Interesting analogy, but the pessimistic view does not bring any progress. The optimistic view may often be wrong, but when it is actually right, it makes our civilization progress.
Not so long ago, the experts knew and said it was impossible to land and reuse a rocket.
What you call the "pessimistic view" might be very enthusiastic about projects that are orders of magnitude more likely to work. So in my opinion, the relevant axis is not optimistic vs. pessimistic, but informed vs. uninformed.
Can an outsider with no knowledge of the field make a chance discovery? It's theoretically possible, and we might want to support undirected experimentation for the one in a billion chance that something might work. But we should take those probabilities into account when allocating our collective resources.
This this this. Sooooo much. This always was flying spaghetti monster level unlikely. People simply do not understand how tightly constrained things like this are by existing evidence. And almost uniquely so.
When experimentalists announced they had discovered super-luminal neutrinos, everybody reacted exactly like all the amateurs here are saying we should: Explore potential theoretical ramifications while being aware that it's most likely measurement error and encourage further study until we're sure.
Not so for the EMDrive. Why? Because the idea that using nothing but electromagnetism you can violate this foundational property of every physical theory since Newton (_including_ electromagnetism) means that somehow your electromagnetic device, while operating well within the range of millions of other experiments out there, somehow triggered behaviour fundamentally different from electromagnetism. Different from the type of theory that electromagnetism is. Different from the type of theory that every physical theory ever found to describe reality is. And yet this phenomenon that is utterly _different_ than anything else we have ever seen went completely unnoticed in the millions of em sensitive experiments performed by everyone from undergrads to international labs over the last two centuries.
In the strictest sense it's not logically impossible that this could happen. But only in the sense that it's not logically impossible that the flying spaghetti monster is manipulating every scientific experiment performed to hide itself.
Correction: We had rockets that could land on the moon. There was no orbital-class rocket that could launch from Earth and land back at launch site (or further downrange). The difference in gravity + lack of atmosphere is a big factor.
It is like saying that the Apollo ascent module could get to orbit and therefore we can clearly make practical SSTO vehicles.
> We had rockets that were reused with the shuttle.
If you consider throwing away the massive external tank and external SRBs (without which it couldn't fly) and completely re-building the three engines after every flight.
> It is like saying that the Apollo ascent module could get to orbit and therefore we can clearly make practical SSTO vehicles.
If we didn't have to worry about fuel, we could. "put in enough fuel so it can land" is not a hard problem to solve with earth rockets, it just costs payload.
> completely re-building the three engines after every flight
It's complicated but the level of rebuilding was driven more by very high standards than it actually being necessary to make it launchable.
> "put in enough fuel so it can land" is not a hard problem to solve with earth rockets, it just costs payload.
I am pretty sure that the convex optimization algorithms used by SpaceX's landing systems are a very recent invention (http://larsblackmore.com/publications.htm). As for worrying about fuel, what's the point of a rocket if it cannot take any payload?
There was huge skepticism in the industry that landing and re-using first stage boosters was viable. Tony Bruno of ULA in 1015, 7 months before SpaceX recoverd their first booster, on their plan to recover the engine module of Vulcan:
"Our concept for reusability is inside that mathematics and realizes that maybe reusability does not start with entire stages,"
It's not that they thought it was impossible to do at all, but they thought the weight requirement of fuel, landing legs, heat protection, etc would be so high that it would eat up too much of your payload capacity.
Right, the reason ULA can't adapt Vulcan to land the first stage propulsively is it can't throttle down low enough. It's only barely possible with the F9, and that's because it uses a cluster of smaller engines and happens to have one in the middle.
That wasn't a design choice driven by reusability, originally SpaceX were aiming to recover the stage using parachutes. The original plan was a falcon with a cluster of 5 engines in a ring. Using a cluster of smaller engines was for two reasons. First because they didn't have the resources to develop a new larger engine. Second they wanted to mass produce the engines to get production efficiencies and a bigger run of smaller engines worked better for that.
It just worked out serendipitously that this lead to a design that was very well suited to adapting for propulsive landings. If they'd developed F5 first instead, they wouldn't have been able to land it because without a single centre engine they would need to use at least 2 engines to land, and the thrust would be way too high. Even with a design with a centre engine, the Merlin would be too powerful for the weight of an F5.
This isn't a criticism, it's just the way things worked out.
I see one little caveat with this argument, though: is it always known in advance that something really is impossible?
As long as there's enough resources to go around to pay 8 figure meme-value dollar amounts for digital assets without intrinsic value, I think there's no harm in spending a fraction of that on possibly fruitless research.
> 8 figure meme-value dollar amounts for digital assets without intrinsic value
The problem is the people who do that are hell bent against letting the government tax them to fund such research, so our research resources are immensely more constrained than they could be.
Not so long ago, the experts knew and said it was impossible to land and reuse a rocket.