The RDA can't do physics

[Clarke]

Originally Posted by Human No More

You need if it you want your 'hurrrrrr remote control the robots' argument.

Oh, yes. Sorry. Derp.

Yes, in parallel - requiring a different multi-billion dollar system for each one. I thought you were the one complaining about cost, now you've switched track to fantasising about the film you wanted to see?

You need a few micrometers of electronics for each particle. That's going to cost, what, a fraction of a penny each? I'm talking about building more cells of a SSD, not building multiple disks. You don't need most of the actual device to add more particles to it, you just need to duplicate a very small chunk.

Without distances on an astronomical scale, the time change is less than a single clock cycle. You can not do anything on a single clock cycle other than the very basic instructions. If they do attempt to use it as such, then the data needs to return, with a 4.4 year light speed lag.

Gravitational time dilation. The time dilation factor between here and geosynchronous orbit is +664ps... per second. Put one end of your FTL communicator on a satellite for a year, and then take it back down to Earth, and the accumulated time difference will be as long as 20ms. In the context of high-frequency trading, that's a very long time indeed.

I said potentially. If it isn't, that just digs your hole even deeper.

In the 1950s, CPUs cost millions of dollars each. Look what happened.

Only if each bit had a defined exact time to transmit or each transmitter was only used for a single bit per packet.

Isn't that the default? That there is a window of a few fractions of a second where you transmit?

I know. Don't profess to greater knowledge here than me - each error bit reduces the chance of detection of others, and increases overhead - in usual applications, bandwidth is far less of a concern than noise itself, as the more complicated an ECC implementation, the more data overhead it creates.

Yet you said earlier that you would run into problems because you need to receive all the bits

That seems marginally more realistic an assumption, but it still doesn't show that other humans should not be there.

Each human costs you literally astronomical amounts of energy. That's why they shouldn't be there.

Depends on what's at stake. Determining the difference between lifeforms and making a judgement that could have such huge repercussions is very different from following defined traffic laws (and assuming that others will as well).

What do you mean, "assuming others will?" Nobody will, because they're human. A huge amount of the difficulty of this sort of thing is predicting how the world behaves and reacting to that before it's happened. That's the exact same "skills" that would be underpinning a military defense computer as well. And what do you mean, "huge repurcussions?" People die if your auto-car does something wrong, just as much as they die if your military computer does something wrong.

...Except the circumstances where a human will be in its range will be exceedingly rare indeed, because normal operation will have all the humans cosy inside Hell's Gate.

That may potentially work, but there has to be some reason it wasn't used - there could still be a reason it doesn't work - even if it's signal strength or light speed lag.

It worked fine for the avatars. Then again, radio is easier to get working.

Hang on, I thought we were talking about communication within the magnetic field here. Stop jumping form one topic to another for the same argument.

I am. IMO, the most practical way for the robots to communicate is with radio transmitters connected to them via optic fiber. That cancels out almost all your EM interference. The R&D comment was about what you do if you don't have these technologies completely lined up; you develop them, because they will save you tons of energy.

I'd contest the second, certainly, but that's beside the point. A rate at which things are developed does not mean any research will instantly complete, ever.

It means that gambling that the technology does not advance gets riskier and riskier as time goes on.

...unless they were looking for minerals available outside Earth's solar system, exactly as they were. They most likely already are utilising resources found elsewhere in Earth's system including asteroids and potentially comets, the next logical step is further. No resource in the solar system lasts forever even if humanity WAS planning to only ever live there.

The location is irrelevant to the odds of whether the gamble will pay off. You're gambling astronomical amounts of energy on the assumption that technology will not advance in the wrong way. If it does, the price of unobtanium plummets and you've lost everything. Accepting that gamble is not good business sense.

And you'll find that the resources in the solar system will quite easily last long enough for the Sun to expand and destroy the inner planets. They just need to be exploited properly, which the RDA don't have the technology to do circa 2154.

Useful? Possibly (yet you fail to give any examples) - but profitable?
Surprisingly, there isn't unlimited use for energy anyway.

Rent out O'Neil cylinders.

...and when it is reachable?

...It'll be reachable if and only if it's cheaper.

Yet it doesn't instantly make it possible. As far as we can see, the former has not happened yet, and while the latter MAY be possible, it's still a long way off if so.

If it's conceivable that ti could happen, it's not good business to send an ISV.

That level also varies by superconductor, with some available today requiring >100T to do so.

There's still an upper limit somewhere, probably limited by current more than B-field.

No, but if the amount of said infrastructure is growing, the savings are too.

If the amount of infrastructure is growing, then you're pumping money into it. Obviously you're savings increase if your investment increases.

For about the fifth time, option 2 is not an option at this point in time.

Then wait for it to be an option. You might have to invest trillions of dollars in research, but you will save 6 orders of magnitude for every penny you spend thereafter.

[Human No More]

[QUOTE=Clarke;154446]
You need a few micrometers of electronics for each particle. That's going to cost, what, a fraction of a penny each? I'm talking about building more cells of a SSD, not building multiple disks. You don't need most of the actual device to add more particles to it, you just need to duplicate a very small chunk.

I understand your logic there, but if a single one costs $7500/bit, then the cost is in the encoding process itself and not in anything associated with processing, or that would have already been done.

Gravitational time dilation. The time dilation factor between here and geosynchronous orbit is +664ps... per second. Put one end of your FTL communicator on a satellite for a year, and then take it back down to Earth, and the accumulated time difference will be as long as 20ms. In the context of high-frequency trading, that's a very long time indeed.

Now that is actually an interesting idea, but also completely irrelevant to ways people complain about Avatar.

In the 1950s, CPUs cost millions of dollars each. Look what happened.

I am going to assume you'd be pissed off if I said the same thing about energy.
The $7500/bit is form an official source, and your personal belief doesn't change that. Indeed, the cost per bit for FTL communication as it is is technically infinite today

Isn't that the default? That there is a window of a few fractions of a second where you transmit?

No, I meant that if you were fragmenting a packet across multiple simultaneous transmissions, there needs to be an indicator (e.g. 'this one is bytes 128-255').

Yet you said earlier that you would run into problems because you need to receive all the bits

Yes - do you even know what overhead is? It is the data that is necessary to include such as checksums, which is not a part of the actual data payload.
In order to determine if each bit is valid, there needs to be a value for every bit in the packet there. 0 is not the same as null. If one is missing, it will make other bits impossible to checksum. The greater the proportion of error bits, the chance is much higher that they can not be corrected and instead, the packet is simply flagged as corrupt and require retransmission. In addition, the more complex the scheme used, the greater the number of additional bits required for transmission - these bits can not be used to carry data.

What do you mean, "assuming others will?" Nobody will, because they're human.

Nobody follows traffic laws? I beg to differ.
Drive behind a driverless car at 90mph in a 30mph speed limit and you will cause a huge accident, becuase that is not within its parameters.

Seriously, WTF at the above quote.

And what do you mean, "huge repurcussions?" People die if your auto-car does something wrong, just as much as they die if your military computer does something wrong.

Someone potentially dying is not the same as losing the entire contract for unobtainium and likely facing huge sanctions and investigations.

It worked fine for the avatars. Then again, radio is easier to get working.

That was my point several posts ago. There is no reason to attempt to produce entirely synthetic brains to create a (two-way only) link (even if it would apply to such a signal rather than basic neural activity only) when radio does work, just not perfectly.

I am. IMO, the most practical way for the robots to communicate is with radio transmitters connected to them via optic fiber. That cancels out almost all your EM interference. The R&D comment was about what you do if you don't have these technologies completely lined up; you develop them, because they will save you tons of energy.

You seem to have this weird belief that R&D will give perfect results instantly and for free if it would be appropriate to perform - so why does Earth today not have fusion reactors?

It means that gambling that the technology does not advance gets riskier and riskier as time goes on.

For about the seventh time, that does not mean that things spontaneously appear. Such items required a convergence of numerous technologies which were in many cases decades in development, if not longer.

The location is irrelevant to the odds of whether the gamble will pay off. You're gambling astronomical amounts of energy on the assumption that technology will not advance in the wrong way. If it does, the price of unobtanium plummets and you've lost everything. Accepting that gamble is not good business sense.

Great way to ifnore what I actually said. Here's it rephrased in a form you should be able to understand:
1. Earth's system's resources will not last forever even with your nanotechnology fantasy
2. If humanity wants to keep growing, eventually it will run out of room
3. Based on 1 and 2, human expansion is inevitable if the conditions for 2 are met.
4. It does not matter what specific resources are there so much as that something useful is, and on a habitable world at that. A probe was even sent there on no other information than that it was potentially a habitable world.

Clear enough now?

And you'll find that the resources in the solar system will quite easily last long enough for the Sun to expand and destroy the inner planets. They just need to be exploited properly, which the RDA don't have the technology to do circa 2154.

...then they look for ones that are easier to do so with, which was the entire point.

Rent out O'Neil cylinders.

WTF does that have to do with 'better uses for energy' past getting to/from them?

...It'll be reachable if and only if it's cheaper.

But it is not currently reachable, which is the entire point.

If it's conceivable that ti could happen, it's not good business to send an ISV.

It's not good business to sit there without any (including the samples needed for your synthesis fantasy to be developed) waiting for 60+ years for it to become possible either, when all that time, an increasing amount could be in place and generating profit without subsequent fixed costs - so the earlier it is started, the higher profit is in the long term, and, as you conveniently convince yourself otherwise, even if synthesis did become possible, it does not suddenly destroy all ISVs, which should be used for exploration of other, further systems even. They are also a long term investment - you're thinking too short term again.

There's still an upper limit somewhere, probably limited by current more than B-field.

Yet based on current understanding, there's absolutely no indication of what it might be. 100T is over twice the strength of the highest sustained magnetic field that can currently be generated.

If the amount of infrastructure is growing, then you're pumping money into it. Obviously you're savings increase if your investment increases.

Exactly. Now you're stating the obvious, and actually backing up both points at once.

Then wait for it to be an option. You might have to invest trillions of dollars in research, but you will save 6 orders of magnitude for every penny you spend thereafter.

...and what happens if it proves impossible for long enough that those trillions cause a complete lack of funding in other areas?

[Clarke]

Originally Posted by Human No More

I understand your logic there, but if a single one costs $7500/bit, then the cost is in the encoding process itself and not in anything associated with processing, or that would have already been done.

But the encoding process was already established as being magnetic field messing around. That doesn't cost anything, once the hardware's there.

Now that is actually an interesting idea, but also completely irrelevant to ways people complain about Avatar.

It's still a small wrench in the works of the film's consistency. "Let's just throw this time machine under the rug, nobody'll notice it."

I am going to assume you'd be pissed off if I said the same thing about energy.

Energy can get 500 times cheaper and it won't make much difference to the economics involved.

The $7500/bit is form an official source, and your personal belief doesn't change that. Indeed, the cost per bit for FTL communication as it is is technically infinite today

There's no possible way for it to be $7500/bit, regardless of what the sources say. I can understand if it was some sort of phlebotinum that humans can't invent themselves, but it's a completely human-designed and -manufactured device, and we are good at mass-manufacturing of small things.

The FTL communicator isn't plausible as a real-world object; it's a plot device on a stick. It's like the Doctor's sonic screwdriver, except we're supposed to believe it works on something resembling real world science, not the "timey-wimey spacey-wacey" science Dr. Who presents.

No, I meant that if you were fragmenting a packet across multiple simultaneous transmissions, there needs to be an indicator (e.g. 'this one is bytes 128-255').

Then you pre-agree on what each transmitter means at specific times.

Yes - do you even know what overhead is? It is the data that is necessary to include such as checksums, which is not a part of the actual data payload.
In order to determine if each bit is valid, there needs to be a value for every bit in the packet there. 0 is not the same as null. If one is missing, it will make other bits impossible to checksum. The greater the proportion of error bits, the chance is much higher that they can not be corrected and instead, the packet is simply flagged as corrupt and require retransmission. In addition, the more complex the scheme used, the greater the number of additional bits required for transmission - these bits can not be used to carry data.

If I'm completely stupid in this context and send my message in triplicate, then once two copies of the message come in and agree, I don't have to wait for the third one; its data is irrelevant.

Nobody follows traffic laws? I beg to differ.

Not to the letter, rigidly, consistently.

Drive behind a driverless car at 90mph in a 30mph speed limit and you will cause a huge accident, becuase that is not within its parameters.

"Not within its parameters?" What do you think it is, a COBOL batch script? It's a predictive AI, it can deal with the unexpected. Possibly with more organization than a human, since it can't panic.

Someone potentially dying is not the same as losing the entire contract for unobtainium and likely facing huge sanctions and investigations.

And you have less trust in the computer than say, Wainfleet, ...why? You know exactly what the computer has been instructed to do, and you know it'll execute those instructions absolutely rigidly, unlike Wainfleet, who is human. Humans have these annoyingly unpredictable things called "unconscious impulses" and "emotions." Both of these often come to the fore when the human is under stress, and one major cause of stress is mortal danger...

Why are humans a good idea again?

That was my point several posts ago. There is no reason to attempt to produce entirely synthetic brains to create a (two-way only) link (even if it would apply to such a signal rather than basic neural activity only) when radio does work, just not perfectly.

Well then yes, use radio. You said earlier that radio doesn't work well enough, which is what the alternate solutions were for.

You seem to have this weird belief that R&D will give perfect results instantly and for free if it would be appropriate to perform - so why does Earth today not have fusion reactors?

Because fossil fuels and fission work well enough.

For about the seventh time, that does not mean that things spontaneously appear. Such items required a convergence of numerous technologies which were in many cases decades in development, if not longer.

...And? So what if it takes decades? It takes 20+ years to make any return on unobtanium. No business can accurately predict technology 20 years in the future, so no business would take the risk.

Great way to ifnore what I actually said. Here's it rephrased in a form you should be able to understand:
1. Earth's system's resources will not last forever even with your nanotechnology fantasy
2. If humanity wants to keep growing, eventually it will run out of room
3. Based on 1 and 2, human expansion is inevitable if the conditions for 2 are met.
4. It does not matter what specific resources are there so much as that something useful is, and on a habitable world at that. A probe was even sent there on no other information than that it was potentially a habitable world.

Clear enough now?

It completely matters what resources are there when you're shipping them back to Earth for profit. The RDA aren't in the philanthropic "expand humanity" business.

...then they look for ones that are easier to do so with, which was the entire point.

"Separate star system" is not easier to deal with, by (at a guess) 5 or 6 orders of magnitude.

WTF does that have to do with 'better uses for energy' past getting to/from them?

The energy you use to get one ship, one way, to Pandora could be used to build and maintain an O'Neil cylinder, as an example. Or mine the entirety of the asteroid belt for conventional minerals. 49kgc^2/kg (ignore the odd unit) is expensive.

But it is not currently reachable, which is the entire point.

And so nobody's tried to mine the asteroids, despite the multi-trillion dollars they're worth.

It's not good business to sit there without any (including the samples needed for your synthesis fantasy to be developed) waiting for 60+ years for it to become possible either, when all that time, an increasing amount could be in place and generating profit without subsequent fixed costs - so the earlier it is started, the higher profit is in the long term, and, as you conveniently convince yourself otherwise, even if synthesis did become possible, it does not suddenly destroy all ISVs, which should be used for exploration of other, further systems even. They are also a long term investment - you're thinking too short term again.

(Why would you need samples? You know what you're trying to build, right? )

As I said earlier, you can't possibly get unobtanium from Pandora for a profit, unless you have dictator-like control over vast portions of Earth's entire power output. You need ~100,000 1GW generators to fuel your tiny 50-ton ship in anything approaching a reasonable amount of time.

Yet based on current understanding, there's absolutely no indication of what it might be. 100T is over twice the strength of the highest sustained magnetic field that can currently be generated.

We've got 97T, we're trying for 100 right now. AFAIK, high magnetic fields behave like high temperatures, so if Unobtanium's critical temperature is only just above room temp, you won't be able to put a lot of power through it.

Exactly. Now you're stating the obvious, and actually backing up both points at once.

Your original statement was that your savings increase exponentially. That's only true if you continually reinvest your savings, which eventually falls apart. (because there is no more unobtanium to gather.)

...and what happens if it proves impossible for long enough that those trillions cause a complete lack of funding in other areas?

Then you don't get unobtanium.

[Human No More]

Originally Posted by Clarke

But the encoding process was already established as being magnetic field messing around. That doesn't cost anything, once the hardware's there.

So now you're questioning actual canon? It's $7500 - i's not stated exactly why, but for that reason, there's no way to show it does not without building a working one for less.

There's no possible way for it to be $7500/bit, regardless of what the sources say.

Early computing was more expensive - FTL communication doesn't even exist yet.

...it's a completely human-designed and -manufactured device, and we are good at mass-manufacturing of small things.

So by your logic, nothing at all should cost more than maybe a few hundred?
Wow.

The FTL communicator isn't plausible as a real-world object; it's a plot device on a stick. It's like the Doctor's sonic screwdriver, except we're supposed to believe it works on something resembling real world science, not the "timey-wimey spacey-wacey" science Dr. Who presents.

Admittedly, I don't watch doctor who at all so may be completely on the wrong track here, but if it works via sound, then isn't that a mechanism right there? - to expand that into a real point, if there is a mechanism by which something works provided, then doesn't that allow assumptions?
The FTL communications are completely unnecessary as far as plot goes - they were included most likely in an attempt to avoid complaints from people like you who wanted to see their specific favourite technology there. If it wasn't there, you'd be going 'BAAWWW NO FTL COMM'. Until fairly recently, you were pointing out possible mechanisms for one, even.

Then you pre-agree on what each transmitter means at specific times.

Then, as I mentioned many posts ago now, you're limited to parallel-only and can only send or receive at one time.

If I'm completely stupid in this context and send my message in triplicate, then once two copies of the message come in and agree, I don't have to wait for the third one; its data is irrelevant.

I thought we were discussing the bandwidth issue? Tripling the amount of data required will not speed up the latency at all, and will triple the bandwidth requirement. Sending it three times has noting at all to do with error correction, stop trying to Gish Gallop me or I will stop responding.

Not to the letter, rigidly, consistently.

...and that's part of its program. What would happen if you behaved in a wholly unexpected way around it and not just inconsistently within the margin it was developed for?

"Not within its parameters?" What do you think it is, a COBOL batch script? It's a predictive AI, it can deal with the unexpected. Possibly with more organization than a human, since it can't panic.

It can't deal with every single eventuality, simply because it is not AI. A human needs to establish what the potential procedures are.

And you have less trust in the computer than say, Wainfleet, ...why? You know exactly what the computer has been instructed to do, and you know it'll execute those instructions absolutely rigidly, unlike Wainfleet, who is human. Humans have these annoyingly unpredictable things called "unconscious impulses" and "emotions." Both of these often come to the fore when the human is under stress, and one major cause of stress is mortal danger...

Building an entirely new system that does not exist at all is a good idea... why?
Your original argument said there should be no human oversight at all, and now you're attacking a rebuttal of that point from an entirely different position.

Well then yes, use radio. You said earlier that radio doesn't work well enough, which is what the alternate solutions were for.[/quote]
I said not well enough for a 4.4ly distance with a 4.4 year latency. That's a big difference from communicating with someone near- adjacent.

Because fossil fuels and fission work well enough.

So you somehow, almost religiously, believe that if fusion was needed, it would spontaneously materialise in human knowledge?
Well, it's not only under research right now but anyone with any knowledge of energy knows that there is an impending world energy crisis as the required capacity is not available due to underinvestment in fission, but fusion has not yet magically and instantly succeeded as you believe it would.

...And? So what if it takes decades? It takes 20+ years to make any return on unobtanium. No business can accurately predict technology 20 years in the future, so no business would take the risk.

What I mean is that your pet ideas are themselves based on technologies that may well still only be under development in the future.
There are plenty of stories about such things, I do not understand why you bash Avatar for not having things you wanted to see.

It completely matters what resources are there when you're shipping them back to Earth for profit. The RDA aren't in the philanthropic "expand humanity" business.

No, but they would for profit when contracts came along. If the local resources in a properly habitable world were not profitable enough to be brought to Earth, they can still be utilised locally.

"Separate star system" is not easier to deal with, by (at a guess) 5 or 6 orders of magnitude.

Are you doing this intentionally? This particular thread of the argument was about the asteroid belt in Earth's solar system. What I said is that the asteroids that ARE usable will be - the smaller and less reachable ones will not unless demand outstrips supply by enough at some point in the future.

The energy you use to get one ship, one way, to Pandora could be used to build and maintain an O'Neil cylinder, as an example. Or mine the entirety of the asteroid belt for conventional minerals. 49kgc^2/kg (ignore the odd unit) is expensive.

...then what makes you think it hasn't been done?
PS. you can not build anything out of energy. It isn't a substance like the swords made 'out of energy' you might find in bad scifi.

And so nobody's tried to mine the asteroids, despite the multi-trillion dollars they're worth.

...and in 143 years?

(Why would you need samples? You know what you're trying to build, right? )

Not if you've never been there so don't even now what you're trying to produce, not to mention a detailed analysis of its composition and structure (even probes within the solar system are remarkably basic due to the hostile environment they need to survive years in, many even lack colour photography - a simple test of properties is likely all it managed)

As I said earlier, you can't possibly get unobtanium from Pandora for a profit, unless you have dictator-like control over vast portions of Earth's entire power output. You need ~100,000 1GW generators to fuel your tiny 50-ton ship in anything approaching a reasonable amount of time.

Yet again - there is no reason Earth's power output isn't higher. You're saying it's unrealistic with current rates, but that does not make it impossible. you also called it 'inconsistent' several times before I pointed out that is is anything but as there is never ANY figure given for Earth's energy production. Later, you then diverged onto building space megastructures which would be capable of producing exactly the same power you concurrnetyluy bash as impossible ("why don't they build an O'neil Cylinder?" and "they can't build a Dyson shell" do not resemble portions of the same argument).

We've got 97T, we're trying for 100 right now.

Emphasis on stable. I was just providing it as an aside to illustrate the actual strength of 100T - it is not something reachable so easily. In retropspect, a better one that wouldn't have let you off would have been stating that Earth's field is 3x10^-5T.

AFAIK, high magnetic fields behave like high temperatures, so if Unobtanium's critical temperature is only just above room temp, you won't be able to put a lot of power through it.

1. That's a huge assumption to make.
2. They behave like it in limiting superconductivity, yes, but 100T is NOT at all equal to even a low temperature.

Your original statement was that your savings increase exponentially. That's only true if you continually reinvest your savings, which eventually falls apart. (because there is no more unobtanium to gather.)

Yes, but that's a long time in the future. Unobtainium-related profit may well be funding your research fantasy.

[Clarke]

Originally Posted by Human No More

So now you're questioning actual canon? It's $7500 [etc]

I'm fairly sure I can get a hardware quantum RNG for less than that, which is pretty much the same device. The only difference is the electromagnet for the inducing, and those appear by the million in HDDs and the like.

Early computing was more expensive - FTL communication doesn't even exist yet.

It was more expensive due to a combination of scarcity, and demand. I bet it wasn't that expensive if you somehow had a monopoly on one.

So by your logic, nothing at all should cost more than maybe a few hundred?
Wow.

Once the fabrication process is constructed, which can potentially take millions, then no. If you're smart about it, you might spend $2bn on a new fabricaton facility... which can then spit FTL radios out at 1,000 units a day.

Admittedly, I don't watch doctor who at all so may be completely on the wrong track here, but if it works via sound, then isn't that a mechanism right there? - to expand that into a real point, if there is a mechanism by which something works provided, then doesn't that allow assumptions?

It repairs barbed wire. It makes (non-explosive) things explode. It can switch on car alarms at a distance. It's not sound; It's Time Lord Science™. As I said, we're not supposed to expect it to be founded on any sort of physically correct physics, unlike Avatar. There, we are expected to believe that everything is at least scientifically plausible.

The FTL communications are completely unnecessary as far as plot goes - they were included most likely in an attempt to avoid complaints from people like you who wanted to see their specific favourite technology there. If it wasn't there, you'd be going 'BAAWWW NO FTL COMM'. Until fairly recently, you were pointing out possible mechanisms for one, even.

...Was I? Current physics doesn't recognize any method of FTL comms. I was only extrapolating from what the sources said about how Cameron-FTL™ works.

Also, if he'd taken it and the magnetic explanation for the flying mountains out, he'd have scored points for using real (extrapolated) physics. Writing hard sci-fi is all about inventing as few rules as possible, and sticking by them once you've invented them. You can make two rules go poof without affecting the plot? Yay!

(Though it does affect the plot, however slightly: without it, Quaritch couldn't offer Jake's legs back.)

Then, as I mentioned many posts ago now, you're limited to parallel-only and can only send or receive at one time.

Depending on the amount of pairs you have and the packet size. Also, using more than one clock cycle to transmit one packet is a possibility.

I thought we were discussing the bandwidth issue? Tripling the amount of data required will not speed up the latency at all, and will triple the bandwidth requirement. Sending it three times has noting at all to do with error correction, stop trying to Gish Gallop me or I will stop responding.

We were; I was demonstrating that there are some ECC schemes that will let you ignore some incoming bits and not have to wait for them. (Though triplicate messages do affect ECC in that you can basically ignore it.) Also, I know it'd be stupid to send a triplicate message when we're short on bandwidth, hence my comment about being "completely stupid in this context."

...and that's part of its program. What would happen if you behaved in a wholly unexpected way around it and not just inconsistently within the margin it was developed for?

It will do its best to predict you. Of course it won't be able to do that perfectly; only posthuman AI could do that.

It can't deal with every single eventuality, simply because it is not AI. A human needs to establish what the potential procedures are.

Have you looked at AI research in the last 20 years or so? We have machines that can learn now. Re: "every single eventuality," neither can a human. A human usually does a lot worse.

Building an entirely new system that does not exist at all is a good idea... why?

Cheaper than sending humans. You can build the majority of Skynet your defense system when you get there.

Your original argument said there should be no human oversight at all, and now you're attacking a rebuttal of that point from an entirely different position.

Ideally, there should be no on-site humans at all, but since you don't like the idea of high-bandwidth FTL, I'm settling for a few individuals.

I said not well enough for a 4.4ly distance with a 4.4 year latency. That's a big difference from communicating with someone near- adjacent.

Why are we talking about radio between there and Earth? (FTL, FTW.) You mentioned earlier that remote control robots don't work because of Pandora's magnetic field, and I presented how to engineer around that.

So you somehow, almost religiously, believe that if fusion was needed, it would spontaneously materialise in human knowledge?

If fusion were needed, you'd have every energy production company in the world investing billions each in R&D, so you'll certainly see progress. You'll get something similar to the computing boom of 1990-2005.

Well, it's not only under research right now but anyone with any knowledge of energy knows that there is an impending world energy crisis as the required capacity is not available due to underinvestment in fission, but fusion has not yet magically and instantly succeeded as you believe it would.

Humans are short-sighted? Also, startup cost for fusion is tremendous. Also also, terawatts of power output is hardly a "crisis." What you might get is a stasis, which would be bad, but it'd hardly be the end of the world. It'd just mean power would get more expensive.

There are plenty of stories about such things, I do not understand why you bash Avatar for not having things you wanted to see.

I don't understand why you think I'm bashing Avatar for not having tech I think are "cool." If I wanted posthuman supertech I'd be reading Disapora or something. I'm bashing Avatar for not playing by its own rules, specifically in the areas of spaceship fuel and the completely stupid idea of interstellar mining.

No, but they would for profit when contracts came along.

...Contracts for what? (Also, good luck utilising Unobtanium without industry.)

What I said is that the asteroids that ARE usable will be - the smaller and less reachable ones will not unless demand outstrips supply by enough at some point in the future.

Yes, but exploiting even the most useless of asteroids requires maybe 100,000 times less energy than shipping stuff to Pandora. The delta-v required to completely escape the solar system is something like 30km/s. Getting to Pandora requires 1,000x that delta-v, and the rocket equation means that you need exponentially more fuel for a given increase in delta-v.

...then what makes you think it hasn't been done?

A dead Earth isn't much of a concern if you can build O'Neil cylinders.

PS. you can not build anything out of energy.

'Course you can, just put it into some baryons and mash them together.
Though I was more talking about the energy budget for gathering and constructing the existing materials for an O'Neil cylinder, since the former is just inefficient.

...and in 143 years?

Still won't happen, probably, unless we get space elevators or something.

Not if you've never been there so don't even now what you're trying to produce, not to mention a detailed analysis of its composition and structure...

Whereas the probes we're sending into the solar sytem aren't looking for anything new, they're looking for geological/biological evidence. I'm sure if there was a reason to include colour photography, NASA would stick one on. Also, remember Moore's Law.

Yet again - there is no reason Earth's power output isn't higher. You're saying it's unrealistic with current rates, but that does not make it impossible. [etc]

I'm saying it's unrealistic with completely ridiculous future rates. The 890GJ/year figure I gave above translates to 22.2kW, every second of every day. I don't use that much, and I wouldn't have any idea what to do with it if I did. Do you?

And again, megastructures are unlikely because 1) the manual doesn't mention them, 2) the manual(s) mention fusion power. Fusion power is nigh-irrelevant in the shadow of a megastructure.

Emphasis on stable. I was just providing it as an aside to illustrate the actual strength of 100T - it is not something reachable so easily. In retropspect, a better one that wouldn't have let you off would have been stating that Earth's field is 3x10^-5T.

That was "stable." (Though they degrade quite quickly for obvious reasons.) I think the maximum unstable one was in the range of 700T.

...And yeah, Earth's magnetic field is 30uT. How high is Pandora's supposed to be? Even 1mT, standing, would have negligible effect on any sort of electronics or sensor, AFAIK.

1. That's a huge assumption to make.
2. They behave like it in limiting superconductivity, yes, but 100T is NOT at all equal to even a low temperature.

That's what I meant. A magnetic field is going to have similar effects to high temperature in a superconductor.

Yes, but that's a long time in the future. Unobtainium-related profit may well be funding your research fantasy.

That'd be... amazing, considering the energy content involved.

[Human No More]

Originally Posted by Clarke

I'm fairly sure I can get a hardware quantum RNG for less than that, which is pretty much the same device. The only difference is the electromagnet for the inducing, and those appear by the million in HDDs and the like.
...
If you're smart about it, you might spend $2bn on a new fabricaton facility... which can then spit FTL radios out at 1,000 units a day.

Even if it doesn't seem plausible - it's stated as $7500, and that seems unlikely to be retconned. After all, it isn't anything that humans are anywhere near doing today.
In this case, it doesn't matter how it works so much as that it does. e.g. fair enough on the sonic screwdrive then (Well, technically, setting off a car alarm with ultrasound (would interfere with many motion detectors)/infrasound(resonance) is not impossible, but fair enough on the rest of those )

It repairs barbed wire. It makes (non-explosive) things explode. It can switch on car alarms at a distance. It's not sound; It's Time Lord Science™. As I said, we're not supposed to expect it to be founded on any sort of physically correct physics, unlike Avatar. There, we are expected to believe that everything is at least scientifically plausible.

Well, as I said, I didn't make any assumption of its actual capabilities. In that case then perhaps it doesn't refer to a mechanism of action, but is just a name. This is the central point of the argument, I think. I maintain that being physically possible and plausible are different to resolving to current day energy production - indeed, the FTL communication is a far less plausible entity than slower than light travel to nearby stars, or even the avatars. So while yes, it takes a lot of energy, it isn't 'impossible'.

...if he'd taken it and the magnetic explanation for the flying mountains out, he'd have scored points for using real (extrapolated) physics. Writing hard sci-fi is all about inventing as few rules as possible, and sticking by them once you've invented them. You can make two rules go poof without affecting the plot? Yay!

It's possibly true, but am I the only person who can see how that would make people ask how they worked? I really hope not. As it is, they aren't even in the main film at all (or even in any version of the film, for the former), but there for people with the interest.

(Though it does affect the plot, however slightly: without it, Quaritch couldn't offer Jake's legs back.)

I don't see how not, honestly. That's medical technology, which could be anything from a complete replacement spine via stem cells or even an artificial one, to simply reconnecting and regenerating damaged nerve tissue.

We were; I was demonstrating that there are some ECC schemes that will let you ignore some incoming bits and not have to wait for them. (Though triplicate messages do affect ECC in that you can basically ignore it.) Also, I know it'd be stupid to send a triplicate message when we're short on bandwidth, hence my comment about being "completely stupid in this context."

Fair enough, then. My initial point was actually about latency rather than bandwidth anyway - if 3 bits/hour is assumed to be an average and the success rate is known as 0.1% for actually encoding the bit.

Have you looked at AI research in the last 20 years or so? We have machines that can learn now. Re: "every single eventuality," neither can a human. A human usually does a lot worse.

Only when the expert system/AI is programmed with the knowledge of multiple humans. Producing a completely autonomous system for an unknown environment in hazardous conditions is not so simple as controlled conditions on Earth.

Why are we talking about radio between there and Earth? (FTL, FTW.) You mentioned earlier that remote control robots don't work because of Pandora's magnetic field, and I presented how to engineer around that.

My point was about remote control from Earth (where signals will be far weaker), and was actually an aside from the 8.8 year latency of such a setup.

If fusion were needed, you'd have every energy production company in the world investing billions each in R&D, so you'll certainly see progress. You'll get something similar to the computing boom of 1990-2005.

...but they are. It's far from being at all energy efficient, but it's getting there. The reason is that most 'energy companies' deal in one type, such as oil and gas, coal, solar or nuclear, or else in grid distribution where they buy the actual energy from others, and therefore have vested interests against other types, while fusion itself attracts funding on government levels.

Humans are short-sighted? Also, startup cost for fusion is tremendous. Also also, terawatts of power output is hardly a "crisis." What you might get is a stasis, which would be bad, but it'd hardly be the end of the world. It'd just mean power would get more expensive.

Note the word 'impending' - much of that multi-terawatt capacity is fossil fuel, and nuclear in most countries has a significant gap due to underinvestment that will require either significant life-extension of old reactors or the construction and first criticality of new ones at completely unprecedented speeds.

I don't understand why you think I'm bashing Avatar for not having tech I think are "cool." If I wanted posthuman supertech I'd be reading Disapora or something. I'm bashing Avatar for not playing by its own rules, specifically in the areas of spaceship fuel and the completely stupid idea of interstellar mining.

You might say space exploration itself is 'completely stupid' due to the costs compared to probes and simply sending packages of DNA towards habitable worlds in unmanned probes. That doesn't stop stories from making it their premise, or indeed, humanity's efforts towards it in real life. Nothing in Avatar's 'rules' contradicts anything else - everything is physically possible. It's a fictionalised earth with higher availability of energy, but still physically possible.

...Contracts for what? (Also, good luck utilising Unobtanium without industry.)

Are you even reading my previous message? You said they wouldn't be involved in space exploration/colonisation for free,and I pointed out that they need to get there somehow, and a company with a fleet of spaceships and possibly even small presences on habitable worlds would be in a good position there - even in real life, since NASA got ruined, they are going to have to rely on private businesses too now - life imitates art?

A dead Earth isn't much of a concern if you can build O'Neil cylinders.

You need something to build them out of, energy for life support, food, water (which weighs a LOT).

Still won't happen, probably, unless we get space elevators or something.

Perfectly plausible, at least locally to Earth.

Whereas the probes we're sending into the solar sytem aren't looking for anything new, they're looking for geological/biological evidence. I'm sure if there was a reason to include colour photography, NASA would stick one on. Also, remember Moore's Law.

The point was that it would be looking for something new, but discovering it on a remote moon would not instantly be enough to synthesise it.

...I don't use that much, and I wouldn't have any idea what to do with it if I did. Do you?

I know what I'd go: Go into space. If it's there, use it, because eventually you will run out of uses for energy.

And again, megastructures are unlikely because 1) the manual doesn't mention them, 2) the manual(s) mention fusion power. Fusion power is nigh-irrelevant in the shadow of a megastructure.

So it can be assumed they don't exist. That doesn't disprove any actual energy figure - just establishes a level that it is unlikely they have progressed beyond.

That's what I meant. A magnetic field is going to have similar effects to high temperature in a superconductor.

Now I don't follow. Yes, it should, but superconductors have different tolerances to them, just like they do to temperature.

[Clarke]

Originally Posted by Human No More

Even if it doesn't seem plausible - it's stated as $7500, and that seems unlikely to be retconned. After all, it isn't anything that humans are anywhere near doing today.
In this case, it doesn't matter how it works so much as that it does. e.g. fair enough on the sonic screwdrive then (Well, technically, setting off a car alarm with ultrasound (would interfere with many motion detectors)/infrasound(resonance) is not impossible, but fair enough on the rest of those )

It still looks like Cameron trying to have his realism cake (i.e. no FTL comms) and eat it too. (i.e. we need FTL comms to make the plot make sense)

I maintain that being physically possible and plausible are different to resolving to current day energy production - indeed, the FTL communication is a far less plausible entity than slower than light travel to nearby stars, or even the avatars. So while yes, it takes a lot of energy, it isn't 'impossible'.

Of course plausibility and possibility are disconnected. However, it seems disingenuous to try and be plausible while being entirely aware you're not actually succeeding; you can't have your realism cake and eat it too. Either you have FTL, in which case you might as well have arbitrary bandwidth because you have no real-physics justification, or you don't have it at all.

It's possibly true, but am I the only person who can see how that would make people ask how they worked? I really hope not. As it is, they aren't even in the main film at all (or even in any version of the film, for the former), but there for people with the interest.

Nobody asks how the Avatars work... except when they keep working in areas where regular electronics fails. We accept that they're some part of 22nd century supertechnology. We don't need an explanation beyond that. But that does mean our belief stretches when other 22nd century supertechnology fails and the Avatars continue for unexplained reasons.

I don't see how not, honestly. That's medical technology, which could be anything from a complete replacement spine via stem cells or even an artificial one, to simply reconnecting and regenerating damaged nerve tissue.

Specifically, they let him offer Jake's legs back early. Otherwise, the response would take 8.8 years, and Jake would be long gone by then.

Fair enough, then. My initial point was actually about latency rather than bandwidth anyway - if 3 bits/hour is assumed to be an average and the success rate is known as 0.1% for actually encoding the bit.

You can get around the latency problem by multiplying your bandwidth, sending in parallel, and shifting the timesteps of the chunks. How much that gains you obviously depends on exactly how much bandwidth you have to play with.

Only when the expert system/AI is programmed with the knowledge of multiple humans. Producing a completely autonomous system for an unknown environment in hazardous conditions is not so simple as controlled conditions on Earth.

I would have thought your probe would have enough sensory hoohah on it to tell you that you're faced with a jungle of some sort.

My point was about remote control from Earth (where signals will be far weaker), and was actually an aside from the 8.8 year latency of such a setup.

Well yeah, remote controlling them from Earth is impractical unless/until you multiply your FTL bandwidth. You'd remote control them from inside Hell's Gate, or (better) give them rudimentary AI and only have humans involved when something goes wrong.

...but they are. It's far from being at all energy efficient, but it's getting there. The reason is that most 'energy companies' deal in one type, such as oil and gas, coal, solar or nuclear, or else in grid distribution where they buy the actual energy from others, and therefore have vested interests against other types, while fusion itself attracts funding on government levels.

I mean billions invested specifically in fusion research. The only plans I know of are ITER and DEMO, and they don't seem to be very high priority.

Note the word 'impending' - much of that multi-terawatt capacity is fossil fuel, and nuclear in most countries has a significant gap due to underinvestment that will require either significant life-extension of old reactors or the construction and first criticality of new ones at completely unprecedented speeds.

Fossils fuels don't go "poof" suddenly. They'll peter out, and when they do, the price goes up, which means the economics of using them go down, which means they get used less, which means the price doesn't go up as fast...

Which takes a few decades at least, so we can get nuclear going in that time.

You might say space exploration itself is 'completely stupid' due to the costs compared to probes and simply sending packages of DNA towards habitable worlds in unmanned probes. That doesn't stop stories from making it their premise, or indeed, humanity's efforts towards it in real life. Nothing in Avatar's 'rules' contradicts anything else - everything is physically possible. It's a fictionalised earth with higher availability of energy, but still physically possible.

Well, manned interstellar space exploration is mostly completely stupid under real physics.

Also, I don't mind when a story invents its own rules to tell a story, such as in the Who example where you get "science" like, "This is my timey-wimey detector. It goes 'ding' when there's stuff." However, if you want to tell a story about a interstellar mining operation, you've got to pick rules that permit that plot, and RL physics isn't one of them. He could have fixed that easily enough with a FTL drive the RDA have a monpoly on or something.

Are you even reading my previous message? You said they wouldn't be involved in space exploration/colonisation for free,and I pointed out that they need to get there somehow, and a company with a fleet of spaceships and possibly even small presences on habitable worlds would be in a good position there - even in real life, since NASA got ruined, they are going to have to rely on private businesses too now - life imitates art?

I don't think anyone is doing space research for the profit margins. It was my understanding it was for PR.

You need something to build them out of, energy for life support, food, water (which weighs a LOT).

Near-Earth objects. Or even main-belt objects, that just takes longer.

Perfectly plausible, at least locally to Earth.

Yeah, a space elevator would make it cheaper to get to space, which means it's more economical to mine stuff out of the local solar system.

The point was that it would be looking for something new, but discovering it on a remote moon would not instantly be enough to synthesise it.

Depends how much you learn from the moon. If you learn enough to only leave one possibility for its structure/properties/etc, then you can model it, despite never having seen a sample. Once you can model it, you know what you're building.

I know what I'd go: Go into space. If it's there, use it, because eventually you will run out of uses for energy.

This is presumably after the solar system's been taken to pieces and used as computronium?

So it can be assumed they don't exist. That doesn't disprove any actual energy figure - just establishes a level that it is unlikely they have progressed beyond.

Yes: roughly 20% or so of the power you need for (small!) antimatter-rockets in a reasonable timescale. That's the issue.

Now I don't follow. Yes, it should, but superconductors have different tolerances to them, just like they do to temperature.

Yes, and we have no figure on unobtanium. However, the original statement was "unlimited power" which isn't feasible.

[Moco Loco]

Originally Posted by Clarke

we need FTL comms to make the plot make sense

Why? It would be more convenient for sure, but I don't think it's necessary.

[Clarke]

Originally Posted by Moco Loco

Why? It would be more convenient for sure, but I don't think it's necessary.

Well, he included it for a reason.

[Moco Loco]

Sure, but it didn't exactly have a prominent role in the movie. It makes sense, but I doubt many people would've noticed if it weren't there Glad it is though.

[Clarke]

Why? I mean, it doesn't have much effect.

(And oops, I am actually done with the last megapost. Silly me for leaving the note.)

[Moco Loco]

Why what, why am I glad it's there? Because it's fun, realistic, and would make things a lot more convenient for everyone all around.

[Human No More]

Originally Posted by Clarke

It still looks like Cameron trying to have his realism cake (i.e. no FTL comms) and eat it too. (i.e. we need FTL comms to make the plot make sense)

Actually, we don't. There's no problem with 4.4 year latency for high priority data and nearly 6 years for low priority.

Either you have FTL, in which case you might as well have arbitrary bandwidth because you have no real-physics justification, or you don't have it at all.

Adding arbitrary bandwidth just opens an attack surface for accusations of being unrealistic.

Nobody asks how the Avatars work... except when they keep working in areas where regular electronics fails. We accept that they're some part of 22nd century supertechnology. We don't need an explanation beyond that. But that does mean our belief stretches when other 22nd century supertechnology fails and the Avatars continue for unexplained reasons.

Technically, apart from the FTL communication, they are the only 'supertechnology'.

Specifically, they let him offer Jake's legs back early. Otherwise, the response would take 8.8 years, and Jake would be long gone by then.

It's far more likely they were going to put him back onto the shuttle and into cryo, sending authorisation with him to have it done back on Earth, rather than on Pandora.

I would have thought your probe would have enough sensory hoohah on it to tell you that you're faced with a jungle of some sort.

Yes, but not necessarily that there are 'angtsik running around

Well yeah, remote controlling them from Earth is impractical unless/until you multiply your FTL bandwidth. You'd remote control them from inside Hell's Gate, or (better) give them rudimentary AI and only have humans involved when something goes wrong.

It's still overcomplicating it compared to using a human if the capacity to bring them there is available (and still requiring more intelligent and better paid humans to maintain the equipment if not).

I mean billions invested specifically in fusion research. The only plans I know of are ITER and DEMO, and they don't seem to be very high priority.

NIF, JET, LLE, HiPER is another potential one.

Fossils fuels don't go "poof" suddenly. They'll peter out, and when they do, the price goes up, which means the economics of using them go down, which means they get used less, which means the price doesn't go up as fast...

That doesn't make them unlimited. Indeed, when remaining deposits become too marginal to use is exactly the same as exhaustion in effective terms.

Which takes a few decades at least, so we can get nuclear going in that time.

The issue is that the majority of existing reactors are near the end of their lifespan and require upgrades or replacement within the next 5-10.

Well, manned interstellar space exploration is mostly completely stupid under real physics.

Again - yes, humanity is working on it because they have that mentality.

Also, I don't mind when a story invents its own rules to tell a story, such as in the Who example where you get "science" like, "This is my timey-wimey detector. It goes 'ding' when there's stuff." However, if you want to tell a story about a interstellar mining operation, you've got to pick rules that permit that plot, and RL physics isn't one of them. He could have fixed that easily enough with a FTL drive the RDA have a monpoly on or something.

That would create all sorts of issues if travel because so casually easy - it would completely change the scope of operations there.

I don't think anyone is doing space research for the profit margins. It was my understanding it was for PR.

Tell that to companies entering things like the X Prize.

Depends how much you learn from the moon. If you learn enough to only leave one possibility for its structure/properties/etc, then you can model it, despite never having seen a sample. Once you can model it, you know what you're building.

It would be possible, particularly for simple covalent bonds between a few elements, but the structure increases the difficulty - probes are very specialised in their function. It may well be possible, but the simple fact that it didn't happen implies that fell down somewhere along the route.

Yes, and we have no figure on unobtanium. However, the original statement was "unlimited power" which isn't feasible.

No, but the limit could still be high enough as to be near-arbitrary. It's a fictional compound, it can have parameters of whatever it is required to be.

[Clarke]

Originally Posted by Human No More

Actually, we don't. There's no problem with 4.4 year latency for high priority data and nearly 6 years for low priority.

Well, I'd have thought you'd add them for the plot, because I can't think of another reason he's put it in.

Adding arbitrary bandwidth just opens an attack surface for accusations of being unrealistic.

Having FTL at all is a attack surface for unrealism. "FTL + unmentioned though large bandwidth" is no more unrealistic than just plain "FTL."

Technically, apart from the FTL communication, they are the only 'supertechnology'.

Well, and the implied Dyson infrastructure, the Venture Star, and whatever medicine is used to regenerate Jake's spine.

It's far more likely they were going to put him back onto the shuttle and into cryo, sending authorisation with him to have it done back on Earth, rather than on Pandora.

Shuttles only arrive every 6 months, at maximum.

Yes, but not necessarily that there are 'angtsik running around

For sufficiently large firepower, what's the problem? (Also, in this case, energy weapons might work better than kinetic rounds. Limitless, massless ammo simplifies a lot of the mechanics.)

It's still overcomplicating it compared to using a human if the capacity to bring them there is available (and still requiring more intelligent and better paid humans to maintain the equipment if not).

The complication is worth the 80TW/person, IMO. Knowledge is massless.

NIF, JET, LLE, HiPER is another potential one.

True enough, though I would suspect their budgets are vastly smaller than research into oil/gas/coal.

That doesn't make them unlimited. Indeed, when remaining deposits become too marginal to use is exactly the same as exhaustion in effective terms.

Well, yes, but that takes a long time, because it doesn't all happen at once.

The issue is that the majority of existing reactors are near the end of their lifespan and require upgrades or replacement within the next 5-10.

Well, someone will find a way to maintain most of the power output. (probably by keeping the reactors running past their lifetime date)

Again - yes, humanity is working on it because they have that mentality.

There are no serious proposals to get to Mars, (perhaps experiments along the lines of, "Are we sure this is feasible?" but nothing like a planned mission) let alone out of the solar system.

That would create all sorts of issues if travel because so casually easy - it would completely change the scope of operations there.

Then make it take 3 years each way and use unobtanium as fuel. FTL does not necessarily need to equal cheap, if the writer commands it so.

Tell that to companies entering things like the X Prize.

OK, profit that isn't artificially generated for PR/tech growth purposes.

It would be possible, particularly for simple covalent bonds between a few elements, but the structure increases the difficulty - probes are very specialised in their function. It may well be possible, but the simple fact that it didn't happen implies that fell down somewhere along the route.

You seem to be suggesting that Cameron has considered all this and knows it's unworkable, rather than considering that he could have overlooked something that spoils his plot/world.

As a more blatant example, nobody asks "What the hell are we doing?" with respect to giving Jake an Avatar. Or to be more specific, giving Jake an Avatar, then expecting him to give it up again. That's basic Hierarchy of Needs stuff.

No, but the limit could still be high enough as to be near-arbitrary. It's a fictional compound, it can have parameters of whatever it is required to be.

So long as they're physically feasible, assuming you want to be maintain "realism." A melting point of 100,000K, or a magnetic field strong enough to support 100 tons at 100m is just dancing around suspension of disbelief. At that point, you might as well call it... unobtanium phlebotinum.

[Human No More]

Originally Posted by Clarke

Well, I'd have thought you'd add them for the plot, because I can't think of another reason he's put it in.

Having FTL at all is a attack surface for unrealism. "FTL + unmentioned though large bandwidth" is no more unrealistic than just plain "FTL."

Yes, but still an order of magnitude less realistic.

...the Venture Star, and whatever medicine is used to regenerate Jake's spine.

Energy is intentionally vague. Everything in the Venture Star is possible, and as for Jake's spine, that's possibly the most ridiculous accusation yet.
Child Receives Trachea Organ Transplant Created With Own Stem Cells | Singularity Hub - we are now in the future. From there to a spine in 143 years seems reasonable.

Shuttles only arrive every 6 months, at maximum.

They can be sent up and down as needed. you're thinking of ISVs, which arrive approximately annually. It would have worked like this.
-Quaritch gives approval for Jake's spine to be fixed
-Puts Jake on a shuttle and into cryo on the ship
-Sends authorisation as stored data on the ISV, to arrive on Earth with the return trip, operation is done there.
I know you're getting desperate for points to use, but there's no flaw there.

For sufficiently large firepower, what's the problem? (Also, in this case, energy weapons might work better than kinetic rounds. Limitless, massless ammo simplifies a lot of the mechanics.)

Also reduced effectiveness in Pandora's atmosphere, increased maintenance (atmosphere again), and the necessary support structure for expensive experimental weapons.

The complication is worth the 80TW/person, IMO. Knowledge is massless.

Equipment, however, is not. Neither is bandwidth unlimited.

True enough, though I would suspect their budgets are vastly smaller than research into oil/gas/coal.

That is because they still have private industry behind them - when they start to become less profitable, all that research funding gets switched - but it hasn't yet.

Well, yes, but that takes a long time, because it doesn't all happen at once.

Nobody ever said it would.

Well, someone will find a way to maintain most of the power output. (probably by keeping the reactors running past their lifetime date)

Yes, it's certainly possible, but the truth is, no such steps have yet been taken and the deadline in terms of not creating an energy gap is rapidly approaching.

There are no serious proposals to get to Mars, (perhaps experiments along the lines of, "Are we sure this is feasible?" but nothing like a planned mission) let alone out of the solar system.

Only thanks to Obama undoing one the the few good things that Bush had actually tried to do. Constellation would have done it.
The entire point of Mars500 is a proper feasibility study - and indeed, there are huge numbers of people who would go, even if it was a one-way trip (something which has generally not been looked into as an option in such detail, while doing so would make it far easier.
The simple fact that it's being looked at shows human ambition - don't complain because it hasn't happened yet.

Then make it take 3 years each way and use unobtanium as fuel. FTL does not necessarily need to equal cheap, if the writer commands it so.

It also breaks the basis of real physics, as well as allowing people to go 'why travel FTL with the only destination being get something that is used for FTL?', not to mention complaints of it taking 3 years, still. Even moderately intelligent people profess to greater knowledge, and by having it slower than light, it seems more realistic than 'slow FTL' where a lot of people would expect all FTL to be very fast. It's a simple determination of which is perceived as realistic by people with a low to moderate understanding, who will be the most likely to complain.

OK, profit that isn't artificially generated for PR/tech growth purposes.

Congratulations on missing the point. The point was that what they are developing as a result can be used for large-scale 'real' applications. I thought that was obvious.

You seem to be suggesting that Cameron has considered all this and knows it's unworkable, rather than considering that he could have overlooked something that spoils his plot/world.

You seem to be looking for any way you can to bash Avatar. You'd be surprised at the detail, you need only look at the design process for the animals for an example - nothing made it into it that didn't work logically and physically.

As a more blatant example, nobody asks "What the hell are we doing?" with respect to giving Jake an Avatar.

No, because it was intended for Tom and they were trying to make the best of a bad situation. Did you even watch Avatar?

Or to be more specific, giving Jake an Avatar, then expecting him to give it up again. That's basic Hierarchy of Needs stuff.

Again - they would have lost more by leaving it on Earth and not trying to get some use of it with Jake, while from their perspective, even if Jake was only there for a short time and still benefited them in some way, they have recovered even a small amount of investment in Tom that would otherwise have been gone.

So long as they're physically feasible, assuming you want to be maintain "realism." A melting point of 100,000K, or a magnetic field strong enough to support 100 tons at 100m is just dancing around suspension of disbelief. At that point, you might as well call it... unobtanium phlebotinum.

Because Reality Is Unrealistic, right?
You're bashing a work of fiction with a habitable world in the closest star system to Earth, populated by a sentient species, with a perfect superconductor present there, over something that has never even been mentioned in canon.
I really couldn't make this up