Re:Efficient,reliable,cheap - chose any 2 :-)
on
NASA's Shuttle Plans
·
· Score: 1
NASA has been doing a lot of good basic research. Things like Ion drives, solar sails, and scram jets which are needed for long term development but don't get as in your face as manned space flight. I think it's best to keep them around as long as they are the only ones interested in doing this type of research.
But, the long-term goal of manned space flight is to make viable off world colonies. However, for that to work they need to be self-sustaining. So we need to develop methods of building everything from glass, to rubber, to CPU's on Mars if we want to truly get off this rock. Until that time it's always going to be cheaper to sit on earth and develop things like fusion than it is to maintain a manned mission in space.
I think there is a lot of value in the research done on how humans react to prolonged weightlessness, but for the near term we need to focus on developing things like biosphere II which will let us live among the stars as apposed to trying to get people to space cheaply or sending tunes of cash to get a few people to sight see on mars for a few months. It's cheaper to send a rocket into space with an experiment than it is to send a person and that experiment up at the same time.
NASA should try and get private companies interested in space but they are doing a great job with things like the mars probes. Even when these projects fail they still provide value if nothing more than keeping people interested in the stars.
Once companies get interested in researching space for profit then let them at it but as long as it's all subsidized by the US Gov then there is little value in having private companies doing basic research.
Fame goes to those who dance in the lights not the behind the seines characters. Lara Craft has same type of fame as Lennon, if not the same level, but the Carmack's of the world are going to be more like George Lucas as apposed to say Keanu Reeves.
The basic rules of real life is about as complex as this game of life. What this game simulates is closer to the chemical reactions of life as apposed to life. You get vary complex behavior not from the rule set but from the size of the board. The simple forms of "life" you see on tiny boards are not going to be anything like what you would see on say 10^1000 X 10^1000 board.
The idea is it's not the specific rule set that changes to evolve life but the level of complexity allowed by the size of the board and a simple rule set. Oxygen, Hydrogen, Carbon,... chains are the building blocks of life. But real life's rule set is not built on the level of those huge and complex changes it's build on the atomic level with simple rules but a HUGE board.
People who want to simulate things don't even try and use the real worlds rule set because it's takes so much space so they design simpler simulations. But that does not change the validity of the simulation. As you can evolve a world from half a page of code if you can run a simulation of that size. Clearly the complexity is a result of evolution not the rule set. Hell the guiding hand is a RND function but you still get increases in complexity over time that's about as close to demonstrating evolution as you can get without simulating a planet for billions of years.
I'm not arguing against QM here, get that straight. vs. I'm saying QM is a new theory, sure it makes a lot of sense, but there may very well be a good reason why I won't instantly teleport across the galaxy.
Ok let's think about this in one case you think about the odds of you specifically popping say 1 foot to the left. Now I agree this is not going to happen in your lifetime. QM gives the type of odds, which would say it's unlikely for you to pop that foot given thousands of times the lifetime of the universe let alone 100 years.
But life is not a specific 'You". Hell if you showed up in the primordial goop you would have quickly died all that this is based on is one of the huge number of possible self replicating life forms showed up at some time over billions of years. On one of billions of worlds in all the worlds oceans. Seeing as that basic life form could be much simpler than you or me or any of the single celled life on this planet right now it is much more probable.
The separation is going from a specific event to a wide range of possible events that are much more likely than you popping even 1 inch to the left vs. teleporting across the galaxy.
PS: I was not trying to imply you had any specific faith. I could have used Buda vs. DC or any other figure but after arguing the same point enough it just started rolling off the fingers as it where. My point is separating the truly un-testable hypothesis from the above line of argument by pointing out not a specific case "JC" but a more general idea of anyone including DC (David C). I also like saying JC vs. DC as sort of a play on AC vs. DC.
On a deep level most people dislike the idea that random chance created the world. People like the idea that life has meaning and there was some creator involved in the mix. However, using that approach solves nothing as either that creator popped into existence, which would seem less likely than a much simpler life form or they evolved from simpler life thus moving evolution back one step. However, I can't really debate these ideas, as there is no way to demonstrate anything about something with godlike power.
ID is like a kid saying "well I know the magician does not know magic so it must be his invisible friend who knows the magic." in that it does not solve the problem merely abstracting it by one level.
I enjoy a good debate. I tend to be fairly proactive in my attack but don't take that to mean that I was angry with what or how you said something.
Anyway cutting though your last rebuttal: I wasn't disputing the knowledge we have about quantum mechanics. All I was saying is that this cannot be used to prove evolution because it is non-falsifiable. Quantum mechanics becomes God in a sense. That's what I meant by it "sounds" religious.
There are several incorrect assumptions in that last statement. As the rest of the argument hangs on this statement "All I was saying is that this cannot be used to prove evolution because it is non-falsifiable." I would like to point out that in scientific debate you don't prove things you try and establish probabilities. I can't prove the world was created 999.999999 seconds ago, but that's not the type of proof science goes for. I say simple life can self assemble with some probability per cubic nanometer of water per second as specific reactions take place. Now each of those stages and assumptions can be tested. I then estimate the age and size of the earth and the number of earth like worlds and say something like "Given those probabilities there is a 95% chance that life should show up within 1 billion years on _ number of worlds. "
At this point you could attack ANY of my assumptions and get me to alter those probability you can't prove I am wrong just as you can't prove the world is older than 1000 seconds but you can demonstrate that it is unlikely. If you can show the odds where worse than say 1 in 20 then I would go ok while the theory was not disproved given those assumptions the theory is probably not true.
PS: Not that I really like semantic arguments but "it is amazing to see so many people absolutely certain that 'infinite' time, space, and energy can explain the origin of everything. Is anyone else uncomfortable about that?" is begging for it. And I am having trouble resisting...
"is amazing to see" that's nice but being surprised does not help your argument.
"'infinite' time, space, and energy can explain the origin of everything." I never said that we had infinite time. But shocking, as it might seem QM implies that with infinite time every possibility will happen. It's one of many things that bothers people but so does the double slit experiment but there is not that says the world has to work in a way that you feel comfortable about.
"I was generalizing." Reduxio ad insertom (SP?) it might win some arguments, but it does not work when there is anything else wrong with your argument.
Anyway, you can't win this type of argument with analogies so please stop trying to use time it's a waste of time.
"but common sense will tell you that that can't happen!"
Wow, that sounds just like; "But common sense tells you that the world is flat!"
You and he both fell for the same trap. There is a huge difference between direct extrapolation from observed phenomenon than your gut feelings. We can all agree that that specific event is so unlikely it's not worth talking about but if you are going to talk about it you have to base your judgment on something. Now I don't know what the world was like 300 years ago let alone 3,000,000,000 years ago. But, I recognize I don't know what's going on back then so I don't say well this story sounds good so that's what I am going to think.
Let say you where to take your fingers and place them close together and in front of a bright light. Now based on your day-to-day world you would expect that nothing unusual is going to happen but... black lines show up. Now clearly something is not going on as you think it should so umm let's say god did it.
Now F != Mass * A but it works most of the time. Light does not really work the way you think it does but your ideas work most of the time. This suggests that our day-to-day understanding of the world does not extrapolate well to things outside of our understanding. But the fundamental difference between trusting scientific theory's and the good book is we understand the theory's to be educated guesses that we can test where the good book is a umm book. Now someone could suggest that good old JC was a hypnotist, but that's not really a testable theory. The thery that people like David Corresh can convince people they need to follow them is. Now that's a big hole in religious argument, but hey have faith you might be right just try not to make things up to "prove" your point.
PS: I am not going to bet that a chair is going to self assemble in my lifetime but I am willing to be CPU's will work most of the time which works under the same principle.
If it was not of "supernatural origin" then it showed up as part of nature. At which point you just added intelligence with no cause, or you are saying that it evolved on another world and then showed up to direct us Ect.
Saying "God did it" is the same as saying it just showed up because at some level God just showed up. It's not an explanation so much as pointing at a magician and saying see he can't do magic it's all done by his invisible friend who can do magic.
it sounds just as religious as any other religious point of view.
It's a direct extrapolation from observed phenomenon. Take a single electron in a "bottle" now it lacks the energy to jump the lid but given time there is a probability that it will jump that lid based on the size of the trap. The problem with your argument is you are saying well ok that works when it's at unit 1 or 10 or 100, but there it is never going to happen when it's unit 10 ^(100 ^ 1000) which not supported by the given evidence.
In the real world you would probably never see something like this happening, but in science you don't get to say this can't happen.
So tell me what do you think the odds of simple life showing up on the earth though random chance. If you say it's unlikely give me some logic given estimated probabilities over volume and time not just a gut feeling that that sounds like faith. If the idea gives you pause then feel free to investigate the subject, make a theory, test it, get it peer reviewed for methodology, and published and you can join the debate. But if your going with an idea because it sounds good then clearly you're the one going on faith.
We are going over a wide range of topics and I don't know which of them you agree with. So picking a single point of attack that you have stated: Here I defend why.
The laws of physics don't change.
Take the game of life on a small board that fallows the basic rule set things get boring vary fast but take a bored that's 10,000 X 10,000 (12 MB of ram) set it up using a random arrangement so 1-25% of the bored is covered. Now to add a dash of mutation on each cycle to give each square a.01% chance to flip. Ok running this you get a simple runner and then "guns" that shoot those runners then a slew of complex gunners. The bored is limited in size so it can only support life of a given complexity but with an undirected and simple rule set you give rise to complex "life".
This is completely undirected and the rules are extremely simple but if you run this simulation with different starting conditions you can see all the basic "valid" life forms show up. The larger the bored and the longer the simulation you run the more complex life you get but the complexity is all centered around the same basic laws of physics for that world.
Now change the starting conditions so the board starts blank and you just use the random mutation factor to add random cells. It takes longer but you still end up with life and given time you end up with all the same life forms.
All this life is limited by the laws of the world but given a large enough board you see predators, prey and even viruses. It's all undirected. The complexity is only limited by the size of the world. More mutation tends to evolve the life faster but over time you should start to see things that try and counteract this mutation. They will always fail some of the time but given time and enough board space it should evolve all the way to intelligent life.
PS: A tick has some level of intelligence as does, a bat, a rat, a dolphin and a human.
The 100CPU chip is based around comparing a P4 with a 8080 and trying to picture a chip that's that much more complex than the P4.
20 years is not that that long in terms of OS evolution. But, in that time frame we should see chips with GB or TB's of cache on them.
Now Intel can try and stick with a small number of CPU's on that chip but it's going to be a lot easer for them to build a chip like a geforce-7800 (http://techreport.com/reviews/2005q2/geforce-7800 gtx/index.x?pg=1) that way they can test each component and cut off the pipelines that don't work. The top of the line chip could have say 10gig's of L2 cache and 500megs of L1 cache split among 100cpu's and with a good design they can still use chips that have significant damage in there Celeron brand which could be based around having 1/2 of those CPU's working with 1/2 the cache.
The basic problem is moving from 4 > 8 > 16 > 32 > 64 bit CPU's is useful but there is little value in a 128bit CPU vs. 2 X 64 bit CPUs. The 64bit CPU can address 2 ^ 32 X 4gigs of memory so it's going to be a while before we need 128bit CPU's but Single instruction Multiple data instructions are only so useful. Now we can always increase the on chip cash but it's a lot easer to have 100 64bit CPU's than it is to use that many transistors on 4 X 64 bit CPUs such that you are not just wasting them.
I don't know if we will ever have 10,000 CPU's on a chip but after a while it's hard to fund something useful to do with all those transistors. And it's going to be hard to build chips if we are still throwing away chips with a small number of defects. Hell, at some point I expect to start using pipelines to verify what other pipelines are doing.
But, I am not a chip designer so I am open to suggestions as to how you think this is going to play out over time.
So ID people accept that mutations are random?
And the earth is several billion years old.
And many species can be traced back to a common ancestor.
But, they feel the need for several ancestors on the root trees? WHY?
All plant cells are basically the same, All vertebrate cells are basically the same, but there is enough difference between plant and animal cells to say that they need to have a different ancestor?
Granted all tree cells are closer to each other than they are to say grass cells so do trees get a common ancestor or just plants? How about Apple trees vs Pines?
Evolution as a thery provides a clean link between all life on earth so which of these are ID people complaning about and what do they find annoying about these links?
laws of nature as we understand them to work cannot explain.
Such as? The human eye is extremely complex but there are a wide variety of eyes in nature all the way down to light sensitive cells on flatworms. But they are all useful and well adapted to the organism. AKA lion eyes are better for them than say frogeyes would be.
"I learned in my earlist years..." that's a simplification which is not realy true. In science everything is always a theory. Something like F = M * A or E = M * C ^2 is still a theory but they have been demonstrated to a level where trying to disprove them seems like a waste of time. Your always going to have a few oddballs that will try and disprove everything but the more established a theory the less time scientists are willing to spend on that specific debate. (But, I am bored at work and willing to waste some time on this topic.)
PS: metal furniture can self assemble see Quantum Mechanics. Now the odds of that happening are hard to comprehend but it's still possible and given sufficient time highly probable. Based on QM saying something can't happen is meaningless but some things are unlikely over a reasonable timeframe so if you want to try and demonstrate that something is extremely unlikely to ever occur on any planet anywhere in the solar system over billions of years that's one thing but saying can't is meaningless.
My previous post was not attacking you. I was pointing out why saying ID vs. Natural Selection was a poor expression of the idea. And if your going to defend Evolution it's best to try and get the ID people to point out a specific problem and why their theory is a better solution.
Anyway, T A saw the light and hopefully some of these people will as well. (Yes, that was meant as a joke.)
Most ID people are aware of that things mutate and natural selection occurs but they like to call this microevolution and when pressed many of them are willing to admit it occurs. However, they like to try and come up with an arbitrary limit that new species can't evolve or simple amino acids can't evolve into simple cells Ect. But in the real world there is a probability that just about anything will happen from being killed by a comet to dieing because all the O2 in the room is not going into your lungs.
Now if they provided some model of the universe, which suggested some reason why evolution was consistently and significantly, to fast we that would be one thing but all I have seen is a lot of hand waving and no real explanation of how the theory of evolution breaks down.
PS: Even Tomas Aquinas recognized building on a foundation of religion is building a house on top of sand.
I have seen vary High Quality Software written by one person. It took a few years to get out but as he never needed to have "design" meetings or talk about how he was changing things so it was vary efferent but a little slow.
Smaller teams are always more efficient but under say 7 - 12 people they tend to take a longer time to get things done.
Some products like windows are never going to get done if one person works on them but smaller teams still tend to be cheaper per feature. So that's Cost VS. Speed issue.
You can skip over some testing and get a Speed and Cost Vs Quality issue.
You can even get several teams and have them work on the same project. Then pick the best one and get Cost vs. Speed and Quality.
Now if you combine the above you can exchange some of Speed, Quality, or Cost to get more of the others.
Re:Efficient,reliable,cheap - chose any 2 :-)
on
NASA's Shuttle Plans
·
· Score: 1
Ok your wrong. Hint: not all NASA launches cost anywhere near what the shuttle does.
He scaled up an in efficient rocket engine so no advantage there.
The idea of using a jet to help things into space had already been used.
He did not deal with any heat dissipation issues because he never got up to Mach 22 / never went into orbit.
They took the lightest pilot they could find and send him up to kiss the sky which is So comparable to sending a ship into orbit that has enough cargo capacity to carry this ship into space.
Now I am all for private companies spending their money to get into space but NASA spend 10 mil on the X prize and as far as I can see they got nothing out of it.
At launch the shuttle is producing 11Gw of thrust, which is an amassing amount of energy. But that's way more thrust than is really needed. Look how the numbers change when you build a craft which:
Uses a jet powered winged first stage to fly closer to the equator and launch at altitude and a little under mach 1. (Look into fuel costs to do this with a 747 vs. rocket fuel. Note: the wings save you from the worst aspects of gravity as the faster your going the lest rocket fuel you waste to keep from falling back trying to accelerate a craft and this is going on while the ship is it's most massive. And, the jet let's you avoid adding all that oxidizer and the structure to hold it, and you get to reduce the amount of low temp rocket fuel your using which saves you even more mass.)
Is only trying to get a 6 people +(7 days food, water, and air) and basically 0 cargo to LEO (Might be a good idea to give all or some of these people Eva capability and parachutes but this adds little to overall mass).
Now at this point the costs of wings on the 2nd / 3rd stage is a hell of a lot less in terms of total fuel but you can now design the ship to do a low temperature re entry on this light craft.
PS: I think building a huge high altitude / speed Para sail * would work much better than this but wings are not that bad of a basic idea if uses in a low temp design. They increase the service area to radiate heat and let you use glide time to lower the amount of incoming heat.
* The goal is to transfer orbital speed to the air and not the craft so a Para sail that glides though the upper atmosphere should be ideal. With the right design you basically added wings at a greatly reduced weight.
What's the practical benefit of wings on a spacecraft?
1) They can provide lift at the beginning of a flight.
2) They can also allow for low temperature reentry.
3) They let you control where and how you land.
4) As a result of 2 and 3 you can make a highly reusable craft.
Now the shuttle only really uses #3. By skipping #2 they have to do a lot of inspections, which waste the value of a reusable craft.
It's not that adding wings to any design is useful, but they are they only way to get to #4 which is the only way to make access to space cheep and safe.
Re:Efficient,reliable,cheap - chose any 2 :-)
on
NASA's Shuttle Plans
·
· Score: 1
Ok 1 team won the prize, but how many failed again? Lots of things look cheeper if you can safely ignore part of their costs.
Of course you do. The question is what do you do when things break? Do you trust the operating system to try to repair itself, or do you shut the OS down, let another OS take over, and call for someone to come take a look at things?
Right now I'd definitely say the latter. Maybe some day things will be different. Maybe some day we'll have strong AI. But even then, there's a limit to the amount of self-fixing that can be done.
It really depends on what your doing. I write real time software and in my problem domain the goal is to keep going if at all possible. Now some people would use a RAID 5 drive and use the parity information to find out if their is an error, but in my world finding out there is a problem is useless I want a fix. I don't want a system that goes OK XOR DISK 1,2 != DISK 3 so BREAK. I want something that keeps parity information so it goes XOR != can I find a parity set that works... OK DISK 2 is bad but I can XOR DISK 1 and 3 and keep going...
In some systems people don't mind halting on errors but in the real time world making 1% more errors is worth a system that does not fail. After it's over you can get a human to look into what's wrong but right now you need as much uptime as possible. A CVS scanner that miss reads 1 item a week but does not go down for 10 years is worth a lot more than a scanner that goes down once a month for repairs.
However, I am more talking about systems where say a LAN card has a bad driver which keeps crashing the system. If the operator decides to reboot on such a failure that's one thing but I don't really want my OS designer to make those choices. It can be as simple as 3 modes, stable (Try and keep going), normal (Pop up a message but let non effected processes continue.), and reboot on error but BSOD is never a reasonable response IMO.
PS: implement high availability through multiple operating systems working in a distributed environment I am talking about 100's of CPU's on the same chip. They are all talking with the same HW and memory so it's not that easy to say ok system 44 is down re run those jobs on _.
The point is you can't trust software, drivers, or hardware. If you want to see a real OS look at what people use on million +$ systems. UNIX is great for the type of hardware most people have on there desktop but it's not what people will be using in 50 years. Intel could put out a CPU with 100's of processors but you start needing an OS that can work even if a few of those CPU's go bad.
If you want stability you need to asume things that can break will break.
I agree it's time to move on, but their are hard limits to how effecnt rockets can be. Space is never going to be cheep if we need to build anything for every launch. It's not going to be cheep if we need to take things apart after every trip.
You can build a ship that get's realy realy hot on reentry or one that opens a few para sails and never hit 1000C but we are never going to move forward if we keep trying to build a single system that we will try to use for the next 30 years. At this point we have little need to send people to orbit and they can spend 6 months up their every trip. So get the shuttle cost down to 100mill a trip and use that extra billion for R&D.
Why should you let fork add more process when their is not enough resorces to handel them? A real os should limit the amount of recorces any one family of processes can use. AKA A forks B1 - B20
B1 Forks C1(1) - C1(20)...
All of these are children of B1 so they all share the same resource pool which is less than the total resources for the system. B1 might be able to fork 50,000 process but that's the limit ALL of it's decendents. You can even set limits so say non OS functions that are spawned on boot are limited to 25% of the CPU.
Slashdot Mirror
NASA has been doing a lot of good basic research. Things like Ion drives, solar sails, and scram jets which are needed for long term development but don't get as in your face as manned space flight. I think it's best to keep them around as long as they are the only ones interested in doing this type of research.
But, the long-term goal of manned space flight is to make viable off world colonies. However, for that to work they need to be self-sustaining. So we need to develop methods of building everything from glass, to rubber, to CPU's on Mars if we want to truly get off this rock. Until that time it's always going to be cheaper to sit on earth and develop things like fusion than it is to maintain a manned mission in space.
I think there is a lot of value in the research done on how humans react to prolonged weightlessness, but for the near term we need to focus on developing things like biosphere II which will let us live among the stars as apposed to trying to get people to space cheaply or sending tunes of cash to get a few people to sight see on mars for a few months. It's cheaper to send a rocket into space with an experiment than it is to send a person and that experiment up at the same time.
NASA should try and get private companies interested in space but they are doing a great job with things like the mars probes. Even when these projects fail they still provide value if nothing more than keeping people interested in the stars.
Once companies get interested in researching space for profit then let them at it but as long as it's all subsidized by the US Gov then there is little value in having private companies doing basic research.
Fame goes to those who dance in the lights not the behind the seines characters. Lara Craft has same type of fame as Lennon, if not the same level, but the Carmack's of the world are going to be more like George Lucas as apposed to say Keanu Reeves.
The basic rules of real life is about as complex as this game of life. What this game simulates is closer to the chemical reactions of life as apposed to life. You get vary complex behavior not from the rule set but from the size of the board. The simple forms of "life" you see on tiny boards are not going to be anything like what you would see on say 10^1000 X 10^1000 board.
... chains are the building blocks of life. But real life's rule set is not built on the level of those huge and complex changes it's build on the atomic level with simple rules but a HUGE board.
The idea is it's not the specific rule set that changes to evolve life but the level of complexity allowed by the size of the board and a simple rule set. Oxygen, Hydrogen, Carbon,
People who want to simulate things don't even try and use the real worlds rule set because it's takes so much space so they design simpler simulations. But that does not change the validity of the simulation. As you can evolve a world from half a page of code if you can run a simulation of that size. Clearly the complexity is a result of evolution not the rule set. Hell the guiding hand is a RND function but you still get increases in complexity over time that's about as close to demonstrating evolution as you can get without simulating a planet for billions of years.
I'm not arguing against QM here, get that straight. vs. I'm saying QM is a new theory, sure it makes a lot of sense, but there may very well be a good reason why I won't instantly teleport across the galaxy.
Ok let's think about this in one case you think about the odds of you specifically popping say 1 foot to the left. Now I agree this is not going to happen in your lifetime. QM gives the type of odds, which would say it's unlikely for you to pop that foot given thousands of times the lifetime of the universe let alone 100 years.
But life is not a specific 'You". Hell if you showed up in the primordial goop you would have quickly died all that this is based on is one of the huge number of possible self replicating life forms showed up at some time over billions of years. On one of billions of worlds in all the worlds oceans. Seeing as that basic life form could be much simpler than you or me or any of the single celled life on this planet right now it is much more probable.
The separation is going from a specific event to a wide range of possible events that are much more likely than you popping even 1 inch to the left vs. teleporting across the galaxy.
PS: I was not trying to imply you had any specific faith. I could have used Buda vs. DC or any other figure but after arguing the same point enough it just started rolling off the fingers as it where. My point is separating the truly un-testable hypothesis from the above line of argument by pointing out not a specific case "JC" but a more general idea of anyone including DC (David C). I also like saying JC vs. DC as sort of a play on AC vs. DC.
On a deep level most people dislike the idea that random chance created the world. People like the idea that life has meaning and there was some creator involved in the mix. However, using that approach solves nothing as either that creator popped into existence, which would seem less likely than a much simpler life form or they evolved from simpler life thus moving evolution back one step. However, I can't really debate these ideas, as there is no way to demonstrate anything about something with godlike power.
ID is like a kid saying "well I know the magician does not know magic so it must be his invisible friend who knows the magic." in that it does not solve the problem merely abstracting it by one level.
I enjoy a good debate. I tend to be fairly proactive in my attack but don't take that to mean that I was angry with what or how you said something.
Anyway cutting though your last rebuttal:
I wasn't disputing the knowledge we have about quantum mechanics. All I was saying is that this cannot be used to prove evolution because it is non-falsifiable. Quantum mechanics becomes God in a sense. That's what I meant by it "sounds" religious.
There are several incorrect assumptions in that last statement. As the rest of the argument hangs on this statement "All I was saying is that this cannot be used to prove evolution because it is non-falsifiable." I would like to point out that in scientific debate you don't prove things you try and establish probabilities. I can't prove the world was created 999.999999 seconds ago, but that's not the type of proof science goes for. I say simple life can self assemble with some probability per cubic nanometer of water per second as specific reactions take place. Now each of those stages and assumptions can be tested. I then estimate the age and size of the earth and the number of earth like worlds and say something like "Given those probabilities there is a 95% chance that life should show up within 1 billion years on _ number of worlds. "
At this point you could attack ANY of my assumptions and get me to alter those probability you can't prove I am wrong just as you can't prove the world is older than 1000 seconds but you can demonstrate that it is unlikely. If you can show the odds where worse than say 1 in 20 then I would go ok while the theory was not disproved given those assumptions the theory is probably not true.
PS: Not that I really like semantic arguments but "it is amazing to see so many people absolutely certain that 'infinite' time, space, and energy can explain the origin of everything. Is anyone else uncomfortable about that?" is begging for it. And I am having trouble resisting...
"is amazing to see" that's nice but being surprised does not help your argument.
"'infinite' time, space, and energy can explain the origin of everything." I never said that we had infinite time. But shocking, as it might seem QM implies that with infinite time every possibility will happen. It's one of many things that bothers people but so does the double slit experiment but there is not that says the world has to work in a way that you feel comfortable about.
"I was generalizing." Reduxio ad insertom (SP?) it might win some arguments, but it does not work when there is anything else wrong with your argument.
Anyway, you can't win this type of argument with analogies so please stop trying to use time it's a waste of time.
"but common sense will tell you that that can't happen!"
... black lines show up. Now clearly something is not going on as you think it should so umm let's say god did it.
Wow, that sounds just like; "But common sense tells you that the world is flat!"
You and he both fell for the same trap. There is a huge difference between direct extrapolation from observed phenomenon than your gut feelings. We can all agree that that specific event is so unlikely it's not worth talking about but if you are going to talk about it you have to base your judgment on something. Now I don't know what the world was like 300 years ago let alone 3,000,000,000 years ago. But, I recognize I don't know what's going on back then so I don't say well this story sounds good so that's what I am going to think.
Let say you where to take your fingers and place them close together and in front of a bright light. Now based on your day-to-day world you would expect that nothing unusual is going to happen but
Now F != Mass * A but it works most of the time. Light does not really work the way you think it does but your ideas work most of the time. This suggests that our day-to-day understanding of the world does not extrapolate well to things outside of our understanding. But the fundamental difference between trusting scientific theory's and the good book is we understand the theory's to be educated guesses that we can test where the good book is a umm book. Now someone could suggest that good old JC was a hypnotist, but that's not really a testable theory. The thery that people like David Corresh can convince people they need to follow them is. Now that's a big hole in religious argument, but hey have faith you might be right just try not to make things up to "prove" your point.
PS: I am not going to bet that a chair is going to self assemble in my lifetime but I am willing to be CPU's will work most of the time which works under the same principle.
If it was not of "supernatural origin" then it showed up as part of nature. At which point you just added intelligence with no cause, or you are saying that it evolved on another world and then showed up to direct us Ect.
Saying "God did it" is the same as saying it just showed up because at some level God just showed up. It's not an explanation so much as pointing at a magician and saying see he can't do magic it's all done by his invisible friend who can do magic.
it sounds just as religious as any other religious point of view.
It's a direct extrapolation from observed phenomenon. Take a single electron in a "bottle" now it lacks the energy to jump the lid but given time there is a probability that it will jump that lid based on the size of the trap. The problem with your argument is you are saying well ok that works when it's at unit 1 or 10 or 100, but there it is never going to happen when it's unit 10 ^(100 ^ 1000) which not supported by the given evidence.
In the real world you would probably never see something like this happening, but in science you don't get to say this can't happen.
So tell me what do you think the odds of simple life showing up on the earth though random chance. If you say it's unlikely give me some logic given estimated probabilities over volume and time not just a gut feeling that that sounds like faith. If the idea gives you pause then feel free to investigate the subject, make a theory, test it, get it peer reviewed for methodology, and published and you can join the debate. But if your going with an idea because it sounds good then clearly you're the one going on faith.
We are going over a wide range of topics and I don't know which of them you agree with. So picking a single point of attack that you have stated:
.01% chance to flip. Ok running this you get a simple runner and then "guns" that shoot those runners then a slew of complex gunners. The bored is limited in size so it can only support life of a given complexity but with an undirected and simple rule set you give rise to complex "life".
Here I defend why.
The laws of physics don't change.
Take the game of life on a small board that fallows the basic rule set things get boring vary fast but take a bored that's 10,000 X 10,000 (12 MB of ram) set it up using a random arrangement so 1-25% of the bored is covered. Now to add a dash of mutation on each cycle to give each square a
This is completely undirected and the rules are extremely simple but if you run this simulation with different starting conditions you can see all the basic "valid" life forms show up. The larger the bored and the longer the simulation you run the more complex life you get but the complexity is all centered around the same basic laws of physics for that world.
Now change the starting conditions so the board starts blank and you just use the random mutation factor to add random cells. It takes longer but you still end up with life and given time you end up with all the same life forms.
All this life is limited by the laws of the world but given a large enough board you see predators, prey and even viruses. It's all undirected. The complexity is only limited by the size of the world. More mutation tends to evolve the life faster but over time you should start to see things that try and counteract this mutation. They will always fail some of the time but given time and enough board space it should evolve all the way to intelligent life.
PS: A tick has some level of intelligence as does, a bat, a rat, a dolphin and a human.
Cool I got a logical point across on /.
0 gtx/index.x?pg=1) that way they can test each component and cut off the pipelines that don't work. The top of the line chip could have say 10gig's of L2 cache and 500megs of L1 cache split among 100cpu's and with a good design they can still use chips that have significant damage in there Celeron brand which could be based around having 1/2 of those CPU's working with 1/2 the cache.
The 100CPU chip is based around comparing a P4 with a 8080 and trying to picture a chip that's that much more complex than the P4.
20 years is not that that long in terms of OS evolution. But, in that time frame we should see chips with GB or TB's of cache on them.
Now Intel can try and stick with a small number of CPU's on that chip but it's going to be a lot easer for them to build a chip like a geforce-7800 (http://techreport.com/reviews/2005q2/geforce-780
The basic problem is moving from 4 > 8 > 16 > 32 > 64 bit CPU's is useful but there is little value in a 128bit CPU vs. 2 X 64 bit CPUs. The 64bit CPU can address 2 ^ 32 X 4gigs of memory so it's going to be a while before we need 128bit CPU's but Single instruction Multiple data instructions are only so useful. Now we can always increase the on chip cash but it's a lot easer to have 100 64bit CPU's than it is to use that many transistors on 4 X 64 bit CPUs such that you are not just wasting them.
I don't know if we will ever have 10,000 CPU's on a chip but after a while it's hard to fund something useful to do with all those transistors. And it's going to be hard to build chips if we are still throwing away chips with a small number of defects. Hell, at some point I expect to start using pipelines to verify what other pipelines are doing.
But, I am not a chip designer so I am open to suggestions as to how you think this is going to play out over time.
So ID people accept that mutations are random?
And the earth is several billion years old.
And many species can be traced back to a common ancestor.
But, they feel the need for several ancestors on the root trees? WHY?
All plant cells are basically the same, All vertebrate cells are basically the same, but there is enough difference between plant and animal cells to say that they need to have a different ancestor?
Granted all tree cells are closer to each other than they are to say grass cells so do trees get a common ancestor or just plants? How about Apple trees vs Pines?
Evolution as a thery provides a clean link between all life on earth so which of these are ID people complaning about and what do they find annoying about these links?
laws of nature as we understand them to work cannot explain.
Such as? The human eye is extremely complex but there are a wide variety of eyes in nature all the way down to light sensitive cells on flatworms. But they are all useful and well adapted to the organism. AKA lion eyes are better for them than say frogeyes would be.
"I learned in my earlist years..." that's a simplification which is not realy true. In science everything is always a theory. Something like F = M * A or E = M * C ^2 is still a theory but they have been demonstrated to a level where trying to disprove them seems like a waste of time. Your always going to have a few oddballs that will try and disprove everything but the more established a theory the less time scientists are willing to spend on that specific debate. (But, I am bored at work and willing to waste some time on this topic.)
PS: metal furniture can self assemble see Quantum Mechanics. Now the odds of that happening are hard to comprehend but it's still possible and given sufficient time highly probable. Based on QM saying something can't happen is meaningless but some things are unlikely over a reasonable timeframe so if you want to try and demonstrate that something is extremely unlikely to ever occur on any planet anywhere in the solar system over billions of years that's one thing but saying can't is meaningless.
My previous post was not attacking you. I was pointing out why saying ID vs. Natural Selection was a poor expression of the idea. And if your going to defend Evolution it's best to try and get the ID people to point out a specific problem and why their theory is a better solution.
Anyway, T A saw the light and hopefully some of these people will as well. (Yes, that was meant as a joke.)
Natural Selection != Evolution.
Mutation != Evolution
Mutation + Natural Selection ~= Darwin's theory ~= Evolution.
Most ID people are aware of that things mutate and natural selection occurs but they like to call this microevolution and when pressed many of them are willing to admit it occurs. However, they like to try and come up with an arbitrary limit that new species can't evolve or simple amino acids can't evolve into simple cells Ect. But in the real world there is a probability that just about anything will happen from being killed by a comet to dieing because all the O2 in the room is not going into your lungs.
Now if they provided some model of the universe, which suggested some reason why evolution was consistently and significantly, to fast we that would be one thing but all I have seen is a lot of hand waving and no real explanation of how the theory of evolution breaks down.
PS: Even Tomas Aquinas recognized building on a foundation of religion is building a house on top of sand.
I have seen vary High Quality Software written by one person. It took a few years to get out but as he never needed to have "design" meetings or talk about how he was changing things so it was vary efferent but a little slow.
Smaller teams are always more efficient but under say 7 - 12 people they tend to take a longer time to get things done.
Some products like windows are never going to get done if one person works on them but smaller teams still tend to be cheaper per feature. So that's Cost VS. Speed issue.
You can skip over some testing and get a Speed and Cost Vs Quality issue.
You can even get several teams and have them work on the same project. Then pick the best one and get Cost vs. Speed and Quality.
Now if you combine the above you can exchange some of Speed, Quality, or Cost to get more of the others.
Ok your wrong. Hint: not all NASA launches cost anywhere near what the shuttle does.
He scaled up an in efficient rocket engine so no advantage there.
The idea of using a jet to help things into space had already been used.
He did not deal with any heat dissipation issues because he never got up to Mach 22 / never went into orbit.
They took the lightest pilot they could find and send him up to kiss the sky which is So comparable to sending a ship into orbit that has enough cargo capacity to carry this ship into space.
Now I am all for private companies spending their money to get into space but NASA spend 10 mil on the X prize and as far as I can see they got nothing out of it.
At launch the shuttle is producing 11Gw of thrust, which is an amassing amount of energy. But that's way more thrust than is really needed. Look how the numbers change when you build a craft which:
Uses a jet powered winged first stage to fly closer to the equator and launch at altitude and a little under mach 1. (Look into fuel costs to do this with a 747 vs. rocket fuel. Note: the wings save you from the worst aspects of gravity as the faster your going the lest rocket fuel you waste to keep from falling back trying to accelerate a craft and this is going on while the ship is it's most massive. And, the jet let's you avoid adding all that oxidizer and the structure to hold it, and you get to reduce the amount of low temp rocket fuel your using which saves you even more mass.)
Is only trying to get a 6 people +(7 days food, water, and air) and basically 0 cargo to LEO (Might be a good idea to give all or some of these people Eva capability and parachutes but this adds little to overall mass).
Now at this point the costs of wings on the 2nd / 3rd stage is a hell of a lot less in terms of total fuel but you can now design the ship to do a low temperature re entry on this light craft.
PS: I think building a huge high altitude / speed Para sail * would work much better than this but wings are not that bad of a basic idea if uses in a low temp design. They increase the service area to radiate heat and let you use glide time to lower the amount of incoming heat.
* The goal is to transfer orbital speed to the air and not the craft so a Para sail that glides though the upper atmosphere should be ideal. With the right design you basically added wings at a greatly reduced weight.
What's the practical benefit of wings on a spacecraft?
1) They can provide lift at the beginning of a flight.
2) They can also allow for low temperature reentry.
3) They let you control where and how you land.
4) As a result of 2 and 3 you can make a highly reusable craft.
Now the shuttle only really uses #3. By skipping #2 they have to do a lot of inspections, which waste the value of a reusable craft.
It's not that adding wings to any design is useful, but they are they only way to get to #4 which is the only way to make access to space cheep and safe.
Ok 1 team won the prize, but how many failed again? Lots of things look cheeper if you can safely ignore part of their costs.
Of course you do. The question is what do you do when things break? Do you trust the operating system to try to repair itself, or do you shut the OS down, let another OS take over, and call for someone to come take a look at things?
... OK DISK 2 is bad but I can XOR DISK 1 and 3 and keep going...
Right now I'd definitely say the latter. Maybe some day things will be different. Maybe some day we'll have strong AI. But even then, there's a limit to the amount of self-fixing that can be done.
It really depends on what your doing. I write real time software and in my problem domain the goal is to keep going if at all possible. Now some people would use a RAID 5 drive and use the parity information to find out if their is an error, but in my world finding out there is a problem is useless I want a fix. I don't want a system that goes OK XOR DISK 1,2 != DISK 3 so BREAK. I want something that keeps parity information so it goes XOR != can I find a parity set that works
In some systems people don't mind halting on errors but in the real time world making 1% more errors is worth a system that does not fail. After it's over you can get a human to look into what's wrong but right now you need as much uptime as possible. A CVS scanner that miss reads 1 item a week but does not go down for 10 years is worth a lot more than a scanner that goes down once a month for repairs.
However, I am more talking about systems where say a LAN card has a bad driver which keeps crashing the system. If the operator decides to reboot on such a failure that's one thing but I don't really want my OS designer to make those choices. It can be as simple as 3 modes, stable (Try and keep going), normal (Pop up a message but let non effected processes continue.), and reboot on error but BSOD is never a reasonable response IMO.
PS: implement high availability through multiple operating systems working in a distributed environment I am talking about 100's of CPU's on the same chip. They are all talking with the same HW and memory so it's not that easy to say ok system 44 is down re run those jobs on _.
Your forgeting about the drag from all the air between the ground and space.
The point is you can't trust software, drivers, or hardware. If you want to see a real OS look at what people use on million +$ systems. UNIX is great for the type of hardware most people have on there desktop but it's not what people will be using in 50 years. Intel could put out a CPU with 100's of processors but you start needing an OS that can work even if a few of those CPU's go bad.
If you want stability you need to asume things that can break will break.
What if the video card driver fails? SSH
I agree it's time to move on, but their are hard limits to how effecnt rockets can be. Space is never going to be cheep if we need to build anything for every launch. It's not going to be cheep if we need to take things apart after every trip.
You can build a ship that get's realy realy hot on reentry or one that opens a few para sails and never hit 1000C but we are never going to move forward if we keep trying to build a single system that we will try to use for the next 30 years. At this point we have little need to send people to orbit and they can spend 6 months up their every trip. So get the shuttle cost down to 100mill a trip and use that extra billion for R&D.
Why should you let fork add more process when their is not enough resorces to handel them? A real os should limit the amount of recorces any one family of processes can use.
AKA A forks B1 - B20
B1 Forks C1(1) - C1(20)...
All of these are children of B1 so they all share the same resource pool which is less than the total resources for the system. B1 might be able to fork 50,000 process but that's the limit ALL of it's decendents. You can even set limits so say non OS functions that are spawned on boot are limited to 25% of the CPU.