ISPs pushing bogus concepts like the exaflood, with the goal of hoisting artificial scarcity upon the market.

That has reduced the financial incentive for carriers to invest in new cables, but they may have to do so by 2014, said Telegeography analyst Erik Kreifeldt.

"The market prices for capacity on the trans-Atlantic routes today don't support the business case for building a new cable system," Kreifeldt said. The coming shift in the industry from oversupply to scarcity is likely to rebalance the economics of cables, raising the prices that service providers pay for capacity for the first time in years, he said.

However, demand is likely to rise by 33 percent per year between 2008 and 2015, Telegeography predicts. That will eat away the over-capacity by 2014, the company said. Before that happens, someone will have to lay new cable or risk not being able to accommodate more traffic.

The problem is that the price of trans-Atlantic bandwidth today only covers the incremental cost of the optical gear on either end of the cable to provision the circuit, according to Telegeography. It doesn't cover the cost of actually laying the fiber, because no one has had to do that for so long.

"You're not going to be swapping that stuff out," Kreifeldt said. Submarine gear that was designed to handle 2.5G bits per second per wavelength has been upgraded for 10G bps links, but it won't be able to support the coming 100G bps technology, he said. Even the 40G bps technology being used today on land is unproven on submarine cables.

said by Karl Bode :

The "infinite bandwidth" argument isn't entirely unlike those who believed (including News Corp's Rupert Murdoch) that the price of voice services would ultimately declining to zero, as software applications made voice communications simply another form of data.

said by funchords :

Now if we could get Verizon to quit blocking port 80 and quit errorvertising, I might be a bigger fan...

It actually makes sense and is pretty unbias (vs purely position or "consumer focused")

said by TKJunkMail :

Seems there ARE limits to ever increasing bandwidth:
»tech.yahoo.com/news/pcworld/2009···edby2014

said by Karl Bode :

On the wireless voice front, control over applications allows carriers to prevent Skype from eroding voice revenues.

said by Karl Bode :

On the data front, high overages can be used to protect TV revenues and manage content competition. (Time Warner Cable protecting TV revenues by imposing 5GB caps and $2 per GB overages)

said by Karl Bode :

If 0.1% of the customers are causing the problem (their own numbers), you put them on business tiers or implement a high cap that catches them in a net. You don't implement an entirely new pricing model that will impact a significant majority of your customers down the road. Particularly in a market that you've bred to be reliant on the simplicity of flat-rate pricing.

Even if they didn't restrict Skype, you've got free applications like Fring that only a few people are able to make work successfully. Wireless data has improved significantly in the last decade, and yet high latency and jitter is still too variable in EVDO / GPRS / EDGE / whatever to work with most of the codecs that these clients use by default. Right now the biggest thing holding back the adoption of the technology is the functionality of the technology itself.

Your argument works a lot better by taking the absolutely lowest tier offered to make your case.

I'm sure at the corporate level there is a little bit of concern about the loss of some video business, but it's about as rational as an elephant being afraid of a mouse.

said by Karl Bode :

Early tests with Skype indicated it worked just fine over HSDPA. I've seen no indication that the technology isn't there, but I've seen plenty of evidence of carriers doing their best to keep mobile voice applications from making voice minutes less relevant by restricting consumer access to such applications on handsets.

said by Karl Bode :

100GB is more reasonable, but the intent is the same, and the overage costs ($1.50 to $2 on TOP of a fairly large monthly fee still aren't reasonable based on carrier costs). Metering bandwidth in turn meters content consumption, preventing life down the road as an unlimited dumb pipe where other companies innovate content and services.

said by Karl Bode :

Again, afraid I disagree when looking at a 20 year window. Time Warner Cable's efforts clearly illustrate they're worried. And while Internet video is only just nipping at the heels of TV, things will change over the next decade. No, TV will remain dominant for many years because of certain inherent advantages (live events), but the fear of lost TV revenue is valid. Especially as speeds increase and home media distribution solutions get simpler/cheaper.

TV viewership still outweighs the entire usage population of the Internet by about 100 million, which will change. Revenue from Web-to-TV streaming services will grow to $2.9 billion by 2013. These execs are crunching the numbers of future TV defections and the metered billing push is driven by that. To deny this motivation denies some very obvious realities IMO...