Meshing Developmental Evolution and Technology

Jerry23 writes "IT Conversations has free audio of a very provocative talk by futurist and developmental systems theorist, John Smart. He weaves a big-picture narrative featuring developmental evolution, technological acceleration, computational autonomy, the emergence and behavior of human social systems, why prediction has such a poor history, the unique growth properties of Information and Communication Technology and the limitations of biotech, finally culminating in his case for the inevitability of digital personality capture and a ubiquitous Linguistic User Interface. Among many other things, he asks 'What will Windows (and the Google Browser) of 2015 look like?'"

Comments

Re:Too Limited

[kaiser423]

What will they look like? I really don't care.

Technology is going to progress, and by 2015 we'll have neat stuff, and by 2215 we'll have even neater stuff. End of story. It's not tied to Google or MS or anything else. It's tech.

Anyone theorizing about stuff now might as well go make their "predictions list" along with all the other Nostradmus' and people talking about flying cars.

Why predictions fail

[quokkapox]

Predicting future scenarios is extremely inaccurate because people tend to focus exclusively on one area and extrapolate it too far into the future without considering the inevitable interactions with other co-evolving technologies, cultural trends, and economic factors.

In a way, we do have our flying cars. It just turns out that most of us don't want/need/afford one parked in our driveway. A helicopter is essentially a flying car, but it's noisy, difficult to operate safely, and expensive to operate and maintain. Likewise, a jetliner is just a flying passenger train.

Nobody, including John Smart or Vernor Vinge, can make meaningful predictions any significant distance into the future. I think things are changing so fast now, that even 10-15 years into the future is almost inscrutable unless you're making very broad generalizations. You can say for sure: We'll have computers. They'll be real fast. But who knows what all we'll be doing with them.

Now with some good Intelligence Amplification, giving you the ability to consider the myriad variables and chart out many possibilities in future space, like a decision tree or a chess-playing program, and prune the unlikely ones, you can maybe construct a fuzzy map of the different courses the future will take. But you'll have to wait and see which one actually happens, just like everybody else.

Alright, I have to get back to brainwashing Jabberwacky ...

Re:Why predictions fail

[timeOday]

I think things are changing so fast now, that even 10-15 years into the future is almost inscrutable unless you're making very broad generalizations.

I'm not so sure about that. 10 years ago the cool thing was... the Internet. Has anything really changed in the last 10 years? Jurassic Park as a special effects feature now 12 years old, and it doesn't look particularly out of date.

My own theory of scientific progress is that while facts and theories proliferate exponentially, they tend to diminish in significance at the same rate. When a field is new, it's easy to make breakthroughs. As it matures, most of the big ideas have been thought but there are armies of people churning out lots of paper.

Look at medical research; the most significant breakthoughs were the easiest to make - like penicillin. Now we're pouring billions into Cancer year after year and making a little progress. Longevity and quality of life are not increasing exponentially.

Look at transportation, rockets and jets were invented 60 years ago and since then nothing has supplanted them. Passenger jets don't look much different from 40 years ago. Cars haven't changed fundamentally in approximately one lifetime.

I'm not saying change has ceased, only that I'm not sure things are really changing any faster now than they were a couple hundred years ago. Some poeple think we're accelerating ever faster towards an incomprehensible future with no continuity, I don't think so.

predicting the future

[argoff]

.. es easy. The least proprietary technology always wins out. Not the prettiest, not the best designed or the most elloquent. Always the least proprietary.

That's how intel made it, that's how windows (ironically) made it, that's how the tcp/ip internet made it, and that's how linux is going to make it today and why it will simply kick butt.

Re:Key breakthrough: the Intentional Stance

[noidentity]

If Google could infer why I am searching instead of just what I am searching, it would be able to do a much better job.

And, as with people, when it gets it wrong it's worse than if it was just a dumb but obedient tool. That's the problem I have with anything that presents itself as a mind-reader: when it doesn't read my mind, I have to read its mind to predict what it will do in response to my input.

In the end it makes things more complex. I'd rather have a tool whose response doesn't depend on what it thinks about me. It's the same with salesmen: don't get in my way when I know what I want to do.

Buzzword Bingo!

[dr.newton]

Welcome to Buzzword Bingo, I AM your host, John Smart. Can you spot the buzzwords?

Let's play!

accelerating change
I am just constantly suprised by new technological emergences
how do we socially interface with those
accelerating change
get in the zone
keep our eye on the ball
accelerating change
you can say this in the mirror every day
the future is now
it's already out there
a can-do, change-aware attitude
accelerating change
accelerating intelligence
intimacy of the human machine
evolutionary development - you're gonna hear this phrase a lot - anybody who uses this phrase thinks deeply about change
accelerating change

But seriously folks, that's about 5 minutes into an hour-long talk. Does this guy take himself seriously? Is he joking?

Smells like a leftover marketing plan from the Dot-com boom.

He needs to actually learn some biology

[Gravlox]

I always love it when smart people "think" they know biology. They always assume that mammals are the pinnacle of evolution, and that man is the ultimate mammal. Two eyes better than many? Obvioulsy this guy has never actually looked at arthropods. Walking upright is the ultimate? Using that as a rule, his Troodont should have been the master. It did walk upright. Running aint everything. Pluse, Troodont and it's kin lasted for longer than the raptors. Not a fair comparison- the really bad day when a huge chunk of flaming rock wipes out 70% of all life on the planet makes all the "who is better" arguments literally go up in smoke. Why do we have two eyes? Bilateral symmetry does seem to be popular amoung terrestrial vetebrates. Not surprising, since all terestrial vertebrates are descendents of fish. Have not seen any fish with 3 fillets on em. I hate when people try and attach meaning and direction to evolution. Evolution direction. There is no anti-entropy going on, other that the usual eating/sleeping/procreating. Ask Flipper if walking upright is a requirement for sentince. I cannot comment on his views on computing, but he comes across as a poseur to this biologist.

Out of touch

[adoarns]

I'm odd man out on this conversation, because I don't think human intelligence is computable, and I don't think we're likely to get things like universal constructors or technological singularities. Not for lack of enthusiasm, unfortunately.

But what strikes me in all the comments heretofore has been this idea of improving usability and efficiency by having the computer anticipate your actions, get to know you, listen in real language.

Well, I don't think this would at all be useful. I switched over to GNU/Linux two years ago on a lark, but I've stayed over because of the absolute richness of the toolset, especially textutils and text editors like vim--as I'm a writer, good text-manipulation is important.

These tools are generic, but precise. I have made my own toolchain to cope with the tasks I have to do each day. On Windows, I had Word. That's it--just Word. On Linux, I use awk, bash-scripting, perl, textutils, darcs, vim, (La)TeX and a host of others.

Having tools is where it's at. Better and more tools. Evolve tools and evolve toolchains.

Natural language is wonderful for human expression, but it's imprecise for detailed specification. Witness the development of mathematical notation, BNR notation, architectural schematics, UML. Programming languages aren't simply weird because it's easier to parse, but because their stilted format gives them predictable behavior. Real human language dips into and out of metaphor freely, invents neologisms, is imbued with dialect, invokes slang, and is more full than not of social and emotional content. Which makes writing stories really fun and easy, but is shit for writing programs--which, let's face it, are just automated tasks.

I don't want Windows Search to tweak my search based on the last fifty items I looked for; I *do* want to be able to tweak the Search myself so that it can bring up relavent text within the file, as well as strip some metainformation I myself added to it and display it. That's my idea of efficient.