Jump to content

2015 budget. NASA gets a raise! Wooo Hoo!


Aethon

Recommended Posts

Time delay, bandwidth. You don't notice, nor can you check as many things when you are (for all practical purposes) passively watching video feed as if you are actively moving around in real time, searching.

I wouldn't be so sure of that. There are multiple teams monitoring a whole slew of camera and sensor feeds transmitted by every camera, over nearly every inch of ground covered by a rover. What do you think they might have missed ?

If anything, a single driver who is concentrated on driving over the terrain would be less efficient.

Time delay is no big deal if you're going slow, and neither is bandwidth. Time is one resource that unmanned missions have plenty of. Again, there is no hurry, the rocks aren't going anywhere.

Edited by Nibb31
Link to comment
Share on other sites

I wouldn't be so sure of that. There are multiple teams monitoring a whole slew of camera and sensor feeds transmitted by every camera, over nearly every inch of ground covered by a rover. What do you think they might have missed ?

Trowing more people at it won't solve the basic problem of inadequacy of passively watching compared to interactively searching.

If anything, a single driver who is concentrated on driving over the terrain would be less efficient.

Why single driver ? why just driving by and not stopping to walk and look around ?

Time delay is no big deal if you're going slow, and neither is bandwidth. Time is one resource that unmanned missions have plenty of. Again, there is no hurry, the rocks aren't going anywhere.

except that the delay and bandwidth issues get more severe the further you go. Just slowing things down is practical on mars, maybe jupiter. But beyond that ? rocks may wait another billion of years, but the rover has limited life expectancy, and so have the humans operating it. So good luck with that strategy researching the kuiper belt and beyond.

Link to comment
Share on other sites

This is one of the saddest things I read this month.

It is still not as sad as the cancellation of the Superconducting Super Collider: The government spent one billion dollars digging a hole and then another billion filling it back up, making it, as Michio Kaku wryly observed, "the most expensive hole in history." And that is what politics does to science. *Sigh*

Link to comment
Share on other sites

Trowing more people at it won't solve the basic problem of inadequacy of passively watching compared to interactively searching.

Do you have any idea of the operations of running something like Curiosity? I wouldn't call those teams "passive" at any rate. Each inch of ground covered by the rover is actively scrutinized and tested. A human simply cannot be as thorough , especially if he also has to concentrate on navigating, avoiding rocks, watching his oxygen and battery levels, checking on his buddy, returning to the base for lunch and a nap, and communicating with the base.

And again, what do you really think they might have missed?

The power, speed, and sensor limitations of rovers and probes are practically all due to the mass limitations of getting stuff from the Earth to Mars. If you are capable of sending a 100 ton manned expedition, then you could also send one hell of a rover, extra batteries, spare parts, literally tons of sensors and experiments, and even a couple of teleoperated robonauts to do an oil change on the rover every 20000 km.

except that the delay and bandwidth issues get more severe the further you go. Just slowing things down is practical on mars, maybe jupiter. But beyond that ? rocks may wait another billion of years, but the rover has limited life expectancy, and so have the humans operating it. So good luck with that strategy researching the kuiper belt and beyond.

We aren't sending people to Jupiter or the Kuiper belt in this century, so that point is moot.

However, if we ever do develop the technology to send people beyond Mars, it's pretty likely that robotics, sensors, AI, and communication bandwidth will also have improved. In fact, I'm willing to believe that those areas will advance much faster than interstellar propulsion and manned spaceflight technology, which will make the decision to send people even harder than it is today.

Link to comment
Share on other sites

I could go on about you massively underestimating the advantage of interactivity while searching, and the diminishing returns of throwing more people at observing the same visual information, but this is more important, because it deals with end results of your approach.

However, if we ever do develop the technology to send people beyond Mars, it's pretty likely that robotics, sensors, AI, and communication bandwidth will also have improved. In fact, I'm willing to believe that those areas will advance much faster than interstellar propulsion and manned spaceflight technology, which will make the decision to send people even harder than it is today.

Sure. AI improved to the point where it can make all the decisions locally, which implies at least human intelligence. So we end up starting an interstellar civilization of not humans but sentient robotic researches, and thinking they will throw some bits of information our way, hoping they won't turn around and start treating us just like any other animal to research. To that I can only say, what's sauce for the goose, it's sauce for the gander. If we shall give up the rest of the universe to the robots, why to hang around and waste oxygen at all, and not replace us with robots down here too ?

Edited by MBobrik
Link to comment
Share on other sites

hoping they won't turn around and start treating us just like any other animal to research.

Something I've never personally understood is this irrational fear of robots and AIs that some people seem to have, why do you guys think like that? Why are advanced robots and AIs immediately "EVIL BAD TERROR NO NO NO"?

I'm not even sure where to start asking or talking regarding that opinion, because it's not built on anything logical or scientific so far as I can tell.

Link to comment
Share on other sites

It's just a silly strawman argument. If we are clever enough to design interplanetary robotic exploration missions, then surely we are clever enough to design them so that they don't want to destroy us.

You don't need AI that can make decisions to wipe out humanity. You only need to make AI that can detect interesting patterns and pre-analyze data so that you can shorten the feedback loop and

But this discussion is an exercice in futility. The truth is, we don't know what technology we might or might not have in 50 years when we decide to focus on exploring Jupiter or the Kuiper belt. I'd rather look at goals that are actually reachable.

Link to comment
Share on other sites

Something I've never personally understood is this irrational fear of robots and AIs that some people seem to have, why do you guys think like that? Why are advanced robots and AIs immediately "EVIL BAD TERROR NO NO NO"?

They probably haven't read enough Iain M Banks.

Link to comment
Share on other sites

Something I've never personally understood is this irrational fear of robots and AIs that some people seem to have, why do you guys think like that? Why are advanced robots and AIs immediately "EVIL BAD TERROR NO NO NO"?

Are we out to exterminate squirrels ? No. I was not talking about AIs being deliberately speciecidal or even malevolent. Just about evolving past their initial programming and going their separate ways to the point where we are just another species of local fauna for them.

Link to comment
Share on other sites

It's just a silly strawman argument. If we are clever enough to design interplanetary robotic exploration missions, then surely we are clever enough to design them so that they don't want to destroy us.

No we won't. You can't predict, control and forever predestine thinking of something as clever as you are. They would most likely not destroy us because they would have no motivation to do so, but that would be beyond our control.

You only need to make AI that can detect interesting patterns and pre-analyze data so that you can shorten the feedback loop

Except the level of pre-analyzing that would allow the robot roam free and search, eventually dealing with contingencies, while the earliest possible human reaction can be next day to decades, would amount to human level od understanding of what's going on.

The truth is, we don't know what technology we might or might not have in 50 years when we decide to focus on exploring Jupiter or the Kuiper belt. I'd rather look at goals that are actually reachable.

The truth is, that short term thinking is one of the mankind's greatest scourges. If it weren't we would be already firmly established on Mars and planning to move to Jupiter next.

Link to comment
Share on other sites

Well, AIs are considered a problem because they are unpredictable. Not only that but they can calculate probabilities and can overreact. Extremely. Plus, what happens when an AI becomes psychotic?

Replace "AIs" with "living beings" (or "humans", if you prefer) and suddenly all is usual and standard fare. The argument is invalid.

Are we out to exterminate squirrels ? No. I was not talking about AIs being deliberately speciecidal or even malevolent. Just about evolving past their initial programming and going their separate ways to the point where we are just another species of local fauna for them.

I don't see where the problem lies there.

---

I apologize if I sound like I'm trolling with my replies, but I am simply and utterly failing to grasp what is behind the idea that advanced robots/AIs are bad.

Link to comment
Share on other sites

I apologize if I sound like I'm trolling with my replies, but I am simply and utterly failing to grasp what is behind the idea that advanced robots/AIs are bad.

No they aren't at all. It's the 'leaving the rest of the universe to them' part of Nibb's version of our future that I don't like.

Link to comment
Share on other sites

No they aren't at all. It's the 'leaving the rest of the universe to them' part of Nibb's version of our future that I don't like.

Another strawman.

Robots and probes are not "them". They are "us". They are the tools that we use to explore places where we can't go and to see stuff that we can't see, because we have biological limits. We have always used tools to improve our limited capabilities. A space probe is no different.

We use robots regularly to repair deep-sea pipelines, to inspect nuclear reactors, or to fly over enemy territory. Do you suggest that we should send people to those places too, just because it's more adventurous?

Edited by Nibb31
Link to comment
Share on other sites

Robots and probes are not "them". They are "us". They are the tools we use to explore places where we can't go and to see stuff that we can't see, because we have biological limits. We have always used tools to improve our limited capabilities. A space probe is no different.

We use robots regularly to repair deep-sea pipelines or to inspect nuclear reactors or to fly over enemy territory. Do you suggest that we should send people in those places too, just because it's more adventurous?

You are comparing the uncomparable. Pipelines with the universe. Simple tools with artificial sentient beings. I am perfectly willing to leave routine maintenance at the bottom of the sea to the robots (we can go there any time we want anyway), but I am not willing to pass on the rest of the entire universe. And while today's simple tools are mere extensions of ourselves, future sentient AIs would be independent beings on their own, not us. We may consider them our progeny, but they won't be us. You may be content with creating a species more resistant to the rigors of space travel, and leaving the future to them, but I am not.

Link to comment
Share on other sites

You are comparing the uncomparable. Pipelines with the universe. Simple tools with artificial sentient beings.

You're the one constructing strawman silliness about artificial sentient beings. I'm merely talking about reducing the communication loop by using space probes that are slightly smarter than the ones we have. For example, instead of beaming back terabytes of data, analyse the data on-site to detect and extract specific patterns and transmit that particular data at a higher resolution or with a higher priority. Or instead of micromanaging each wheel motor from mission control, just input a set of coordinates and let the rover figure out how to get there. That's hardly creating a master race of robot overlords.

Edited by Nibb31
Link to comment
Share on other sites

@Nibb

@MBobrik

I think you each misunderstand the other's point: Nibb wants to have probes study the universe for us, and MBobrik wants to humanity to expand into space. These two ideas are not mutually-exclusive.

-Duxwing

Link to comment
Share on other sites

We certainly need both manned and unmanned exploration.

Probes can go to the new frontiers where we still/nev cannot send people, while manned flight can gradually push forwards, first Moon, then Mars, then asteroid belt, then Callisto etc.

Link to comment
Share on other sites

NASA should spend a lot more on alternative propulsion methods, not building the SLS and Orion. We will not get any humans beyond Mars with out highly efficient engines.

If NASA didn't build the shuttle, and we went straight to NTR's and other high ISP, low trust engines, we would be on Mars by now. I still don't understand why we keep wasting time in LEO.

Link to comment
Share on other sites

You're the one constructing strawman silliness about artificial sentient beings. I'm merely talking about reducing the communication loop by using space probes that are slightly smarter than the ones we have. For example, instead of beaming back terabytes of data, analyse the data on-site to detect and extract specific patterns and transmit that particular data at a higher resolution or with a higher priority. Or instead of micromanaging each wheel motor from mission control, just input a set of coordinates and let the rover figure out how to get there. That's hardly creating a master race of robot overlords.

I am sorry, I failed to notice that even though I started about what would be necessary to explore beyond the distance where the time delay makes things like just better compression, data filtering, autonomous driving, and such insufficient, you can not even imagine that far and thus you couldn't be talking about that situation. Hard to make it obvious to you... so lets crank it up a little... imagine a probe to alpha centauri. how it could gather data efficiently when the response won't come back until long after the mission ends ? The only solution, methinks, is that the probe itself understands what it found and adapts its further exploration accordingly. anything less would lead to gross inefficiencies - first the probe catches a glimpse of something, it will take eight years to redirect it to examine it more, when it turns out to be something unexpected, further eight years to tell the probe to change its approach, and so on...

Edited by MBobrik
Link to comment
Share on other sites

Nibb wants to have probes study the universe for us, and MBobrik wants to humanity to expand into space. These two ideas are not mutually-exclusive.

Actually, you've got it wrong. I want both but Nibb rejects the latter.

Link to comment
Share on other sites

Actually, you've got it wrong. I want both but Nibb rejects the latter.

You don't get it, do you? I want flags and footprints and space hotels on Europa as much as anybody else in this forum. The problem is that what any of us "wants" is irrelevant. The point is that there is no rational justification for manned exploration in the foreseeable future.

And I have no interest in discussing manned travel to Alpha Centauri, Mars colonies, or stuff that might happen in 2 or 300 years. The world, and even humanity, was completely different 200 years ago. Speculations about the future are going to be as laughable then as Flash Gordon and Meliès's Man in the Moon are to us now. We can't even start to comprehend what humanity's aspirations, dreams or motivations will be in 200 years, so making assumptions is pointless.

Edited by Nibb31
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...