The data singularity is here | Hacker News Reader

The data singularity is here

dataspora.com

39 pointsypavan9 comments

"In a nutshell, the Data Singularity is this: humans are being spliced out of the data-driven processes around us, and frequently we aren’t even at the terminal node of action. International cargo shipments, high-frequency stock trades, and genetic diagnoses are all made without us."

I rather think automating mundane tasks is a good thing, because it frees us up to focus on innovative and creative opportunities. Jack Bauer operates at the speed of thought, because his tools take care of details irrelevant to him. IMO the data singularity exists, but it will shepard in the age of the exceptional individual. Startups (with HLLs) foreshadow this outcome.

IMO the data singularity exists, but it will shepard in the age of the exceptional individual.

Exceptional individual = Ubermensch?

I think there, is going to be plenty of scope for well oiled companies, especially small companies.

HLLs? The best I can get from a search is Heritage Learner.

But I agree, I think the startups who are good at managing lots of data, especially new data, extremely quickly will succeed.

My guess is High Level Languages.

Robin Hanson's latest talk* contends that automation is a complement, rather than a replacement, for human work; and that this holds true even when automation gets smarter.

* http://www.overcomingbias.com/2010/03/econ-of-nano-ai.html

I believe when the singularity hits, it won't be in the idea of a smarter machine making smarter machines.

It will be in decreasing communication lag-time and interoperability between /people/ asymptotically to nil.

We will still be individuals, but we will be tied inexorably to not only each other, but the vast amount of data stored on the internet.

What makes us /think/ better, however, is speed and utility of communication. What helps us /remember/ better is the speed and utility of information retrieval.

The latter has been increasing faster than the former. We need fundamentally gigantic changes in human-computer-human communication to affect the former. The first of these was the internet, of course. What we've been doing with it since its inception are comparative baby-steps - someday, perhaps soon, there will be a leap just as big as the Internet is.

And it will be fundamentally destabilizing to everything we know as society today.

The article seems to focus on the big picture of log files of services that are all interconnected in some way... this makes for the most impressive mental picture perhaps, forming "the tributaries feeding an ocean of data in the Cloud"; however, the exponential decrease in cost of computing power, data storage, and bandwidth has this effect on any information that we can digitize. Some of the most interesting developments will likely be in more or less isolated fields. I am excited to see the interplay between field-specific developments (finding patterns in a genome) and broader algorithmic work (finding patterns in any data). I suppose this is already happening, but it will become increasingly visible to laymen over time.

I'm not sure the singularity would make the change much faster. because in most industries the barriers to change are things like capital , time to test , and various human factors like risk aversion , politics , etc...

So change will be somewhat faster. The biggest difference will be higher level of machines taking people jobs , but we already have this. this will just increase.

they said the same things about auto-pilot and its interaction with ground control years ago.

Reply on news.ycombinator.com