Tuesday, October 06, 2009
Nobel Prize honors fiberoptics
Woods Hole, my home town, has six scientific institutions and a year-round population of under 1000. As Gloucester is to fish, as Pittsburgh was to steel, as Palo Alto is to venture capital, so is Woods Hole to science.
As I was growing up, the fall spectator sport wasn't the World Series. It was the Nobel Prize. We'd root, root, root for the home team.
This year's Nobel Prize for Physics goes to Charles Kao, who I had the honor of hosting at Bell Labs when I worked there. Ironically, the two other scientists sharing this year's Physics prize, Willard Boyle and George Smith, were Bell Labs scientists, but I did not know them.
The Royal Swedish Academy of Sciences choice to honor fiber this year underscores how civilization-changing a technology it is. The whole spectrum, from DC to daylight, including all the frequencies that carry wireless communications, are replicated in each glass strand. A cable the width of a broomstick can hold thousands of strands. The leap from electronics to photonics will prove to be as profound as the leap from muscle power to mechanical power.
Wireless communications has its place; I'm confident we'll find the right mix of fiber and wireless communications technologies. But our grandchildren are likely to ask us why we didn't replace copper and coax cables faster. What will we tell them?
As I was growing up, the fall spectator sport wasn't the World Series. It was the Nobel Prize. We'd root, root, root for the home team.
This year's Nobel Prize for Physics goes to Charles Kao, who I had the honor of hosting at Bell Labs when I worked there. Ironically, the two other scientists sharing this year's Physics prize, Willard Boyle and George Smith, were Bell Labs scientists, but I did not know them.
The Royal Swedish Academy of Sciences choice to honor fiber this year underscores how civilization-changing a technology it is. The whole spectrum, from DC to daylight, including all the frequencies that carry wireless communications, are replicated in each glass strand. A cable the width of a broomstick can hold thousands of strands. The leap from electronics to photonics will prove to be as profound as the leap from muscle power to mechanical power.
Wireless communications has its place; I'm confident we'll find the right mix of fiber and wireless communications technologies. But our grandchildren are likely to ask us why we didn't replace copper and coax cables faster. What will we tell them?
Technorati Tags: fiberoptics, NobelPrize, WoodsHole
Comments:
You can tell them that the incumbents (AT+T, the RBOCs, Comcast, etc.) went for short term profits instead of long term gain. They stuck with copper cabling when they could have switched to hybrid (copper + fiber) cabling. The hybrid cabling would have made it easier to roll out Fiber-to-the-Cabinet/-Node (FttC / FttN).
One of the voids in network data is a comprehensive fiber inventory. How much fiber conduit is in place ? How full are the conduits ? Of the cables in place, how much is still dark ? How much of the lit fiber is using obsolete drivers ?
These questions may never be answered because the telcos (that includes Comcast) will fight to keep this data hidden. They'll probably claim that it's proprietary and would give too much data to their competition. But in most places in the U.S. the telcos are public utilities and are subject to regulation. And how can the regulators and the general public determine that the telcos are doing their job in terms of capital investment if they don't publish those numbers ?
Consider how AT+T has been vaguely complaining about network capacity in California. The articles I've seen imply that it's their data backbone, not their mobile phone backhaul. The problem is that if it is their TCP/IP network backbone then it may be their own fault. When Calif. was PacBell territory their ISP, PBI, announced a fiber backbone (OC-3 class IIRC) that tied the San Francisco region to the Los Angeles area. This was in the mid-to-late 1990s. Since then SBC swallowed PacBell and morphed into AT+T. How much of that backbone is lit ? How up to date are the drivers ? What are their upgrade plans for that backbone ?
I hope the upcoming overhaul and expansion of the U.S.'s rail network is used for the installation of co-located fiber trunks. This would be a win-win combination - railroad command and control, passenger wi-fi, and general network capacity expansion.
Post a Comment
One of the voids in network data is a comprehensive fiber inventory. How much fiber conduit is in place ? How full are the conduits ? Of the cables in place, how much is still dark ? How much of the lit fiber is using obsolete drivers ?
These questions may never be answered because the telcos (that includes Comcast) will fight to keep this data hidden. They'll probably claim that it's proprietary and would give too much data to their competition. But in most places in the U.S. the telcos are public utilities and are subject to regulation. And how can the regulators and the general public determine that the telcos are doing their job in terms of capital investment if they don't publish those numbers ?
Consider how AT+T has been vaguely complaining about network capacity in California. The articles I've seen imply that it's their data backbone, not their mobile phone backhaul. The problem is that if it is their TCP/IP network backbone then it may be their own fault. When Calif. was PacBell territory their ISP, PBI, announced a fiber backbone (OC-3 class IIRC) that tied the San Francisco region to the Los Angeles area. This was in the mid-to-late 1990s. Since then SBC swallowed PacBell and morphed into AT+T. How much of that backbone is lit ? How up to date are the drivers ? What are their upgrade plans for that backbone ?
I hope the upcoming overhaul and expansion of the U.S.'s rail network is used for the installation of co-located fiber trunks. This would be a win-win combination - railroad command and control, passenger wi-fi, and general network capacity expansion.
