The problem is that it's not clear what happens when the minimum wage increases. It's also not clear whether something different happens when a city or state does it versus a national change.

Case in point: thirteen states increase the minimum wage and employment increases faster (on average) in those states than in those that do not increase the minimum wage. The presumption in the post is that the causality is that increasing the minimum wage causes employment increases. What if the causality goes the other way? Increasing employment could make states more willing to raise the minimum wage. Correlation does not indicate causality, so economists can't differentiate between the two explanations.

There's actually been quite a bit of study of the effects of raising the minimum wage. The problem is that it's impossible to produce a real double blind study. Without that, there will always be reasonable doubt. In one study, they won't be able to eliminate the possibility that employment would have gone up faster without the change. In another study, they won't be able to tell if people are moving from the comparison area to the change area for the higher wage jobs. In another study, perhaps employment increases occur because kids drop out of school to take jobs.

Economics isn't anywhere near as mature a science as physics or chemistry. It doesn't lend itself to repeatable experiments. Without objective data, subjective opinions take a far greater role.

Actually, these actions seem driven by them not having a contract rather than chosen by Amazon. Without a contract, Amazon can't be sure of filling pre-orders at the promised price. Without a contract, Amazon can't return unsold books. The result is that they can't allow pre-orders or order books they aren't sure they'll sell. It also may be that Amazon normally gets credit for warehousing books with limited sales.

This is what the business looks like when selling without a contract.

Without an actual look at the contract negotiations, it's hard to say who's to blame for the lack of a contract. Perhaps both are. Perhaps authors are to blame for not demanding better contracts (authors often make less for digital books than print books, even though publishers charge Amazon more for them). The only thing that we can say externally is that book prices are often ridiculous. There is no reason for the digital copy of a new bestseller to be more expensive than the print copy.

Not only Jules Verne. What about H.G. Wells? I'd also include Arthur Conan Doyle. Most of the science fiction in Sherlock Holmes has turned into science fact, but in some ways that actually makes it more relevant for your purposes.

More recent classics would include the Heinlein juveniles (almost everything he wrote before Strange in a Strange Land plus some of what he wrote afterwards) and (as others mentioned) the Asimov books. The robot series is more for programmers than scientists, but there's quite a bit of interesting science floating around Asimov.

Not classics, but the Tom Swift and Danny Dunn books are oriented towards juveniles and promote science.