New Fiber Optics In The Works

Logic Bomb writes: "An article from MIT's Technology Review has the details on a new kind of fiber optic cabling that could provide part of the backbone bandwidth increase everyone is looking for. Instead of sending the light through glass, the light is actually sent through nothing but air. The key is a tube lining made of a special class of materials called "photonic-band-gap" which manage to perform an almost-perfect reflection of particular wavelengths of light. I wonder if it'll be cheap enough for home use. :-)"

Comments

Thoughts on fibers

[maggard]

The article isn't clear on how these hollow-core fibers handle multiple frequencies of light (rather it implies different answers at different points.) It appears that (using their trite analogies) one might be trading in a several-hundred lane highway with regular tollbooths for a single lane without need for a tollbooth. Presumably this would be a boon for some applications but it's not a universal revolution.

Indeed with many of the increases in fiber-bandwidth having come from multiple frequencies of light & with greatly improved hardware soon to roll-out ('tunable' lasers & all-optical switches, some using light-frequency as a routing determinant) if these new fibers are truly limited in their frequency-transmisson they could find themselves hobbled when they eventually come to market.

I also wonder about splicing these cables, terminating them, etc. The difficulties of a single fiber were surmounted but with a number of wave-guides closely bonded together I imagine most present technology wouldn't work.

Those concens aside I can see a number of applications where a long-distance non-repeated cable could be of enormous use, particularly in under-sea cables.

Back to the when-can-we-see-this-in-our-homes I doubt we will ever as this particular technology seems unsuited for such an application. If the question were about fiber-in-general expect it to become possible in a few years.

Plastic-based fiber is proving to be cheaper & more versitile then glass based in the sort of mid/high density generally assumed for residential and now the sticking point is the connections & switching. Once cheap optical switches come onto the market it'll just be a matter of physical installation - presumably in about the same pattern cable-TV has used.

If you can get cable-TV now hopefully in about a deacade you'll begin having the option of fiber.

Imagine a Bewulf cluster of these... - sorry, couldn't resist.

As always,

[photozz]

The goal, .. of course,.. is for faster porn.... Anyone got a guess on the availability? The fiber I mean,.. not the porn.....

problem is...

[deran9ed]

Fiber is expensive as hell and many companies like QWEST already have existing dark fiber all over the place. Many corporations have yet taken measures to move unto those strands of unused fibers, and it would cost many companies an arm and leg to replace their cabling, especially when they haven't even used it yet.

Sounds great. So, where's the catch? It's a matter of limits. As communications networks get bigger, busier and more ambitious, the drawbacks of conventional glass fibers are becoming evident, and existing optical-fiber networks will eventually be unable to cope.

This is looking way into the future. Has anyone here actually upgraded to a fiber ethernet based network, or is everyone hoping. In reality its again a very expensive thing to do, cheap to think about, but expensive to do.

With companies like PSInet which is a big ass ISP coming near the brinks of bankruptcy, many companies are in a rush to SAVE money not run out to buy more equipment, upgrade, etc. I would like to see networks get faster, but is it a complete neccessity at this point? ... The answer is sadly no, and who knows by the time this is even feasible, with the way technology changes, there's bound to be something even faster by the time this becomes something close to a standard. Kudos to the MIT people though ;)

Home use?

[sllort]

"I wonder if it'll be cheap enough for home use. :-)"

Considering you can put 100Gbps through 400kilomters on one strand of existing optical fiber, you're gonna have a completely fucked up home if you need more bandwidth than that.

That said, the article claims that this will "revolutionize the telecommunications industry" because it allows for longer-haul fibers without inline optical amplifiers.

That might be true, if we were using the existing fiber we have. But look at the people selling low-power in-line optical amplifiers - namely Corvis. Nobody's buying their shit. We have millions of miles of "dark fiber" in America - fiber that no one is leasing. In addition, no one is using the "long haul" capability provided by the new generation of companies such as Corvis - mainly because policing these long fibers for a break is expensive, in addition to the fact that in a store-and-forward network topology (like IP) you have to route at each hop, so there's no reason to go that far.

The only successful applications of long-haul fiberoptic technologies so far have been underwater trans-oceanic lines. and this technology may help with that. But revolutionize the telecommunications industry? FUD.

What would revolutionize the telco industry would be if Corporate America actually had applications they wanted to buy bandwidth for, and started doing it. Look at all the solid equipment providers with tanked stocks: lucent, cisco - the bandwidth explosion hasn't happened.

sigh. fud.