3 ms·
You've got it wrong, BADLY wrong: For "novel programs" I don't give a rat's ass. "I can't figure out how that's apropos of anything" It's about charging into
by HilbertSpace 16y ago
You've got it wrong, BADLY wrong: For "novel programs" I don't give a rat's ass.
"I can't figure out how that's apropos of anything"
It's about charging into Web 2.0, 'consumer Internet', 'information technology' entrepreneurship, and the future of computing in business, that is, exploiting cheap cores, bytes, bandwidth, and infrastructure software, and making money, $1 billion, 10 times that, maybe 100 times that, maybe from one project, much more in total, and where for this some selected work in math is the crucial key.
E.g., just sitting right in front of us is an opportunity to suck the air out of old media, and for that a key is some math.
"so I imagine there's something that I'm missing".
Yup.
I can't explain it better than I did.
In communicating any of this, there are two BIG problems:
First, there are nearly no examples in, say, 'information technology' entrepreneurship. Lacking examples, it's super tough for people to see it.
Second, the math I'm talking about is not often taught, even in the best US research universities. It's rock solid math, but it's just not very popular.
So people are missing both the examples and the math. So, the whole thing looks empty or worse.
Sorry 'bout that.
I'm telling you the truth, but I can't explain it better than I did.
If many people readily understood the work, then the business opportunity would no longer be there. A good business opportunity and difficulty understanding it are love birds.
Broadly this situation is not very surprising: Anyone talking advanced math in Web 2.0, etc. is doing 'field crossing', but that technique, say, using clipper ships to bring pepper from India to Boston, make enough in one trip to pay for the ship, has to its credit a major fraction of the best business successes.
So, bringing math to Web 2.0 is field crossing. I'm here to advise you, for the future of cheap cores, etc. in business, a major key is selected work in math; I tried to give the reasons at a high level, e.g., about 'information'. Maybe someday I will underline this point with an example. For now, that's the best I can explain.