Yet Gates' and Jobs' gold came from edge clients, not web media. Clients have never been more powerful, why is there no modern gold rush in client software? Is it merely that SaaS/hosted software does not suffer from piracy?
Other things is that the web is more universal. Say I have a great client app. It works perfectly on a client Windows machine. If I want to tap the larger market I have to have an OS X version, an Android version, iOS (depending if I didn't target both when doing the OS X version), and a Linux version. To keep all of these people happy, the versions should be pretty much the same. That's hard to do with client software on multiple targets.
The web targets the lowest common denominator: browsers. Fortunately as those go, the VM that is the browser is pretty powerful. Sure I might not be in my preferred language, but I can get things done. If done right, powers are added as the device becomes more capable my leveraging existing frameworks like Bootstrap, etc.
Gates and Jobs are pop-stars of technology. A few hundred years into the future, we're more likely to see Charles Babbage and Alan Turing in text books than Gates and Jobs. Those text books are unlikely to be read by humans in the same physical form we exist today (unless something catastrophic happens), and the class of intelligent beings dominating that period will trace their lineage directly to inventors and scientists than businessmen.
Whom should we remember: Gödel, Turing, or Cohen?
Unfair, that is not an answer to what he actually said:
> Those text books are unlikely to be read by humans in the same physical form we exist today
You're talking humans, he is not.
I love how computer scientists seem to decide they've "solved" AI because neural nets can sorta- kinda- do something that we do unconsciously (like see) and blithely assume that we just need a bigger net to get "thinking". When have we ever built something more complex than, say, a bunch of rocks piled on each other, without understanding its basic workings?
> I love how computer scientists seem to decide they've "solved" AI because neural nets can sorta- kinda- do something
Citation needed. It turns out there are exactly zero computer scientists who have ever said anything even remotely like that.
But again, that's wildly off topic.
Sounds like assumed strong AI to me.
> Citation needed. It turns out there are exactly zero computer scientists who have ever said anything even remotely like that.
Why cite, you've apparently done the lit search.
From memory there's:
"Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended."
That was in 1993 -- so we've got almost ten years left. Phew.
http://mindstalk.net/vinge/vinge-sing.html
https://en.wikipedia.org/wiki/Vernor_Vinge
Now he's a math professor, although his profession is currently "computer scientist" but I expect it to be very easy to find many other computer scientists who've said similar things, even going back to before neural nets. They may not be computer scientists you (or I) respect but...
I think some computer scientists spoke with similar confidence about LISP, but can't be bothered to find citations. (Obviously hard, since there aren't any, right?)
For really off-topic assumptions of this kind that push my button -- slower-than-light travel to other stars is often discussed as though it's a solved problem (e.g. Fermi's Paradox) despite the fact that it would require a really complex machine that was able to work flawlessly for hundreds or thousands of years. Again, something we're ludicrously far away from.
Far fewer are interested in computing so much as a gold rush that happens to use computing as its backdrop, which is overwhelmingly heading towards web-based media companies.