Smart phones are disappointing in a lot of ways but they're a big improvement over J2ME flip phones, dumb phones, and PDAs.
The problem with the Amiga was AFIAK Commodore's management. They should have divested themselves of all the old stuff and doubled down on marketing and supporting Amiga. Instead they kept lots of irons in the fire, wasted a lot of money, and in the end got killed by the IBM PC platform juggernaut. I wonder if the clear superiority of their tech made them over-confident.
Perl was a mature, full-featured and very useful programming language when the earliest versions of Java rolled around. Java relied on the bloated, super slow JVM, was super buggy, and really did not add any value. The "write once, run anywhere" hype turned out to be just that. So I really don't see how it was a "huge step up" in comparison. More like a huge step down.
So that's about how Java was back then, and whether it was a "huge step up" from Perl. It wasn't.
Yes, Java improved greatly over time, and yes it's more popular now. Whether its current popularity is deserved or not is debatable, and a separate issue.
Slow compared to what? Both making phone call and sending text message are as fast as dumb phones used to be. The other functions like browsing or playing does not have much to compare to. Maybe some PDA somewhere is faster, but I cant make calls with it so it is irrelevant.
Customers don't care about bloat, unless it interferes with them somehow. Nor have reason to care about it. Bloat is just functions they do not use nor care about.
There is some theoretical network effect in place when it comes to phones. Even if you would make great linux or whatever phone, if I could not run my android games on it I would not want it. However, linux phone I have seen was neither fast nor impressive, so it does not really matter.
Not in my experience. Anything less than a new flagship model and slowdowns are not uncommon, especially when the phone is updating or syncing or whatever.
Much more important than technological features are things like marketing, conformity, and groupthink.
The lesson was reinforced when so many developers and companies started jumping all over the earliest versions of Java, which was complete and utter garbage, and was untested, slow, and immature back then. Yet people were drooling over it.
Today's smartphones are yet another instance of this: they're slow, bloated, and spyware-infested, yet people love them.
Similar things have happened again and again, making me sometimes suspect that if the worse a technology is, the more successful it is likely to be. Worse is better indeed.