FiOS, FTTN as Marketing Platforms

BT plans to launch its 24 Mbps service in April 2008 with available coverage reaching in excess of 50 percent of the U.K. market by April 2009. It will be using a fiber-to-node network very similar to at&t's FTTN network, with copper drops. Both BT and at&t believe that is sufficient bandwidth for anticipated customer demand. In at&t's case that also includes IPTV and HDTV services.

They might be right. But there is something more than bandwidth at issue here, and that is the marketing platform. If you have Millenial children, ask yourself what their preferences are in the area of broadband and video entertainment providers (You know they all rely on their mobiles).

Up to this point, though, it has been the common pattern to buy both video and broadband from the cable company. What we need to watch is what happens when services such as Verizon's FiOS fiber to home service are available. The issue is not just how much bandwidth they need or will pay for.

The issue is whether FiOS or fiber to home services are a more compelling product than cable modem services.

TowerStream: 8 Mbps for $1,000 a Month

Wireless has been the perennial favorite for believers in facilities-based access competition to the entrenched telephone and cable companies. Some 25 years ago, proponents argued that Multichannel Multipoint Distribution Systems (MMDS) based on 2.5 GHz spectrum were going to be the way new video entertainment providers would gain a foothold.

That effort failed. Similar spectrum then was touted by the likes of Winstar, Teligent and others as a solution for high-speed access in the business market. The effort failed.

Much spectrum then was acquired by firms such as Sprint Nextel, BellSouth and MCI and spend years essentially languishing. Now Clearwire and Sprint say the former MMDS spectrum will be the foundation for WiMAX.

We shall see. A smaller new company, Towerstream Corp., is selling 8 Mbps broadband connections for $1,000 a month in eight markets, and currently plans to operate in 20 cities within two years.

In its Seattle market, starting February 1, new customers will be able to buy 3 Mbps connections bandwidth for $499 a month, with free installation. Towerstream offers businesses a range of bandwidth options including T1, T3, 100 and 1000 Mbps connections

The company has established networks in Los Angeles, Miami, Chicago, Seattle, the San Francisco Bay Area, and the greater Boston, Providence and Newport, R.I.

Using WiMAX technology, the company can “light" a city with just a few antennas. Its New York City network uses four antennas, including one on the Empire State Building.

TowerStream undercuts competitor prices for a T1 line by 50 percent or better. The small antennas that the company locates at the customers’ premises are installed by contract DISH or DirecTV installers.

Provisioning intervals normally are two or three days, compared with three to six weeks for a T1 line from a telephone company or competitive local exchange carrier.

Mid-band speeds in the 8 Mbps to 10 Mbps range seem to be the "sweet spot."

TowerStream appears to be using both telesales and direct sales approaches. It is said to have a 180-seat telemarketing center and is in the midst of expanding its sales force to 160 people, according to Morgan Joseph analysts, who say the company won 27 contracts in eight days, on the strength of 58 proposals. The company appears to have 100 or so direct sales reps trained and ready to call on prospects.

If history is any guide the company should enjoy at least modest success. By avoiding the mass market, it stays out of the way of 3G and other 4G networks aimed at consumers and small business. That's a strategy that lots of other wireless access providers also use.

So far, however, no single entity has managed to build a big business on the backs of fixed wireless broadband in the small business, medium-sized business or enterprise markets. And it may be that the path to success is precisely to operate as a niche provider, in high-density markets, without getting grandiose. That's typically where operators have stumbled in the past. But we'll have to watch and see.

In many cases the business case rests on prosaic concerns. LMDS operators found they had trouble getting access to rooftops once landlords decided they were sitting on a gold mine. It wasn't, but the incremental real estate access charges were enough to kill the business case.

Then there is the availability of riser and conduit space, access to it and the cost of new cabling. Assuming those sorts of issues can be managed, TowerStream might have a shot, at least in some markets, such as New York.

Bandwidth in the 8 Mbps to 10 Mbps range is a bit more than the 4 Mbps to 6 Mbps mid-band Ethernet service some other providers are finding attractive.

"Year of the..."

Be careful when anybody declares this or next year the "year of X." Such predictions inevitably are wrong. That doesn't mean the direction of a trend is wrong, just the timing.

So when Google CEO Eric Schmidt says the recreation of the PC and Internet stories are before us, he's right about the direction. When he says it is "very likely it will happen in the next year," he's most likely wrong about the timing.

The mobile Web will be a "huge revolution", as Schmidt argues. But it isn't going to reach the tipping point next year. Proclamations of the "year of the anything" are universally incorrect.

BlackBerry Consumer Push

Research In Motion's move into the retail consumer market, including lifestyle features such as television, music players, cameras and Facebook social-networking software, is a good thing for consumers. That that includes a goodly number of professionals and workers who use email a lot for work.

Obviously a consumer device has to be priced lower than a "business class" device. But one thing I do notice, as a "business" BlackBerry user, is that the keyboards being supplied on devices such as the Pearl and Curve have a distinctly unpleasant feel. RIM might be doing this on purpose, but the feel of the keyboard is as important to this user as the keyboard is on a PC.

Every other element of the experience is outweighed by this one fact. Again, RIM might be doing that on purpose, to differentiate the market segments each device appeals to. If so, it's working. The 8800 class of devices are the only ones with a tactile experience I can tolerate. That's one way to create differentiation of user experience, I will say.

The omission of cameras and so forth also are design features intended to make the 8800 appeal to enterprises. But sometimes it comes down to other simple features. Like the feel of a keyboard.

What Future for Downloading, Streaming Video?

The conventional wisdom now is that movie downloading will replace DVD sales and rentals, and that this replacement is only a matter of time. The conventional wisdom may well be correct, up to a point. On-demand viewing, in one form or another, has been increasingly for decades.

To some extent, the rise of the cable industry was an early and crude form of on-demand viewing, to the extent that viewers began to break away from the "three networks" experience, starting a process of audience fragmentation that continues today in much more diverse forms.

But movie downloading isn't the only future. In fact, the way new visual media are being used suggests that consumers are taking an "all of the above" approach to media.

People might continue to rent DVDs as well. But maybe not in the same way. "Unless video stores are reinvented, it may be that in five years, there are tens of thousands of kiosks, millions of online DVD renters and very few video stores," says Reed Hastings, Netflix CEO.

If you look at any sort of DVD media as an example of "sideloading," as people sideload music onto their iPods and MP3 players, you get the idea. People download songs. But they also may stream or sideload. And though one often thinks online delivery is the only viable business format, one can imagine other ways to do things.

Price, for example, might be one way to differentiate the market. Online or to-the-TV downloads or streaming will have a higher price point, with a more "immediate" delivery format. But mailed DVDs will have a much-lower price point, with less immediate delivery. But the point is that the delivery time might not matter.

On the Netflix unlimited three DVD plan, can have as many as three movies "checked out" at any one time. And if a person is busy enough, viewing of those movies only happens on weekends. So "immediate" availability isn't required. The three selections have to be available on the weekend.

Release windows still are a factor as well. If you want to see a movie, and missed it in the theaters, you can view it about a month to 45 days sooner than any "on demand" outlet has the content, if you watch on a DVD. On an unlimited rental plan, the cost of any viewing is arbitrary.

Cable and Internet VOD costs something on the order of $4.00 per movie, and the content has to be viewed within a certain period of time, sometimes within 24 hours.

Selection probably will be an issue as well. It is hard to imagine an equivalent lineup of online titles as the Netflix catalog represents, especially in the "long tail" area of niche content.

"Despite the growth of VOD over the last five years, DVD rental has been stable, with online rental and kiosk rental making up for store losses," says Hastings. "In the United States, DVD spending, including purchase, is still approximately 20 times larger than cable and Internet VoD combined, according to Adams Media Research."

Just about everybody thinks this will change, at some point. The issue is whether online delivery is the only choice, or whether other delivery methods still will remain a significant part of the mix. Price, release windows, immediacy, and depth of catalog suggest there is room for multiple consumption modes.

Mobile Search is Different

Google OneBox is an example of how search will change as mobile queries increase. In an enterprise setting, OneBox delivers real-time information from enterprise sources, such as CRM, ERP and business intelligence systems, based on a user's search query. In a consumer application, entering a movie title might yield a top result showing screening times for that movie at the closest theater.

The search algorithms have to anticipate what need a user has for a particular bit of information based in part on what device is used to make a query. In a mobile setting, it is a fair bet that a query for any type of product is related to some immediate need for using that product. Starbucks coffee, Italian food or bagel, perhaps.

So adapting applications such as search for a mobile use case requires more than adapting the display for a smaller screen, oriented in a different horizontal-vertical dimension and often with limited navigation tools and less bandwidth than a wired environment provides.

The reason for queries, as well as the types of queries, arguably are different in a mobile context. One is more likely to be querying a customer or inventory database in the office. One is more likely to be looking for someplace to eat when out of the office.

And then there's there the time of day, day of week dimension. People will be asking different questions on weekends than weekdays. They'll be seeking different answers after 5 p.m. or 7 p.m. than at 10 a.m.

All of which increases the value of locational knowledge and mapping. But you probably already had that figured out.

Video Delivery: Some Ways Better Than Others

At some point, as much more video starts to be delivered using IP networks, network marketers and engineers are going to have to come up with ways to entice people to use alternate means of delivery, when it is feasible to do so. At some point, it simply will not make sense to chew up valuable voice and interactive data bandwidth for relatively low value YouTube clips, as entertaining as they might be.

Consider for example what Qwest is doing: it has esentially decided to keep all traditional linear video programming, including high definition TV and on-demand programming intended for TV screen viewing, off its IP pipe. It is doing so by delivering linear TV in the most bandwidth efficient means possible, namely by satellite, streaming point-to-multipoint.

As would be the case for IP multicasting, the idea is simple" launch one single copy of each program to a virtually unlimited number of users who can view the stream at the same time or on a store-and-watch-later basis (TiVo or another digital video recorder).

That will reserve the IP connection for unicast video and other interactive applications. The same sort of "offloading" principle is used by Netflix with its "DVD in the mail" approach. The point is that we do not have to force everybody to use IP bandwidth for watching unicast video when multicasting, sideloading, satellite, physical media or some other approach, including time-shifted delivery, might work just as well.

The baleful alternatives will find service providers unable to meet customer demand for bandwidth because there no longer is any money to be made; a dramatic increase in monthly prices; or both. Consumers are smart. Given a reasonable set of different ways to get video, at discrete prices for different delivery times and media, they'll make choices that relieve pressure on access bandwidth bottlenecks.