Not sophisticated at all. Due to a combination of the Second Law of Thermodynamics (time moves in one direction) and the Epistemological boundaries of Humans (we are limited in what we can Know).
That’s the problem with Unknown Unknowns - you can’t anticipate them 
I just reread that, and not really. The reason you might think this is that some of the hippies adopted terminology from the book, so the “prediction” kind of made itself come true.
Clarke just noted that a geostationary orbit would be useful. Relay satellites are kind of like relay stations for phone service, and anyway George O. Smith predicted them (but not in Earth orbit) before Clarke’s paper.
Clarke, like everyone else, predicted that these would be manned. He missed transistors, which allowed even the first satellites to be unmanned, and thus a lot cheaper.
A better prediction he made was that someday long distance would cost no more than local - pretty much where we are today within the US.
And everywhere if you use Skype or an equivalent. I talk to my daughter in Hong Kong all the time for free.
Mine too. If you look at sf magazines from 1970 or so, especially Galaxy/If, the consensus was that 2000 would be polluted and horribly overpopulated in the US, with cities unlivable. Make Room, Make Room is just one example. For once reality was far better than the consensus.
The TOS utopian future is a minority view, and even those who took over the franchise seemed to hate it.
Verne was kinda right about getting news on your phone. All news outlets now have websites. And Apple is bigger than my head. ( well Iphone x is kinda large).
Rooms full of people delivering news, uh, I don’t know, telemarketing maybe?
Some of us (and by “us” I mean me) would argue that is a fairly good description of the US.
I have seen a bunch of these based around a ‘resource shortage’.
We’re all screwed because we’ll run out of freshwater and we’ll all die of thirst. We’re all screwed because Saudi Arabia will finally run dry any day now and we won’t be able to find replacements fast enough. We’re all screwed because we’re about to run out of copper or cobalt or natural gas, etc.
The problem with these predictions is all of them ignore scale. Scale wise, we have barely touched the most easily found resources within a mile or 2 of the surface of the crust in the most easily accessible landmasses on one planet. Big picture wise, we are many orders of magnitude away from real resource shortages (we run out of mineable solid matter in our star system).
Instead they just look at what resources that mining/hydrocarbon companies have bothered to pay exploration teams to go find. No company is going to be interested in finding more than they plan to actually collect within some timespan of 10-20 years, give or take.
With all that said, the looming threat of nuclear war is a big one. It’s a series of accidents/human screwups waiting to happen. The fact it hasn’t happened yet is because the daily probability of it happening is low, but it doesn’t make the risk not there. A marginally sane man is in charge of the US nuclear arsenal, and a strongman in russia has found it politically viable to build a bunch more nuclear weapons.
Science Fiction doesn’t try to predict the future. It’s always about the present and any extrapolation is just using the model “If this goes on…”
It has occasionally come up with guesses that were pretty accurate (see “A Logic Named Joe”), but that’s just happenstance.
Something will replace the tools we use now that we haven’t quite identified yet. All current issues regarding these changes are just like hats in the 1900,s.
Everyone had a hat.
If Verne was writing predictions about 2889, isn’t it a little presumptive to be declaring how, why, or even if he was mistaken?
Some books which try to explain why most prediction is poor:
Ivor Baddiel and Jonny Zucker Never in a Million Years: This is a funny catalog of poor predictions, although it’s not very good at explaining why those predictions are bad, some of the jokes don’t work, and the people who did the predictions often should never have been listened to in the first place.
Dan Gardner Future Babble: Why Expert Predictions Fail – and Why We Believe Them Anyway: This is a better book than the previous one, since it looks strictly at predictions by people who would be expected to do better at predicting and gives some reasons for why making predictions is so hard.
Philip E. Tetlock Expert Political Judgment: How Good Is It? How Can We Know?: This is about a decades-long experiment by a psychologist who looked at the predictions of good contemporary political experts and showed that they aren’t really very good at predicting the future situation in their own specialties. Some moderately difficult mathematics is involved in testing the predictions. There’s a mathematical technique called Brier scores that allow you to accurately measure the usefulness of predictions.
Philip E. Tetlock and Dan Gardner Superforecasting: The Art and Science of Prediction: A team of psychologists led by Tetlock discovered a way to teach people to make more accurate predictions. They were able to show that only 2% even of reasonably intelligent people could learn to do this. These people are not the ones who were deeply trained in the subject that they were making predictions about. Experts often get stuck in their theories and can’t see when they have to move on because of new evidence. The people that are best at prediction are those who know how to evaluate both new and old evidence and humbly acknowledge that their previous predictions (and ideas in general) were sometimes inaccurate. Gardner is a journalist, not a psychologist, so his contribution to the book was rewriting it to explain the experiments to a more general audience.
Daniel Kahneman Thinking, Fast and Slow: This is a popularized account of Kahneman’s (and Amos Tversky’s) theories about how people are affected by their unconscious assumptions. Being good at predicting is partly about getting past your unconscious assumptions. Kahneman won a Nobel Prize in Economics (Tversky had already died by that time, so he didn’t get one) for showing how unconscious assumptions affect economic choices.
Also, “popular magazines” are especially susceptible to applying very little rigor to their predictions, preferring a sensational story based on something that “could” happen. So Popular Mechanics in any particular decade may have been writing about Mach 3 Flying Wing airliners or about moving-conveyor interstates or atomic home furnaces or airplane runways built as bridged platforms across the tops of skyscrapers. But at any place where real money got invested, these were just the stuff of brainstorming sessions, actual money was getting put into the things that ended up actually built (OK so *some *notions of the “city of the future” *were *made into policy, often with calamitous results in quality of life – high-rise public housing blocks, anyone?). Similarly, the serious scientists at the academic departments in the mid 70s were not the ones announcing an imminent new Ice Age.
As mentioned before many precictions were based on a fallacy of extrapolation in a vacuum. We go from the Wright Flyer in 1903 to the 747 and Concorde and Apollo LEM in 1969, so of course by the 2020s we’ll be taking regular scheduled flights to Mars and Jupiter. Except that at 1969 we were at the point where any further improvement was barely marginal and any attempt at something spectacularly greater would be a money pit. Or we are going to end up reprocessing our excess population into Soylent Green… except that the birthrate in industrial societies plummets at the same time as agricultural yields expand. Or automation is going to lead to a life of leisure and 15 hour workweeks for everyone… except employers will just fire your now-superfluous ass and you’ll have to work at three menial jobs to survive.
On the Science Fiction side, as RealityChuck points out, SF is not about reliably and objectively predicting the future. It is about making some point on ***our ***human condition, by setting the story in a world that has been altered by technological or social transformations so that the usual answers no longer apply. For which you can choose to strictly stretch only those bits that make a difference for the story, right to the point they serve the story, or else go whole hog and apply Clarke’s Third Law to make your technological and social development so advanced or incomprehensible it’s de facto magic.
From today’s news, my interest was piqued by the 5G story. The spectrum will be opened for 20 years, but there is loud insistence by some for 25, which seems like a trivial difference. What do they think is going to happen between 2040 and 2045, that makes this time frame so critical? What could they know, that we can only guess?
I thought Hitler would turn up by now. Or Hillary.
We’ve had flying cars for a long time. They’re not practical because extra cost is involved to make something dual purpose. The end result is something that doesn’t drive or fly well but costs more than the individual vehicles that do a better job.
We’ve had videophone wrist watches. Again, they weren’t practical. It’s awkward to talk into your wrist when a blue tooth ear piece is so much easier and their’s not much point looking at everyone you talk to. We have Skype for that.
What was predicted weredeath rays and robots and cars that drive themselves. All of that is coming true.
The free love, communal living and new religion items were surely not done just to live out a book although it may well have had an influence. One of the functions of art is as a sort of “pre-prefrontal cortex” where we run simulations of novel practices and communicate them to others who can then perform them.
2001 Space Odyssey has ipads and Skype. 2001 is also notable in having few crewmembers aboard the ship which implies a high level of automation. All of which mean that Kubrick predicted growth in the capabilities of computers. Sci-fi of that era seemed much more focused on mechanical, chemical and aeronautical engineering than computers.
Perhaps we could name other examples of works that got predictions right so we can learn from that.
Are there works which predicted the use of IT in warfare beyond fantastical robots? Things like guided munitions, drones, networked warfare, multispectral sensors?
Another important note is that 2001, etc, had fusion powered spacecraft because there was a compelling reason - to investigate alien artifacts. I think if we’d found alien artifacts on the Moon and knew there were more at jupiter - especially since the monoliths aren’t just artifacts, they use active systems and were clearly still functional and possibly intelligent - we’d find a way to get there.
Also, didn’t the monoliths manipulate the minds of humans in close proximity through some alien Clark-Tech? Maybe the critical innovations needed to make a fusion drive work came from scientists who saw a monolith.
I’ll second Gardner’s Future Babble on this list, a great look at exactly this topic. If you’re really interested, it’s well worth a read, and he goes into a bit of what makes people better predictors too.