4 - one reason the P4's FP sucks) that instruction's 'free' - it takes 0 clock cycles to process. There's some overhead involved with it, but it's not clear to me that the small gain from fixing the overhead loss would offset the large loss of not being compatible with large portions of x87 software. And it's not clear to me that the overhead couldn't be compensated for in some other way, as well.
Slashdot Mirror
The only serious concern left is an astronomical disaster, such as a meteor strike. It seems that the reasonable thing to do would be to focus resources on a defense system for that.
No matter what you do, you are never going to develop a defense system against resource depletion. While many people constantly say that there's plenty of X on Earth for humans to use (where X is anything consumable), they're crazy. Helium, for instance, is a decidedly depletable resource, and one that is being used up quite rapidly. It's doubtful that the Earth's helium supply will last much past the end of the 21st century - and yes, this is true, even with people using helium to lift balloons.
There are plenty of other resources that're being heavily depleted as well. Yes, there are more sources for them, but it will not be economically feasible to recover them. Which, of course, means that they might as well not exist.
Plus there are other disasters to be concerned about: a magnetic pole reversal, an Ice Age, a sudden rise of the sea levels, etc. - none of which you can reasonably protect against. Earth is fragile, and it will always be fragile. It's also not permanent. It will die. It has a finite lifespan in the neighborhood of a few hundred million years left before the oceans boil off. The reasonable thing to do is get the hell off the planet.
Once people migrated to North America, suddenly oceanbound travel started to explode. And likewise, ship technology increased dramatically. There's no reason to believe that the same thing wouldn't happen here.
There's also no reason to believe that colonization wouldn't provide the same benefits it did in the 1600s-1700s: a fresh view of the world from a different perspective.
There've been many people that have said that the reason the Internet boomed so well in the US was due to free local phone access, because the phone infrastructure in the US is so good. This is because the US is a large country with lots of open land - certain technological advances started here because it was best suited for them. There's no reason to believe that a Martian colony wouldn't be subject to the same pressures.
The point is that human beings do best in adversity - "necessity is the mother of invention." There are surely people working on radiation treatments, space health issue, space transport mechanisms, etc., but there's no real need for them now. If there IS a need, then those sectors of science will literally explode, and the secondary benefits will be very hard to imagine.
It's important to realize that one can -never- estimate the benefit of a colony to the home country, virtually by design - a colony is a new settlement, with new needs, and new ideas. And nothing - nothing - is more valuable to the human race than new ideas.
So maybe you're right. Maybe off-planet colonization isn't the answer to all of our problems. But it might be the answer to a whole, whole lot of them. You simply don't know until you go there, and find out.
The weird thing about this is that you can obviously reverse time, and say "how much does Jupiter have to contract to release as much energy as it does?" and "how big must it have been?" When you do that, if Jupiter's output is even roughly of the same order of magnitude as it is now, Jupiter must have been on the order of the size of -the Sun- as recently as the Jurassic.
That's just weird to think about - looking up in the sky at night and seeing a visibly large disk. Not as large as the Sun or the Moon, obviously - Jupiter's about what, 4-6 times farther away? - but still decidedly very, very big.
No, it isn't. The most abundant elements in the Universe are H, He, C, N, and O, and then "everything else". H (and deuterons) is what everything started out as. He is the end product of the pp chain, along with trace amounts of lithium that are so trace they don't matter. After that, the next chain is the triple-alpha process (3 He -> C,N,O) which is why C, N, O are all around.
We have rocky planets because this close to the Sun, H, He, C, N, and O are all gaseous/liquid (well, most of their compounds are). Besides, that's what you want. Solids can't move around very well, so they can't form new compounds easily.
In addition, silicon isn't chemically identical to carbon. It's too much of a metal - the electrons in its valence shell are too weakly bound, so it tends to ionize pretty easily. Silicon CAN form long chains, but the Si-Si bond is much weaker than the C-C bond, meaning that these long chains are much more fragile than carbon chains, which produce things like diamond and carbon nanotubes.
Molecules aren't fragile - electromagnetic fields are (comparatively). The problem isn't existence - it's formation. Same reason that silicon life, even simple, would never exist: carbon life would always beat it out, because any place you find silicon, you find carbon (carbon is made in an earlier point of stellar nucleosynthesis, and it's thousands of times more abundant than silicon).
Electromagnetic fields aren't long lived like molecules are - if you form a bit of amino acid, it isn't going to go away easily. The environment that allowed it to form has to be relatively 'benign' to it (since it allowed it to form) and thus it should survive long enough to encounter other amino acids, etc.
I'm not saying it's not possible. It's just not likely. And in the space of the universe, it probably does exist. But chemical based life is much easier - it's more stable - and therefore much more likely, and easier to find.
We'll look for the evil energy creatures later.
We use calcium as a skeleton, and corals use other things as skeletons as well. But when I said skeleton, I meant -chemical- skeleton, not physical skeleton: you need a strong-linked chain that you can build really huge molecules out of, and then tack wacko things like nucleic acids, phosphorus groups, or iron onto. Carbon's the only way to go. Silicon can't form long chains.
Carbon dioxide is not a dipole (at least, not to first order). The carbon is slightly positive, and the oxygens are slightly negative, but the oxygens are equally distributed from the center of the molecule. No dipole moment. Won't dissolve things anywhere near as powerfully as water does.
Water's often referred to as the "universal solvent" - it is very good at its job.
Eh? You're giving water too much credit, now. The stuff on which all our beloved complex molecules depend is carbon--water is just a useful solvent. In and of itself, water creates structures like ice crystals or cages (e.g. clathrates). Interesting for a number of reasons, sometimes very pretty, but not particularly 'complex' in the sense that you mean. For us, water is a very nice solvent because it is polar (you can dissolve ions in it) and amphoteric (it can act as a proton donor or acceptor depending on ambient conditions). Liquid ammonia (NH_3) would do almost as well, in principle.
Water is the solvent. It's what allows complex chemistries to exist. As I said, -maybe- ammonia, but ammonia lacks oxygen, and the reactiveness of oxygen is a major benefit to complex chemistries. The other thing to remember is that wacko liquids of more than about 2 elements just aren't going to happen. The phase space is too small.
I'm not quite sure what you're getting at here. HX--where X is anything but hydrogen--is always a dipole.
Sorry - missed a word - covalent dipole. HF, HCl, etc. are ionic, not covalent. Fluorine, chlorine, etc. are too electronegative - I think the higher-ups might be covalent, but they're Covalent dipoles are 'stable': in a liquid form, they remain dipoles, rather than just charges migrating around. Simply put, water doesn't dissociate easily.
Because there aren't any simple atoms that we haven't discovered yet. We know them all. Water takes two hydrogen atoms, and one oxygen atom. A liquid that we haven't discovered yet would be very complicated, and thus, not likely to form.
We might not 'know' that water+carbon is the only way to go, but being 'pretty incredibly damned sure' is good enough, sometimes.
And Carbon doesn't have to be the building blocks of life forms, it just so happens it has good properties for such on earth. Elsewhere in the universe, Carbon may not be as easily found in solid form...
Care to propose another one? Life doesn't care whether or not carbon, phosphorus, nickel, or molybdenum is liquid, solid, or encased in a purple gel. It'll -make- it liquid, it just needs to have it, and of all of the elements on the periodic table, carbon is unique. You think it's a coincidence that life is almost entirely made out of C, H, N, and O, and that all of those elements are within the first and second row of the periodic table?
The properties that carbon has aren't physical - they're chemical. It's simple physics. Nothing you can say or do will ever make any other atom like carbon, and even if some wacko silicon based life started forming in an area, carbon life would rapidly outstrip it, because it simply is better suited to the task. More combinations. Silicon life would never be complicated enough to beat out carbon-based, and you will -never- find a place in the universe that has silicon, but no carbon. Carbon's about the 3rd most abundant element in the universe (actually, CNO are roughly the same).
Life didn't base itself on water and carbon on Earth because of dumb luck. It based itself on water and carbon because it's the best way to do it. The -only- other plausible life combination might be carbon and ammonia, but it may be that there's a gotcha somewhere in the chemistry there. (Lacking large amounts of oxygen is definitely a downer, chemistry wise - yes, I know it's caustic and abrasive, but that's what you want in a life-based chemistry - something that generates a ton of reactions).
That's not to say that life cannot exist without water, but it certainly makes life much more plausible.
You're not giving water enough credit. Basically, the important thing is to qualify what is life: life is the creation of complex systems that can adapt and increase in complexity over time. That's a decent definition of life - it excludes fire, for one, which is always a difficult one. In order to satisfy that definition, you need a framework which allows you tons of complexity, which is what water gives you. Gotta love water.
Water is the simplest dipole that can form. You can't make a dipole out of HX, and if you want H2X, water's the easiest choice. Is it really any amazing wonder that nature, needing a dipole (which allows for complex arrangements), chose the simplest one? Hmm. Bout as surprising that the elements used in life happen to be the most common in the universe (barring helium).
-Maybe- ammonia. Maybe.
Life -needs- a dipole. Life also needs a 'backbone' - a framework. Carbon's your only choice for that.
A carbon-silicon combo might work.
Why in the world would life EVER use silicon, when carbon is so much more abundant than it, and will be no matter where you go in the universe, and carbon doesn't need silicon? All it does is weaken the structure.
Carbon's a given - it's the only one that'd work.
No, but I could exceed the speed of light via a poor explanation really easily.
Want to get to Alpha Centauri in, say, a year? No problem - that's about gamma = 4, or about 0.97 c. Of course, someone not familiar with relativity would say that you're travelling at 4c, but that's because of an improper definition of distance - in truth, you only traveled 1 light year, not 4 - the distance Lorentz-contracts.
Same situation here. Standard stupid way of calculating speeds against the sky don't work because your estimation of what the distance travelled is is wrong. It's just geometry.
Here for the math. You're not tacking. It's assuming that a distance in one reference frame holds in another, highly relativistic frame.
There are plenty of places to attack standard current astrophysics: the measurement of the fine structure constant changing? Very weak. SN 1a distance determination? A little odd, considering we don't really understand all the classifications we have. But superluminal motion is just a reference frame mistake.
Definition, actually. Quasars that don't have their beams pointed at us (and, incidentally, they have 2-pi steradians to point at, not 4-pi: they have two beams, not one) aren't called quasars. They're called radio galaxies. Still the same object.
Quasars/blazars/BL Lacs/radio galaxies/Seyfert galaxies are all (now) beginning to be understood to be the same thing, just looked at from a different point of view.
Apparent superluminal motion really is just a completely normal point-of-view problem. It's a standard undergraduate problem in relativity.
Because the definition of a "particle" in this case is completely arbitrary. It's a charm and an antistrange quark in a bound configuration. You can 'imagine' it as the charm and the antistrange orbiting each other (though this isn't strictly true!) There's a 'ground state' for a D_s+, which is like 1970 MeV as I said above. This is an 'excited state'. In particle physics you call excited states new 'particles'.
We know we didn't 'invent' it because a c and an s_bar existed a long time before this guy. We just put them together in a weird way.
Might be a little too early in the morning. This guy is just a resonance of D_s+, which has a mass of 1968 MeV, and also is made up of c & s_bar. Naked charm just means the particle has a c and no c_bar, which is perfectly fine. Charmed particles have been around for a while now.
But what's really interesting is that all of the reasons for their decisions as well a a thourough discussion of other possible excemptions which were considered but not selected! Included among these are most of slashdot's favorites, such as DVDs, video games, reverse engineering, research, etc. Yes it's a government document and looks intimidating, but it's well worth a close read by everybody. As noted a new round of hearings is underway; I encourage everybody to read up on why many exemptions failed last time around and what needs to be done to present a better case next time.
Actually, DVDs, video games are protected under that clause.
"protected by access control mechanisms that fail to permit access because of malfunction, damage, or obsoleteness."
If a video game fails to boot because it is scratched, then it stands to reason that you can copy it to circumvent the copy protection that is damaged - it's a legitimate work that you're trying to access, and it's preventing you from doing so.
It also allows you to circumvent copy protection on a DVD if DVD players become obsolete. You could probably make a case for circumventing CD protection in a few years to rip to MP3, as well.
It's a bit of a stretch, but I think it'd stand. The first clause actually covers a lot more of what people would consider "fair use" than one would think.
Not true. USB needs memory. RS232 does not. It generates a byte stream. Hell, RS232 UARTs even typically have a scratch register you can use if the system you're on doesn't have one.
As for how this applies to PCs: there are many points in a PC's boot cycle that the above situations may apply. Before the processor is initialized, before memory is initialized, etc. USB takes too much work - from a hardware point of view - to get it up and running. Serial just works right out of the box.
Oh, and the whole "IRQ", device driver, etc. thing is just nuts. The IRQ implementation is a BIOS thing that's there for compatibility. You don't have to get rid of the COM ports to get rid of the IRQ implementation. Just fix the BIOS.
As for the device driver: RS232 isn't for permanent peripherals. It's for communication. Plug it in, communicate, and unplug it. If it's a persistent communication, then you have a device driver, sure. But I have *never* seen a device which accesses the serial port have problems unless the machine is set up weird (as in, multiplexing with IrDA).
You don't need to write a device driver for RS232. That's what kicks butt. Under linux, it's just getc and putc. Just that simple. No device setup, no nothing. I can write a program which talks to another device over a serial link in five minutes. USB? Takes a bit longer. Say it with me if I don't need what USB offers, why should I go through the hoops?
RS232 is here to stay. The PC industry will have to violently and vehemently yank the rest of the world along to get rid of a beautifully simple-to-implement standard.
Takes up Connector Space
Many, many ways around this. Have a "port extender" for laptops. Move all the legacy ports to a removable header on motherboards. Don't get rid of them on the motherboard.
They don't take up space on the motherboard. Not significantly. Not on laptops, Palm Pilots, or anything else. RS232 interfaces are implemented in silicon on most of these things! All you need is the connector, and for that, a 2x5 header takes up remarkably little space.
Power Supply issues - Each interface type that needs yet another voltage level is going to affect the power supply and/or motherboard. It's much nicer if you can avoid all that stuff.
RS232 (or EIA-232)'s voltage levels are there for a reason. They're very noise tolerant. They work in nasty environments. A TTL-to-RS232 converter is cheap, small (a DS235 - I think that's what it is...) and low power, and doesn't require extra voltages. They're neat that way.
I don't know whether or not a USB implementation is power-competitive with an RS232 implementation. I would guess not - you need very very little to interface to RS232, and it's very very tolerant of "poor" implementations ("I'd like 12V, but if you give me about 4, I'll be happy!"). USB isn't, so it won't be able to "skimp" as much as RS232 does.
For instance, that DS235 (It's a Dallas Semiconductor RS232 transceiver... can't remember if that's the right part number) uses the negative voltage of the communcating partner to generate the negative voltages, and sneaks power from there when it can. If the partner only goes to -4V, then it will too.
Oh, and the speed thing is just plain wrong - large cache UARTs that do 1Mbps+ have been available for a long time now. You don't have them on motherboards because you don't need them.
This is a common argument, and it is also wrong. Legacy components are addressed via a completely modern addressing scheme. Look it up - they're all attached to the LPC bus, which is a low-pin count bus that connects to all the legacy devices. It's not ISA, and it's not addressed like ISA.
The 16-bit thing is completely wrong, unfortunately - addressing 16-bit devices is as simple as declaring (unsigned short *) rather than (unsigned int *), and then a couple of address lines acting differently.
Now, if it's implemented differently in a BIOS, that's the BIOS's fault, not the device's.
Complain to the motherboard manufacturers. Get them to have an all-USB (et al) backplane with a few headers on the board to go out to a PCI cover slot that contains the legacy stuff. If you don't want it, don't use it.
Removing it from the board is what bothers me. Also removing it from laptops, and there your argument would hold water - thin laptops simply are thinner than a DB-9. But it's still annoying. Laptops are exactly when you -want- a simple serial port.
Though with those, you could just put the legacy components on an (unpowered) "device expander". That'd work.
See points to counter all of his.
And last time I checked, the distance limitation is a killer. I've had a 40 foot long serial cable. I simply couldn't've used a USB implementation there.
(And don't suggest USB->serial converters. They're such hacks, they barely work right, and more importantly, it's so trivial to implement on a motherboard there's no reason to get rid of them for the benefit they offer.)
OK, I could debunk about -all- of those statements, but I'll keep it down to a few.
.. because you get better noise performance that way. RS232 works in more "hostile" environments than USB does.
... problems with DB-9 connectors
Standard serial ports don't have a power supply with a well-specified current budget (you have to use wierd parasitic power supplies that don't always work on laptop serial ports).
+plug and unplug
+"device types"
+"hubs"
+"isochronous transfer modes"
What the hell? Serial ports are for communication, not powering things. They shouldn't supply power. Yes, in certain cases, they used to, but that was bad design because USB didn't exist. Now it does. Device needs power -> use USB.
They're also not for interfacing with generic devices. They're for communication. You talk over them with RS232. That's it. That's all. That's all they do, that's all they SHOULD do. If you want more, use USB. But if all you want is to send characters back and forth slowly, you shouldn't need to go through the gigantic hoops that USB offers.
Serial ports require negative voltages (more workarounds with switched-capacitor inverters).
Serial ports require an interrupt per byte and are connected on the legacy ISA bus - each I/O cycle takes nearly a microsecond (thousands of cycles on a modern PC!). A USB controller is a bus-mastering PCI device with a scheduler driven by table data structures.
This is simply wrong. This is the implementation of a serial port from years gone by. Modern 16750's support DMA transfer to/from their buffers and are easily interfaceable to modern systems. Typically serial ports now go through the LPC bus rather than the ISA bus, and those overheads are completely wrong. The IRQ requirement is a BIOS limitation.
I'm sorry, but your statements are really quite bad. The retaining screws are there for a reason - to ensure complete mating. While USB is well designed, it's not as rugged a connector as a DB9 is.
DB-9s do have recessed pins. Grab a male serial port. The shell is in front of the pins. Ground will mate before any of the pins mate. The whole "USB has better ESD because of its connector!" is just bull. As for the hot-insertion and removal, RS232 doesn't HAVE power and ground pins. It has communication pins.
So that's the basic response. USB is an improvement upon RS232, but it is not a replacement for it. If you don't need USB, you shouldn't have to have it. UARTs are just -way- too easy to interface to. When I say "interface", I mean "hardware", not "software". If your BIOS is stupid, complain to the motherboard manufacturer, not the UART manufacturer. I've designed interfaces to UARTs. It's really really easy. You can even get purely serial UARTS so the total pin count is really low.
The day UARTs get end-of-lifed is the day I sigh in agony. I'm already annoyed that DRAM is going away. Not all of us want the whiz-bang superfast crap - we just want to make hardware that works that we don't have to spend a long time worrying about stray capacitances and such.
So, to sum up:
Serial ports: good
Devices which use the serial port in bizarre ways (grabbing power, using them for something other than simple communication): bad
USB: good
Ditto with the parallel port.
No. Because an idiot can integrate a UART. It's known hardware. You know what it needs. You know how to integrate it. In fact, in a lot of cases, it gets integrated into silicon inside ASICs, and all people know is "hook this up to a connector."
Serial ports, parallel ports, legacy ports, etc. - all cheap, and easy. Not as general purpose as USB, true - but not everything needs what USB offers.
Jeez, do you know how many ISA->IDE adapters I have lying around? Dozens. Every computer had them through 486s. And as for only being able to access the first 512M only, wouldn't the hacks like EZ-BIOS etc. work there as well?
OK. Ignore the x86 architecture bit completely, and just look at the add-ons you're addressing - ISA slots, floppy drives, RS232 ports, parallel ports, PS/2 keyboard/mouse adapters.
For instance, RS232 ports: What exactly is wrong with an RS232 port? Why is it "worse" than a USB port? There's no difficulty in actually using an RS232 port - UARTs are cheap, they're brain-dead easy to interface to, and they support rather modern interface methods (DMA, etc.). They are, however, low speed - but of course, for low speed operations that's all you need. You will never need high-speed data transfer to your keyboard or mouse - they're inherently low data transfer devices, since humans are slow.
Same goes for ISA slots and parallel ports. They don't hold back the state of the art. They're add-ons. If you don't use them, they don't do anything. It's just a memory space that doesn't get accessed. If you're complaining about their implementation on current PCs (the fact that they sit in I/O space, take up IRQs, etc.) then you're complaining about the BIOS, not the peripherals. I really didn't see the point of replacing the PS/2 keyboard and mouse. They're just serial devices - they interface via the same method that UARTs, etc. get addressed, which is ridiculously easy to interface to.
There is nothing fundamentally wrong with legacy components. Interfacing to a UART is trivial. Much more trivial than with USB, in fact. There's no reason a "clean" design of a PC couldn't have a serial port, ISA slot, ATA hard drives, and everything else.
Even the x86 architecture thing is 'not that bad'. Take the x87 architecture - everyone complains about the FXCH instruction, because it IS stupid, but on the P3 and Athlon (but not the P4 - one reason the P4's FP sucks) that instruction's 'free' - it takes 0 clock cycles to process. There's some overhead involved with it, but it's not clear to me that the small gain from fixing the overhead loss would offset the large loss of not being compatible with large portions of x87 software. And it's not clear to me that the overhead couldn't be compensated for in some other way, as well.