Isaac Asimov on what science fiction is: the reaction, not the action

[Admiral2]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

I feel like movie execs who think like you do are the reason we can't have nice things.

On the contrary. We'd have more nice things without the griping of fans as inflexible as you are.

To do the type of adaptation you want of the series so far we'd need a guarantee of ten seasons on Netflix or some other streamer. Only the first two seasons would focus on the core story, the rise of a midshipman in the Royal Navy during an interstellar conflict between empires. For the rest, you'd have to dump in backstory upon backstory about Sphinx, Grayson, the Manticore Junction, the Diaspora, the Solarian League, The Star Kingdom's government, Haven's government, pirates, POW camps, etc. etc.. And by the way, you'd have to spend half an hour every season on the history of treecats and humanity, and in season ten, the treecats have to join a Manticore/Haven alliance against the Solarian League. That's right. Treecats become the ewoks of the Honorverse.

You think this is a "nice thing" that will have any chance of success? I'm happy to take the side of the execs on this one.

 

[USS Triumphant]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

On the contrary. We'd have more nice things without the griping of fans as inflexible as you are.
<snip>
You think this is a "nice thing" that will have any chance of success? I'm happy to take the side of the execs on this one.

1. You really need to learn to recognize when someone is ribbing you.

2. I LOVE the Ewoks, so that all sounds pretty freakin' awesome. Ridiculous in some ways? Certainly. I think you're forgetting how long Hercules and Xena lasted, though.

 

[lpetrich]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

The Honor Harrington series is essentially just Horatio Hornblower IN SPACE! You can establish that with the opening battle scene, with vaguely galleon-shaped starships with energy sails firing missile broadsides at each other.

Is that what it's like in the Honor Harrington series? That's sort of like the car-as-mechanical-horse absurdity that I quoted in my OP.

Here's an example of that absurdity in visual-media SF: the design of the Star Wars Star Destroyers. It is designed like a sea warship, with its weapons and observation tower on one side, and with the other side being plain. In outer space, one ought to be able to fight in all directions, at least with one's defensive weapons.

Star Trek has another one: never seeing spaceships approaching sideways or upside down. In outer space, there will not be any well-defined "up" direction. So one can see something like this: Earth - Apollo 9 with an upside-down Lunar Module.

About 80% of the exposition in those books is universe building at the minutia level. Nobody needs to see thirty minutes of exposition about treecats and humans. Just CGI a treecat on its human's shoulder. Problem solved.

In the print version, this seems like something that one ought to add as appendixes to the main story. One might even do them in-universe in some way, like encyclopedia entries or investigators' reports.

 

[Silvercrest]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

But what's important is now how the hardware is constructed, but what one does with it. He got tired of robots killing their creators, so he decided that they must have safety mechanisms. His Three Laws of Robotics. Even here, these have no special connection with robotic hardware, so they may better be called the Three Laws of AI Systems. They can be further generalized as Three Laws of Tool Design, though those laws are often used implicitly. I've turned "robot" into "AI system" here, since one would likely need strong AI to implement them explicitly.

1. An AI system may not injure a human being or, through inaction, allow a human being to come to harm.
2. An AI system must obey the orders given it by human beings except where such orders would conflict with the First Law.
3. An AI system must protect its own existence as long as such protection does not conflict with the First or Second Laws.

There are lots of difficulties that arise when implementing these laws, difficulties that gave IA plenty of story possibilities.

The interesting thing about the Three Laws, and something that gets overlooked a lot, is this: They weren't added-in safeguards in the usual sense. They were stated to be an integral part of the positronic brain, so fundamental to the design that you couldn't build a positronic brain without including them. At least a working one, anyway.

That would have gotten Asimov around the question of why the military-industrial complex didn't just design "Terminator"- or "Second Variety"-style robots and just omit the First Law for convenience. They couldn't. That way Asimov could move on to the exploration of what it meant to have the Three Laws and what you could do with them.

As such, "difficulties in implementing" takes on a new meaning, since the Laws were already about as implemented as they could be. Some of the difficulties were a result of external situations where two or more Laws conflicted. Others were because of experimental robots. It might not have been possible to break the Three Laws, but they sure could be bent.

Non-positronic (and presumably non-Asenion) AI also existed in that universe, but it was implied they were not all that successful.

 

[Admiral2]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

On the contrary. We'd have more nice things without the griping of fans as inflexible as you are.
<snip>
You think this is a "nice thing" that will have any chance of success? I'm happy to take the side of the execs on this one.

1. You really need to learn to recognize when someone is ribbing you.

2. I LOVE the Ewoks, so that all sounds pretty freakin' awesome. Ridiculous in some ways? Certainly. I think you're forgetting how long Hercules and Xena lasted, though.

Not ten seasons, and the Herc/Xena universe didn't have anywhere near as much peripheral stuff to wade through...

 

[USS Triumphant]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

Not ten seasons, and the Herc/Xena universe didn't have anywhere near as much peripheral stuff to wade through...

Together (which I think is fair since they're a directly shared universe) they had 12, actually.

But never mind the HH series now - instead, I'm eagerly awaiting lpetrich's new novel. Since lpetrich seems to think he has writing advice for New York Times bestselling author David Weber, I'm just certain he's working on his own book, and it will be totally bitchin'.

 

[JoeZhang]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

As to SF being about the reaction and not the action, IA notes the issue of positronic brains, which is what his robots have. He introduced that as a future-tech detail, though it must be conceded that a positronic brain would be impractical. It takes a *lot* of energy to make a positron, energy released when a positron combined with an electron. About a MeV of energy, the energy that an electron or a proton will get when dropped through a million-volt potential difference. That's far more than the amount of energy necessary to knock an electron out of the outer layers of atoms, and such knocking is indeed what ionizing radiation does.

Just as an aside, he picks Positron because of its recent discovery but... do any of the stories actually mention that they have any to do with positrons?

 

[Silvercrest]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

I haven't read all the Robot-related books by any means, but I'm familiar with just one vague reference: Something to the effect that such brains "generate short-lived particles called positrons. During their brief existence, the positrons travel along paths that resemble a simplified version of the human brain."

If we can take "positron" to mean some technobabble particle instead of the real thing, this provides a certain rationale for my previous post. That is: why the Three Laws are so embedded and so unalterable.

Perhaps the positronic brain isn't written software and doesn't have "ethical subroutines" that can be deleted with a click. Instead, it was developed from some sort of active holographic model of brain functions. The developer learned to simulate brain patterns (including thought) using patterns of "positrons". The result is an AI that mirrors the basic human psyche that it was based on. This would include concepts of basic human ethics which could be codified as the Three Laws. You might be able to shift some of those positron flows to alter the brain's perception of the Three Laws, as with Cutie, Nestor, Speedy, etc. But you can't remove them completely without messing up the whole pattern.

The problem with this is that it implies the developer doesn't really know how the positronic brain works — he just mimicked something that already existed. Beyond a certain level, it would be "a wizard did it".

And I suspect there are other problems with this model that I can't see. Any opinions?

 

[USS Triumphant]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

If I recall correctly, it was mentioned in one of Asimov's stories that the problem of developing a positronic brain without the Three Laws wasn't a purely technical one, but also a financial and regulatory one. The brain fabrication equipment that US Robotics and Mechanical Men had was developed to only produce brains with the Three Laws - and it would have required decades and billions of dollars for them to recreate the wheel to make brains without them. The same would have been true for any other company, with the added difficulty that the government had given US Robotics a limited monopoly on producing robot brains and would have acted to shut down anyone else trying to do so.

 

[Silvercrest]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

That's kind of counterproductive from the government's point of view. If non-Asenion brains were viable (such as those produced by Consolidated), you'd think the government would support them instead. There's a lot of value in an AI that can kill.

It's why my other post implied they weren't all that successful. Other types of brains were possible, but must not have been particularly viable in the long run. The government certainly would have supported them if they were (to the extent of pouring billions of dollars and decades of time into them). As it was, they'd support U.S. Robots and their Asenion brains only because they had proven viability.

Edit to add: I take all that back. I was going by memory of the story "Escape!" in which Consolidated's AI broke because it was substandard to U.S. Robots'. But it was still an Asenion brain because what broke it was a First Law dilemma. So I guess I don't have anything to support the viability of other sorts of brains, and maybe you're correct.

 

[Admiral2]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

The persistence of the Three Laws in the Asimovverse has nothing to do with the technical ability of the manufacturers. Those are only superficial concerns. The whole point of them is to alleviate the innate fears of the public by creating automatons that (theoretically) can't rebel against their masters. If the government grants the creators/manufacturers of positronic brains a monopoly, it's to keep the voters complacent, not because the positronic brain is significantly better than any other kind of operating system.

The stories in I, Robot and the Bailey/Daneel novels pretty much center around what happens when those laws fail, and the worry is always about what happens to human beings when those failures occur. Discussing the robot stories the way you'd discuss Next Generation episodes - as if the technical hows and whys matter - misses the point.

 

[Silvercrest]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

Fair enough. The only reason I addressed it was to underline the author's point that the Three Laws were a constant . This was his shortcut around the concern of machines rebelling so that the stories could concentrate on other things. Discussing them as "safeguards" implies that they can be removed or circumvented — the very implications he was trying to avoid.

So I wanted to see if there was a remotely plausible basis for that kind of AI. But I see that it doesn't bear that sort of analysis very well. And as you say, going further down the rabbit hole does miss the point.

 

[Admiral2]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

Fair enough. The only reason I addressed it was to underline the author's point that the Three Laws were a constant . This was his shortcut around the concern of machines rebelling so that the stories could concentrate on other things. Discussing them as "safeguards" implies that they can be removed or circumvented — the very implications he was trying to avoid.

So I wanted to see if there was a remotely plausible basis for that kind of AI. But I see that it doesn't bear that sort of analysis very well. And as you say, going further down the rabbit hole does miss the point.

Got it. Just remember that that same author also freely admitted he knew nothing about building actual robots, and never approached any of the stories from a truly technical perspective. He considered himself a social scientist, so the robot stories were studies of human beings interacting with their mechanical slaves. The Three Laws are constant because they drive the interactions in the stories. Whether the concept is viable or not for an actual AI is irrelevant, because the guy who made it up didn't care one way or another.

In fact, people who actually build robots would probably tell you immediately that Three Laws couldn't work in real life - and again, most of the robot stories involve the laws not working - so you'd be right in thinking such a thing wouldn't be plausible, but plausibility was never the issue.

 

[Reverend]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

Heinlein - Starship Troopers was a stretch, and its sequels boggled the mind - not in a good way.
But I tend to think his stuff is too irreverent (sexually, too) for wide acceptance. And heady.

True, the later stuff is just downright creepy, but I think something interesting can be still done with 'Puppet Masters', even after several remakes of 'Invasion of the Body Snatchers'. Mind you, given today's social climate they'll probably just try and use it as an allegory for radicalization and domestic terrorism instead of the "red scare" anxiety of the day.
My personal preference would be to go for a faithful retro-future version of the 90's complete with flying cars and analogue electronics, just for the sake of novelty.

On a slightly related note, I'm actually surprised that when I thought about it, it seems the only "Titan" of literary sci-fi to get the most exposure (for better or worse) in cinema is Phillip K. Dick. Probably followed by Wells & Verne, those most of that seems to be re-makes of the same two or three stories over and over..

I suspect that Richard Matheson can give PKD a run for his money when it comes to Hollywood adaptations: The Incredible Shrinking Man, I Am Legend (three times), The Legend of Hell House, Somewhere in Time, What Dreams May Come, Stir of Echoes, Real Steel, The Box . . . and that's not even counting umpteen TV adaptations of his work.

Granted, most of those are more horror/fantasy than sci-fi, but Matheson has always been hard to pigeonhole.

And Wells has had more adaptations than you might think: The Invisible Man, The Island of Doctor Moreau (three times), Things to Come, The Man Who Could Work Miracles, The Time Machine (three times), The War of the Worlds (twice), The Food of the Goods (twice), and The First Men in the Moon.

As for Verne, we've had adaptations of Twenty Thousand Leagues Beneath the Sea (at least twice), Around the World in Eighty Days (twice), The Mysterious Island (twice), Journey to the Center of the Earth (twice?), Master of the World, and Five Weeks in a Balloon.

Meanwhile, Harry Harrison's "Stainless Steel Rat" books have been optioned forever, but we're still waiting for the movie.

You know I didn't even consider Matheson since, as you say, though he's written some of the best fantasy/sci-fi out there I never really think of him as specifically a sci-fi writer as I do for the likes of Wells, Verne, Clarke, Asimov, Heinlein, Niven, Ellison, Niven, etc. etc.


I know it's no excuse but a lot of those Wells & Verne adaptations are way before my time so I was actually only aware of a handful of them.
Anyway, does this count as a true Verne adaptation? It was probably my first ever exposer to his work growing up.

 

[Greg Cox]

Re: Isaac Asimov on what science fiction is: the reaction, not the act

On a slightly related note, I'm actually surprised that when I thought about it, it seems the only "Titan" of literary sci-fi to get the most exposure (for better or worse) in cinema is Phillip K. Dick. Probably followed by Wells & Verne, those most of that seems to be re-makes of the same two or three stories over and over..

I suspect that Richard Matheson can give PKD a run for his money when it comes to Hollywood adaptations: The Incredible Shrinking Man, I Am Legend (three times), The Legend of Hell House, Somewhere in Time, What Dreams May Come, Stir of Echoes, Real Steel, The Box . . . and that's not even counting umpteen TV adaptations of his work.

Granted, most of those are more horror/fantasy than sci-fi, but Matheson has always been hard to pigeonhole.

And Wells has had more adaptations than you might think: The Invisible Man, The Island of Doctor Moreau (three times), Things to Come, The Man Who Could Work Miracles, The Time Machine (three times), The War of the Worlds (twice), The Food of the Goods (twice), and The First Men in the Moon.

As for Verne, we've had adaptations of Twenty Thousand Leagues Beneath the Sea (at least twice), Around the World in Eighty Days (twice), The Mysterious Island (twice), Journey to the Center of the Earth (twice?), Master of the World, and Five Weeks in a Balloon.

Meanwhile, Harry Harrison's "Stainless Steel Rat" books have been optioned forever, but we're still waiting for the movie.

You know I didn't even consider Matheson since, as you say, though he's written some of the best fantasy/sci-fi out there I never really think of him as specifically a sci-fi writer as I do for the likes of Wells, Verne, Clarke, Asimov, Heinlein, Niven, Ellison, Niven, etc. etc.


I know it's no excuse but a lot of those Wells & Verne adaptations are way before my time so I was actually only aware of a handful of them.
Anyway, does this count as a true Verne adaptation? It was probably my first ever exposer to his work growing up.

I confess I was unaware of that "Verne" cartoon!

At the risk of showing my age, my first exposure to Verne was probably the old Disney version "20,000 Leagues" with James Mason, and the 1960s version of "Center to the Earth," also with James Mason!

And, of course, the classic Ray Harryhausen version of "The Mysterious Island" was a staple of Saturday afternoon TV when I was growing up . . . .