I’m talking about what most people will have available in their homes.
I think that’s unanswerable for the most part.
- It’s very hard to predict technology.
- Define internet fastness… It’s different everywhere.
You could graph the speeds of the fastest available connection over the years. Maybe it would turn out something like Moore’s Law and you could predict by that.
I found this table at this site:
Based on the data given here there has been a 5,000-fold increase in speed from 1978 to 1999 (approximately 20 years as the OP asked). If this trend were to continue based only on these numbers we’d be looking at speeds of 7.5 Gbps per second in 20 years, but as aeropl pointed out, it’s difficult to predict technology.
The data presented are also skewed by the large leap in connection speeds listed between 1998 (56Kbps) and 1999 (1.5 Mbps). Before this the speed appears to more or less double every two years. At this pace, perhaps in 20 years the speed would increase a thousand-fold and give us speeds of about 1.0 to 1.5 Gbps per second, which seems more realistic.
Well, let’s speculate here.
“Speed” involves many aspects. Raw fiber speed of the main lines and distribution lines. Speeds of switches and routers, and their loading. Speed of nameservers. And so forth.
With respect to raw fiber bandwidth - it’s more limited by the final quarter-mile to the house, than the main trunks of the Net.
IIRC, the guy that does the Rapidly Changing Face of Computing newsletter (who has been amazingly correct over the years on predicting many things, especially hard drive sizes/price) has talked of the potential for multiple terabytes/sec coming into a home by 2025, delivering a broad mix of TV, VR, movies, radio, and other content - with actual speeds as-implemented in the average household being in the gigabytes/sec range. I’ll see if I can dig up the article that I read on that.
Yes, the speed of the household internet connection will continue to increase…but to a point. There is a limit of the speed of electricity for going through a wire of a certain length (covering a specific distance). Computer makers try to combat this by decreasing the size of the computer (in order to decrease the lengths of the wires) but there is a limit to this, too. If the space between wires and other machinery is too small the computer will overheat.
We could have much faster internet for our personal use. In Korea, they have VDSL which is 20 times faster than DSL. It’s supposed to be better than a T1 line, but I have yet to check it out for myself.
I’ve heard two theories for why we don’t have it in America yet
- customers don’t want it.
- companies want to make money on DSL first.
Don’t know which one is correct.
True, but that has nothing to do with the bandwidth of your internet connection. Data coming into your house from, say, boards.straightdope.com moves at the same speed whether you have a 56k or a T1. The difference is the amount of data that comes in at the same time (bandwidth, not bandspeed).
Are you referring to the public Internet or Internet2 ?
Internet2 moves at warp speed compared to the present-day public Internet.
Ava wrote:
How does that stop someone coming along and starting a new company using the faster technology. They wouldn’t give a hoot in hell about the old companies cost of invested technology.
My 2 cents worth. Not as fast as you might think. Once you get to the point where a page with several pictures opens instantly, or close enough anyway, how much faster do you need or even notice? For instance, if you’re running at 500Gig how much does 4 or 5 trillion really improve your life.
Just a WAG.
There is a limit to the speed which u can send using electrons (to the person talking about banwith vs banspeed, the only way to get more speed out of electrons would be to have multible send and receive lines, and once u start doing that there is no limit to the speed). However the feutrure of internet connections are going to be finber (optical) which allows MUCH greator speed then electrons. it’s faster and allows you to send using multible colors to expand the bandwidth with one fiber.
[ul] **NOT FAST ENOUGH. **[/ul]
Depends what you mean by “most people”. Most people are still on modems, connecting at 28.8 or 56K.
The problem isn’t technological, it’s social and economic–getting the fast connections to people who are willing to pay for them.
“social” should be “political”
We’ll always find something to eat up bandwidth. What about streaming television shows?
Neither.
VDSL can only operate over a copper wire for about 4,000 feet. Bringing VDSL to the masses will require phone companies to run fiber optic cable all the way up to peoples’ neighborhoods. Most cities currently lack this infrastructure. There are a lucky few in select areas in the U.S. who can get VDSL, but most of us will have to wait.
VDSL offers up to 52Mbps downstream and 16Mbps upstream. See this article for more information.
I think you’re wrong. About ten years ago I started using internet, and I was using a dial-up 9600 baud modem. At that point I could see absolutelly no need to increase the speed with more than a factor of two or so. 9600baud meant that I could have a complete refresh of my screen (80cols x 25rows x 10bits) in just over two seconds, without being clever. As long as I was using telnet / gopher / ftp, that was quite enough.
Then came WWW, with fancy pictures, etc, which at once made bandwidth a problem.
For browsing (todays) standard WWW-pages, it’s true that 1Gb/s is overkill, but I predict that within 15 years there will be new applications that use this bandwidth. (eg. video, virtual reality etc)
I predict that in 20 years computers will be twice as fast, and ten times as large.
I agree with Anthracite’s cite of terabit speeds by 2025. Already there’s a FTTH network spreading throughout Tokyo that advertises 100Mbs speeds and comes pretty close to living up to it, in my experience.
The only compelling rationale for speeds beyond current cable is for interactive media, narrowcasting entertainment media and video telephones. At some point it will be the norm for HDTV quality broadcast television/internet/telephone and video phone to be delivered by various appliances connected to the same fiber pipe.
You can look at technologies on the horizon now and get a pretty good idea where we will be 20 years from now. The problem is the pipeline and the carrying capacity of the fiber being installed. Digging up streets and re-laying high capacity fiber to each house is probably a lot more expensive than the backroom hardware necessary to implement gigabyte level ethernet. Ultimately I think the desire for HDTV and the ability to have almost completely immersive media experiences (including games) is going to be driving new fiber installations more than the ability to talk to your pizza delivery guy face to face.
4-5 trillion? Still not fast enough!