Quote:
|
IT’S 8 a.m., Tuesday, Nov. 18, 2008, and you are headed for a business appointment 300 mi. away. You slide into your sleek, two-passenger air-cushion car, press a sequence of buttons and the national traffic computer notes your destination, figures out the current traffic situation and signals your car to slide out of the garage. Hands free, you sit back and begin to read the morning paper—which is flashed on a flat TV screen over the car’s dashboard. Tapping a button changes the page.
|
Laugh all you want, but it's impressive what they got right:
o The importance of computers
o Shopping on-line
o Credit cards
o Direct Deposit
o On-line education
o On-line trip planning
They failed to predict the Internet, of course, which I believe will go down in history as the single most important human achievement (to date). Yes, more than movable type. Yes, more than antibiotics.
Things they completely missed the mark on due to sensationalism-inspired stupidity:
o The idea that entire city structures would change to include domes and central planning. Even though the basic idea of a city hasn't changed in a thousand years, entire metropolises would be torn down and rebuilt over the span of 40 years. Yeah. Right. Fucking retards.
o The idea that computers would control cars. We're
still several generations away from people accepting this today, even though the technology to do it is probably within our grasp. Thing is, a single death related to a computer mistake would be more feared by people than a thousand drunk driver deaths.
o Extrapolating the historic trend of transportation ease forward. The world keeps getting smaller, true, and to writers in 1968 they could only extrapolate the trend of faster, cheaper transportation. Instead we see more people
using airlines than forty years ago, but certainly not intercontinental rockets. And yet the world
has gotten smaller, due to information exchange. What the authors failed to predict was a shift from physical presence to an acceptance of virtual meeting places, conference calls, etc.
o Space travel. Small-minded people typically viewed space as yet another frontier to conquer. Space is a bit different than an undiscovered island or even Antartica, though, ain't it? Add to the fact that there just isn't much gain to be realized from having humans in space, other than an economy of space tourism (which is finally starting to appear). It will be slow in coming. The adventuring part of me is determined to reach at least orbit in my lifetime. It's something I'd pay almost anything to do. The rational part of me realizes that humans have little business being there.
Anyway, fun to read. Want my predictions for 2048?
o People will still drive cars. Many will still be gasoline-based, but at least half will be electric.
o Nuclear power will experience a brief resurgence, however will be rendered mute by more efficient energy use and generation. No single source of power will prevail, but a combination of solar, tidal, wind, nuclear, etc will provide sufficient cleaner alternatives (at least in the first world).
o Computers will be integrated into nearly everything and communicate with each other in ad-hoc networks.
o While advanced nano-tech will still be a ways off, primitave nano-tech will be ubiquitous. Clothing dyes, paint and simple plastic structures will be able to change colors and shapes. Initial medical treatments using the tech will be available.
o Threat of large-scale world war will be practically null due to a level of economic interdependency never before seen. Most economic debates will revolve around world economy as a whole.
o People will travel less, telecommute more.
o Most people will still reject implants.
o Life expectancy will not increase drastically from what it is today.
o There will be an archival crisis in which data stored on digital medium begins to get lost.
o Everything important will be on-line.
o The Internet will fracture into at least 3 separate independent networks.
Those are mine, anyway ;)
Currently we try to pack everything onto a common dumb carrier (dumb carriers are good). We do secure transfers over SSL, companies have WAN-connected VPNs, etc. I just don't see this as continuing.
Already some analysts wonder if Google is positioning themselves off the net for back-end operations. Sure, many companies have dedicated circuits between sites, but Google almost seems to be building an entire private network. It doesn't need to stay private.
So the first fracture may be a network designed for running corporate back-ends. Having this completely separate of the main Internet would protect from failure far more robustly than simply having multiple backbones. Entry points would be closely monitored so it would be very difficult for hackers to gain access to it (they'd need to compromise one of the company's public-facing points first). A level of monitoring not possible or ethical on the public Internet would be available here.
The second fracture would be a split of latency vs. throughput. Take a game server, for example, that requires a ton of small packets quickly. Compare that to a large torrent download which has large packets that can transfer slowly. It's like sending data over a laser or packing hundreds of tapes onto a truck. You just can't have a one-size-fits all here. QoS and other means help on a network you entirely own, but impossible when bouncing between different providers. I think it's entirely probably that a gaming-specific network will be created for low-bandwidth low-latency applications. The public Internet will probably slowly morph into being a high-bandwidth high-latency network, which is ideal for web pages, video, torrents, other p2p, email, etc.
The third fracture would be a split between multicast and unicast. Multicast was basically a terrible failure because it tried to create a multicast backbone virtually tunneled over public links. Yeah. It sucked.
Unicast traffic is how almost all network traffic works. Basically, machine A sends a file to machine B. So, if you're YouTube.com and are serving a popular 10MB video to 10,000 clients, you're basically sending 10MB 10,000 times. Yeah, you can use akamai or other caching services to help, but that simply means that they're sending the data instead of you.
Compare that to how most broadcast media works. A single signal is sent, where it's repeated as many times as needed. That's the idea of multicast. Say you have three zones--East, West and South. Youtube would then send a maximum of three different files--one to each zone. This would then be split as many times as needed at each router (which would have their own zones) until it reached the user's ISP. Of course, data wouldn't be sent to zones in which there is nobody requesting the file.
Illegal activity will always simply tunnel over existing networks. You'll never have a "dark" net, as sexy as that would be ;)
People would simply never accept giving up control. And pretty much everyone has to give up control for a system to work. Making it something that could slowly be phased in would be tenfold more difficult.
And, like I said, a single death caused by a software bug would halt the project in its tracks for years. The project would be stalled an ineffective for decades until people accepted that deaths caused by software errors are inevitable, but at least less than human error.
Self-driving cars are not only an engineering challenge. They're a social one, as well.
Nope, it was a dirt off road...road. Only thing they had were sensors on the car and waypoints on a GPS (given an hour before the race). It was mostly a natural track and they started at like 15 minute intervals.
If you want I could try to find the website about it, it has movies tracking the entire thing with all the separate teams building their cars and the whole race.