Look ma, I shrunk the light beams: Waveguides are hollow tubes that help channel (light) waves and typically the width of the waveguide needs to be similar in magnitude to the wavelength of the guided wave. This has imposed limitations on the size of the optical devices. Now Caltech researchers, co-led by assistant professor of electrical engineering Hyuck Choo, have figured out a way to get around this natural limit. They’ve come up with a way to make light beams smaller and smaller by developing a new kind of waveguide device that can focus light to a few nanometers. What that means is that we can create even smaller optical components, which in turn can be used inside networking gear, data center equipment and even inside imaging devices. It also mean we can use current optical fibers to cram even more data inside them.

The waveguide device is made of amorphous silicon dioxide (much like common glass) and is covered in a thin layer of gold. As light passes through the device, photons interact with electrons at the interface between gold and silicon dioxide. The electrons oscillate and those oscillations propagate along as waves, carrying the same information as the light waves. The new device is built using traditional chip making technologies, thus making it easier to produce and bring to the market.

Move over Silicon, make way for new chip materials. Silicon might be the building block of our digital life, but researchers at MIT believe that it is time to pay attention to new materials. They have come up with what they describe as the smallest indium gallium arsenide transistor, which is 22 nanometers (a billionth of a meter) in length.

Indium gallium arsenide has already found use in fiber-optic communications and radar technologies, but the new small size transistor is aimed at computing devices and replacing silicon, according to Jesús del Alamo, the Donner Professor of Science in MIT’s Department of Electrical Engineering and Computer Science (EECS), who co-developed it with EECS graduate student Jianqian Lin and Dimitri Antoniadis, the Ray and Maria Stata Professor of Electrical Engineering.

Why is this important? Well, MIT researchers argue that the since silicon transistors are getting harder and harder to shrink, the amount of current that can be produced by the devices is also shrinking, limiting their speed of operation. So, in order for Moore’s Law to go on, del Alamo points out that alternatives to silicon are needed. By building a tiny transistor using Indium Gallium Arsenide, they claim to have a viable contender.

Optics & Silicon get hitched: This marriage of optical and silicon is an area of immense focus, and IBM is spearheading the movement. IBM has come up with new chips that combine traditional silicon based electronic parts with optical components that will lead to much faster and beefier networks.  Stacey has all the details.

[Bonus link] Fibers secured: Toshiba and Cambridge University scientists have come up with a way to boost the security of data over fiber optic cables, the New York Times reports. The fiber optic networks are prone to eavesdropping, but these researchers have come up with ways to make the networks more secure. Read more at the New York Times’ website.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The 2013 task management tools market
• The state of cross-platform media measurement
• Listening platforms: finding the value in social media data

However, it looks like those issues may finally be resolved with Google paying the same attachment fees that cable and telecommunications companies pay utilities to use their poles, as opposed to paying additional costs associated with stringing cable higher up on the pole where electricity cables hang. A Google spokeswoman confirmed that the fiber was getting strung, but couldn’t tell me what percentage of the network would be underground as opposed to aerial. She said via email, “We don’t have a percentage estimate but we’re starting on utility poles and down the line, the fiber-to-the-home connections will be terrestrial.” Fiber is far cheaper to deploy aerially than planted in the ground because it avoids the labor costs associated with digging trenches.

Google has said it would start signing up customers by the end of 2011 and that it would begin connecting customers in “early 2012.” If the company has just begun laying its fiber, that may push things back a bit. The project has experienced a few minor delays, but so far seems to be moving ahead. Rather than when, I’m actually far more curious about what Google’s network architecture will be and what it will cost.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Report: Monetizing Digital Content
• The 2013 task management tools market
• How consumer media will change in 2013

I spent some time talking to Dane Jasper, the CEO of Sonic.net about how he plans to take on the incumbents, the cost of a fiber build out and the natives of a city that already have sued to stop AT&T from building out fiber to the node broadband. We also touched on caps and why Jasper can’t see himself offering businesses gigabit services in the near future.

They may be rivals but Sonic.net and AT&T share a common enemy.

When it comes to building out infrastructure, from broadband to roads, someone, be it environmentalists or neighbors leery of the project’s components, are bound to raise a fuss. When it comes to better broadband, the cabinets holding the electronics raise the ire of residents who would rather not have refrigerator-sized boxes on their lawns. For example, residents of San Francisco have banded together to sue to stop AT&T’s planned U-verse deployment, which requires more than 700 cabinets to hold the electronics gear be placed around the city.

Jasper says because Sonic.net is deploying fiber to the home, he will use fewer cabinets (he estimates 188) but he’s still worried that San Franciscans will step up to hold up or halt his permits. AT&T originally had received its permits, but those permits were halted by the court while this suit goes forward.

Jasper is worried that the suit could take another three to six months, and will hold up his deployment, but he’s hoping that fewer cabinets and a willingness to share Sonic.net’s infrastructure with other providers might make city residents view his cabinets with a bit more favor. After all, instead of building new cabinets, competitors interested in the market could share space in the existing boxes. Jasper understands that cabinets aren’t ideal, but he’s also hoping that if he plays it straight with the city, he can convince them that fiber-to-the-home is worth the potential of a couple hundred eyesores.

The Economics of broadband and fiber to the home.

The Sonic.net employees. Look for them in your neighborhood, San Francisco.

Meanwhile, Jasper declined to disclose his costs for providing fiber to the home in San Francisco, but he did say the fiber deployed would be a mix of aerial and underground cabling. Aerial deployments are cheaper because there’s less labor associated with stringing the cable. Jasper said he hasn’t chosen vendors yet, but he is currently using ADTRAN, Clearfield and Corning in the fiber-to-the-home build out he’s building in Sebastopol, Calif. Sonic.net is profitable as a company, and has been in business for 17 years.

The question is if Jasper can keep Sonic.net in the black while building out and selling fiber to the home to consumers for $70 a month. Sonic.net today offers two products in most markets, a $40 ADSL service with one phone line and a $70 a month 40 Mbps bonded DSL service with two phone lines. Replacing the copper with fiber adds costs, but Jasper plans to keep the rate the same. Sonic.net’s well known for declining to cap its broadband service and for adding a variety of services to its bundle without charging more. Jasper says, “fundamentally we recognize that as a competitive service provider we need massively differentiated products and we have done that with our fusion copper products and uncapped service.”

But, as Jasper says, “we recognize that copper is not a long term solution ten to 20 years from now and it’s logical to build fiber out.” So while there’s a bit of a chicken-and-egg situation to building out fiber in terms of the customer demand not necessarily being there right away, Jasper believes that the demand will come and he can deliver the capacity and afford the build out. Jasper is using ist Sabastopol buildout to help model the costs and demand for fiber in San Francisco.

Why businesses won’t get Sonic.net’s Fusion products anytime soon.

No caps and no meters for Sonic.Net

To help keep costs in line, the fiber links are for consumer accounts only, at least so far. Jasper has been an ardent foe of broadband caps, where ISPs place a limit on the amount of data a customer can use each month. However, when it comes to delivering broadband to businesses, he recognizes that a superfast gigabit connection to a business will have a very different usage pattern than one delivered to a consumer. Yet currently Sonic.net only charges businesses a bit more than residential services at $45 and $90 respectively). Under a gigabit network, that lack of price differential and the possibility for a business to use all of their connection (or even half) becomes unsustainable.

“We haven’t built our fiber past any businesses yet, and we did it intentionally,” Jasper said. “With our stance on no capping, I have a little bit of concern delivering 1 gig to a business at $89.95 and them using half of it, because that could really happen.”

Sonic.net has a decade and a half modeling usage for consumers at lower prices than rivals offer, but with businesses and their demand for broadband, Jasper says there are a lot of unknowns. For example, the lack of applications for gigabit networks probably helps Jasper here, as does the fact that most consumers typically use downlink services to consume content. And currently there’s a limit to how much they can consume, even with three or four TVs downloading or streaming HD content.

“Consumption is still constrained by the number of TVs and hard drives and even though everyone eventually has more stuff, practically speaking it really does end up normalizing down to a reasonable level,” Jasper says. He points out that the inbound bandwidth costs and middle mile bandwidth costs are getting less and less expensive, which means that customers downloading content isn’t a giant cost suck. But a business might hook a data center or several servers up on a gigabit connection and use that to send a lot of traffic out. And that could get expensive.

So for those watching U.S. broadband policy, between Google’s plans to deploy fiber to the home in both Kansas Cities, a few municipal networks, Verizon’s FiOS network and Sonic.net’s plans, we’re getting more people to a gigabit. It can be done, so let’s see what we can learn as these companies push ahead. And when others say it can’t be done, perhaps we’ll have the information that proves them wrong.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The future of Wi-Fi in the enterprise
• Netflix may suffer from limited mobility
• 12 tech leaders’ resolutions for 2012