I spent time at a conference last week where one of the prevailing themes was fragmentation. As much as we talk about convergence (of devices, services, applications), the truth is that end-user experiences are fragmenting. It used to be that many of us watched the same shows on the same types of TV sets, communicated primarily on landline phones, and had some kind of stereo system for listening to music at home. Now the communication and entertainment choices are endless, as are the ways we choose to access them.

From one perspective, this fragmentation means that technology has to be extremely adaptable. Service providers are trying to address a million and one different demands while also finding ways to stay cost efficient. So they’re turning to technology providers to make it possible. Here are just a few examples:

• The rise of HD is one of the factors leading to adoption of the MPEG-4 AVC compression scheme. However, it will take a good long while before every consumer has a set-top that can accept MPEG-4. The result: new receivers/transcoders that can use both MPEG-4 and MPEG-2 compression.
• Consumers want more content choice, but bandwidth is not limited. Enter switched digital video. With SDV, operators can switch on specific channels only when they are requested – delivering certain programs when they’re wanted, but saving the bandwidth when they’re not.
• Demand for higher broadband data rates to support Internet media has led to the introduction of DOCSIS 3.0 channel-bonding technology. But consumer modems won’t transition overnight. So technology in cable modem termination systems has to support not only DOCSIS 3.0, but also legacy DOCSIS 1.X and 2.0 modems.

There are many, many other examples, but the point is that technology is increasingly called on to be more flexible. Convergence=Fragmentation=Flexibility.

The cat’s out of the bag. Even though there’s been no official press release, CableLabs leaked and then confirmed that the Motorola MTR700 was one of two tuning adapters to be awarded certification in the recent Cert Wave 60. Quick recap: the tuning resolver is what will allow one-way cable devices to continue to access all cable channels even when some channels are moved to a switched digital network. Current plans are for the MTR700 to ship in the fourth quarter.

Interestingly, Comcast has been as enthusiastic about migrating to all-digital broadcasts recently as deploying switched digital video to save bandwidth. Another type of Motorola adapter, the DTA100, is on order from Comcast and will convert digital signals to analog for subscriber hold-outs still clinging to basic cable service.

Meanwhile, SDV work does continue to move forward. Cable Digital News recently reported that Comcast has added two more trial markets in Minnesota. Motorola has commitments in hand to pass more than 24 million households with SDV service.

Vyyo, loosely translated, means “air” in Hindi. And air is what the Norcross, Ga.-based cable broadband equipment maker with that name might have run out of. The company, which used to trade on Nasdaq under the ticker VYYO, is down to some $5.3 million in cash and cash equivalents, according to an SEC filing. The sad turn of events for the company comes a year after it received a big cash infusion — $35 million — from Goldman Sachs. It now has a market capitalization of around $3 million and trades on the Pink Sheets OTC market.

Vyyo makes spectrum-overlay technology that allows cable providers to give big boosts to both upstream and downstream bandwidth. Unfortunately, with the exception of a handful of buyers, the company couldn’t get its technologies widely deployed. And according to Cable Digital News, Vyyo has been fighting a losing battle with vendors that make switched digital video and 1 GHz bandwidth expansion gear.

A couple of savvy folks picked up on my near-throw-away reference earlier this month to the idea of using switched digital video as a tool for transitioning to MPEG-4. The basic concept is that operators could selectively broadcast MPEG-4 programming by offering it on a switched video tier. This would give them the ability to deliver higher-quality video and save bandwidth whenever MPEG-4 receivers (set-tops) were in use. The idea is tantalizing, but also much more complex than a two-sentence explanation suggests.

In digging through John Schlack’s technical paper on the subject, here are some of the details and clarifications I can offer.

The easiest way to make this work from a technical perspective would be to offer a premium service with additional MPEG-4 HD content. Subscribers to the service would get an MPEG-4 set-top and would be the only ones able to tune to switched channels with MPEG-4 programming. MPEG-2 set-top users simply wouldn’t see the channels as options.

Another way to deploy MPEG-4 content on a switched tier would be to implement forced tuning whenever an MPEG-2 subscriber requested a program being delivered with MPEG-4 encoding. With today’s equipment, this would require storing both an MPEG-2 and an MPEG-4 version of the program – not a problem when storage and streaming capacity are set up to be scaled separately. The system would deliver the MPEG-4 version when requested, and would force tune everyone to an MPEG-2 version if any MPEG-2 users requested the content. Force tuning has the potential to be somewhat disruptive, but there are solutions for mitigating that problem, like making sure both versions of a show are on the same QAM and/or waiting for a commercial break to force tune.

Theoretically, there should be no reason operators couldn’t also implement dynamic transcoding. In that instance, the switched digital video (SDV) system would dynamically transition between an MPEG-2 and an MPEG-4 stream, without the need to have two separate versions of a show queued. However, today’s decoders are not capable of dynamic transcoding, which means operators would have to spend money to deploy new decoders.

The end of John Schlack’s paper describes the benefits of using SDV for MPEG-4 delivery so succinctly that I’m going to quote it directly here. Any further questions? Drop me a line at marisilbey (at) comcast dot net.

Initially with a small number of MPEG‑4 capable settop boxes, SDV will be an enabling technology for deploying MPEG‑4 into the cable plant, allowing delivery of additional HD content and potential bandwidth savings. As the number of MPEG‑4 capable settop boxes grows beyond the deployed legacy settop box count, the cable plant will transition to broadcasting more MPEG‑4 content while delivering the MPEG‑2 content on the switched tier. This will provide further bandwidth savings.

The news is out that the Starz network is making its move toward MPEG-4. It won’t happen all at once, but the programmer is planning to use MPEG-4 compression with its HD channels starting this summer. Like HBO, Starz is using the Motorola DSR-6050 receiver/transcoder, which supports both MPEG-4 and MPEG-2 outputs.

Obviously there’s a benefit to programmers in moving to a more advanced compression scheme (better quality video), but it’s been widely discussed that cable operators can’t take advantage of the bandwidth savings that come with MPEG-4 until they have a large number of MPEG-4 set-tops in the field.

As it turns out, that’s not entirely true.

In theory, an entire service group of subscribers would need to have MPEG-4 set-tops in order for a cable operator to stream a channel in MPEG-4. But by using a switched digital video (SDV) platform, operators could selectively send out MPEG-4 streams only when the subscribers requesting that content had MPEG-4 set-tops. In other words, operators could recoup even more bandwidth over straight SDV by using MPEG-4 compression in a switched environment.

Motorola’s John Schlack presented a paper on this topic at The Cable Show. If anyone’s interested in more detail, drop me a line at marisilbey (at) comcast.net.

First the news: Motorola announced the launch of a new universal edge QAM today, the APEX1000. In brief, edge quadrature amplitude modulators (QAMs) are used to translate signals from IP into radio frequency (RF). (Leslie Ellis explains it by saying this modulation turns spectrum into bandwidth.) Operators are finding they need more QAM channels of late because of growing demand for switched digital video (SDV), on-demand video, broadcast TV and DOCSIS data services. The issue of QAMs has been covered a bit recently because there is a cost associated with deploying more QAM channels even though at the same time they help operators save money and bandwidth by enabling SDV. (I’m hoping to have a deeper cost/benefit analysis some time soon.)

Confused? Don’t worry. The important thing to understand today is that a universal edge QAM makes it possible to use QAM channels for whatever services operators need most. In a single APEX1000 there are 48 QAM channels available, all of which can be used for SDV, VOD, broadcast TV or data service delivery.

Meanwhile, the APEX news is interesting in conjunction with an article Jonathan Tombes has out today. Tombes’ write-up in Cable 360 questions whether the pendulum has started to swing away from switched digital video, one of the big drivers for more QAM channels. I’d say the answer is no, but of course that comes across a little biased given Motorola’s interest in selling both QAMs and SDV technology.

So consider this: everyone has focused on SDV as a direct bandwidth-saving tool, but it has the potential to do much more. If operators can selectively switch on a particular channel, they can also selectively deliver different content formats depending on the technical set-up of a given subscriber. Bottom line: SDV provides an awful lot of flexibility, and plenty of benefit to justify spending a bit more on QAMs – especially when those QAMs can be swapped over for different uses as needed.

There are some very complex equations here, but what they mean is that operators can create the systems that work best for them. All it takes is a little math.

A participant on the NCTA’s switched digital video (SDV) briefing call last week asked why the cable industry isn’t moving more quickly to all-digital broadcast given the bandwidth savings. Verizon’s making the shift, why not traditional cable operators? The answer is that cable companies as a whole have a lot more existing analog customers, and the transition is going to be both complicated and expensive. It makes sense. However, I heard an interesting contradictory anecdote yesterday. The folks in Motorola’s SDV lab talk to operator customers all the time about the logistics of architecting networks that are a mix of elements – a mix of digital and analog, a mix of HD and SD, and a mix of switched and broadcast video. Apparently one customer ran the numbers for his network and came to the conclusion that it would be cheaper for him to go all-digital, all-HD and all-switched now. No mixed environments, everything uniform.

First off, and not surprisingly, the cable operator in question is a relatively small one. Presumably the capital cost of making a complete transition is not nearly what it would be for a larger operator like Comcast, and thus is quickly outweighed by the operating expense of managing a network with a huge range of video outputs. (I’m making the assumption that capital costs grow more or less linearly with more subscribers while operating expenses level out more.)

Secondly, the implications of making a complete transition are fascinating and manifold. On the one hand consumers get better service, and the operator gets more bandwidth to offer more HD content and advanced services. On the other hand, what do subscribers pay for when the transition happens? Will a basic price tier still be available given the level of service being offered?

It will be interesting to watch as different cable operators wrestle with how to make their own transitions.

The NCTA conducted an interesting form of outreach last week. After apparently reading a number of unflattering and possibly inaccurate reports on switched digital video’s impact on CableCARD retail devices, the association put out the word that it would conduct a conference call briefing for any bloggers and journalists interested in discussing the topic. The call happened on Friday and it was truly a party of industry trade folk. Among the people I know were on the call: EngadgetHD, Cable Digital News, Multichannel News, IP and Democracy, TV Technology, Envisioneering, HD Guru, Gizmo Lovers, and Zatz Not Funny. And other than someone who treated us all to loud hold music (participants’ phones were not muted), everyone behaved pretty well.

Digression aside, there wasn’t much in the way of news on the call, but I still applaud the NCTA for holding an open forum. More interesting to me than the NCTA’s presentation were the questions that came after it. Here’s a sample (paraphrased) with the NCTA’s answers (also paraphrased). The entire recorded conference call is also available. Call 1-800-475-6701 and enter access code 920821.

Sampling of Q&As from the 4/25 NCTA conference call:

Q: Given the issue of having no 2-way communications in today’s retail CableCARD devices, why doesn’t the cable industry support DCR+?
A: The same problem exists with DCR+ as with one-way CableCARD devices in that consumers don’t get access to all the cable services they expect. DCR+, for example, won’t receive certain existing services like Time Warner’s StartOver and LookBack, not to mention future applications.

Q: What assurance can you give that Tru2way won’t be superseded by something else in three or four years, making consumer hardware obsolete?
A: Cable operators are putting Tru2way in their own devices so they’re not likely render Tru2way obsolete when they have so many set-tops with it in the field.

Q: Given the issues with switched digital video, why doesn’t the cable industry just move to all-digital like Verizon to reclaim bandwidth?
A: The cable is industry is much bigger than Verizon with a lot more existing analog customers. The transition is going to be much more difficult.

Q: PC companies and portable media player companies have difficulty with the CableLabs certification process, particularly when their headquarters are in places like Shanghai, rather than Boulder, Colorado. How do companies figure out how to work with CableLabs?
A: The CableLabs licensing and certification process is open to anyone who wants to participate, and CableLabs wants more companies to get involved. With all the innovation and energy from some of these CE companies, it seems it would be possible for them to make contact with CableLabs without getting 150 people in a room for a CableLabs event in Shanghai. Several cable industry CEOs have even gone to Japan and Korea to reach out to the Asian CE market. It could be a two-way street.

Q: Will there be any kind of requirement for operators to tell subscribers which channels will be on a switched network?
A: Yes, generally operators have to give thirty-day notice when there’s any kind of programming change.

I’ve gotten more than few friendly nudges lately for an update on the switched digital video (SDV) tuning resolver. The Motorola device is now officially called the MTR700 and it just came through a CableLabs interop with flying colors. Next the product will be submitted to CableLabs’ Cert Wave 60, and then it will be on display (functional and static demos) at the Cable Show in New Orleans.

Hard to believe a cable product could go through testing this quickly, but consumer CableCARD devices have made the technology a priority.

A few points of context:

• Cable operators don’t want to drive subscribers away because they can’t access certain channels on their retail CableCARD devices, hence the reason the industry has moved so quickly on the tuning resolver
• Operators have significant SDV plans for 2008
• SDV is a good thing! It means more capacity for, among other things, high-def TV channels

EngadgetHD ran a poll this weekend on how many HD channels readers have at home. Keeping in mind that the EngadgetHD audience is geekier than your average room full of people, I was surprised by the numbers as of Sunday night. More than half of the respondents acknowledged having fewer than 20 HD channels. That means there’s an awful lot of room for growth, and a lot more bandwidth needed to broadcast the volumes of HD that consumers are starting to demand. A good reason for switched digital video.

Favorite links from a holiday weekend of reading:

Broadband Reports – $2 billion from Intel may bring national WiMAX network

Light Reading - Comcast aiming to implement SDV in 15% of network this year

EngadgetHD - Latest research confirms it – DVRs increase TV consumption

Ars Technica - Retail sales staff don’t know much about the DTV transition

Get ready for a slew of DTV transition posts as we approach the one-year countdown. (I’ll have my own up soon enough.) In the meantime, however, there’s word out of Cox that the operator will continue offering analog signals until February 2012, three years after the official transition date in 2009. That’s good news for analog customers, but what does it mean for Cox cable?

Here’s a look at some of the pros and cons of the decision.

Pros:

• Cox doesn’t have to spend money on DTV converter boxes for all of its analog customers
• There’s no need for a large wave of truck rolls next January
• There won’t be a mass of customer complaints from folks wondering why TV doesn’t just work like it used to

Cons:

• Three-year delay on reclaiming bandwidth from analog channels (somewhat mitigated by Cox’s move to 1 GHz)
• No switched digital video for analog
• Maintains the hurdle of migrating customers to digital before they can subscribe to premium services like on-demand video

As we settle comfortably into 2008, switched digital video (SDV) technology is hitting its stride. First came news of the tuning resolver, a device designed (by Motorola and others) to allow retail products with one-way CableCARDs to access two-way, switched digital services. Light Reading reports that CableLabs officially issued specs for the tuning resolver on January 30^th.

Now Motorola is announcing successful testing of its SDV platform in a multi-vendor environment. This is critical because, like with any other network technology, operators will not rip and replace equipment unless absolutely necessary. At the CableLabs winter conference Motorola showed its SDV solution with multiple electronic program guides and multiple QAM devices.

I do realize this all sounds terribly boring to anyone not making a living off this stuff, but there are a few reasons to care. First, switched digital video means more HD content. Second, SDV is a cost-effective way to add bandwidth to a network, which hopefully means less price escalation for consumers. Third, switched digital video moves us one step closer to a switched unicast television model where everyone gets a personal, and personalized, video stream.

Much more to come on switched digital video in the coming months.

Edge QAMs have been getting a fair bit of industry attention recently, not least of all because of the recent Kagan conference, “QAM Before the Storm”. (Still my favorite name for an event ever…) Some at the Kagan conference were predicting the near-term end of edge QAM devices given the coming shift from RF technology to IP transport over fiber. However, when I queried folks within Motorola, the general consensus was that this prediction is a bit over-the-top. One person agreed that QAM modulators will move to fiber nodes over time, but that we have at least one more cycle for headend-based modulators before that happens.

Another executive went even further suggesting that the transition could take longer than most people expect. “The consensus in the late 1970s was that broadcast television was dead. It’s now 2008, and last I knew, 20% of people still get their TV over-the-air.”

Meanwhile, in the short term, use of QAM devices will only increase as we move to more on-demand content, switched digital video and, in a few years, a switched unicast model. (See above) I’ve talked about switched unicast before, and the potential benefits of the model are high enough to warrant further investments in edge QAMs near term. Specifically there is huge revenue potential from targeted advertising once switched unicast is in place.

L to R: Fred Wright (Motorola) Geoff Roman (Motorola) Craig Cuttner (HBO) John Burke (Motorola)

I promised to write more on the Motorola set-top announcement with BT, and after just sitting through a panel session with several Motorola executives and Dan Marks, the CEO of BT Vision, I feel more than equipped to provide additional context.

Dan Marks started his part of the talk by discussing the importance of individual addressability in set-tops, and he gave a great analogy. Operators have gone from selling TV content to building a virtual retail store for the digital experience. Because of that, the abstraction of traditional television ratings is no longer enough. Operators need to know not just what programs are popular, but what happens when consumers “walk down the aisle.” In other words, how, when and where they’re watching TV.

This is where addressability – or true two-way communications – with set-tops comes into play. The long-term goal and opportunity (using set-tops like the Motorola IP set-tops BT will deploy) is targeted ad insertion. If you know how, when and where people are watching TV, you can sell to them more effectively. In the short-term, though, even just the subscriber data alone is valuable. For example, knowing how much bandwidth is being used for switched digital channels means operators can allocate overall bandwidth as necessary (virtually in real time) to meet subscriber demand.

The one issue, however, that we’ve yet to see play out is the impact of consumer privacy concerns. James McQuivey, the Forrester analyst who moderated the panel session, believes that consumers will make the same compromises between privacy and convenience that we all make every day. (For example, handing a credit card over to a waiter) There will be stutter steps as privacy issues are dealt with, but ultimately privacy concerns will not stop a trend of data collection (and the application of that data) that’s already begun. And that is probably a good thing – as long as data remains disassociated from individual identities.

One last note on this topic: I asked the panel how much service providers are concerned about the privacy issues related to two-way communications between individual set-tops and the operators distributing to those devices. Not surprisingly, the answer is operators are at many different points along a spectrum. Some are very concerned about a privacy backlash, and some believe that if they can see the technology works the way it’s supposed to, they’ll be able to sell it to content providers given the revenue potential of targeted ad insertion.

More to come later on this panel session on the topic of networking video in the home.

Last week’s conversation with Geoff Roman was good for more than the requisite prognostications that fill up blogs and news stories this time of year. The discussion also included the most succinct overview of Motorola’s acquisition strategy that I’ve heard yet. I’m referring specifically to the acquisitions made in the last 24 months by the Home and Networks Mobility business, or as it was formerly known, Motorola Connected Home Solutions.

I’ve touched on the acquisition storyline before, but Geoff Roman did a much better job of explaining the strategy simply.

Prior to 2006, Motorola centered operations primarily on its strength in cable set-tops and cable video encryption (a technology product known as Mediacipher). Beginning in 2006, however, Motorola decided to go after a slate of acquisition targets that would help the company address a changing broadband and video market. Here are the acquisitions outlined by technology category.

• IPTV – The technology had gained traction in Europe and Asia when Motorola bought Kreatel in early 2006.
• MPEG-4 encoding – Motorola acquired both Tut Systems and Modulus this year to build up a strong position in carrier-based as well as cable-based MPEG-4 encoding.
• Transcoding, multiplexing and other video processing – These capabilities came on board with Terayon and also the part of Tut’s business focused on edge multiplexing.
• Switched digital video – Motorola acquired Vertasent in September 2006 for its SDV portfolio.
• ADSL and VDSL CPE – While Motorola cable modems are well-established in consumer homes, DSL equipment was not part of the company’s offerings until the acquisition of Netopia earlier this year. Netopia has also provided the nucleus for Motorola’s network management strategy.
• Convergence software – Motorola completed the acquisition of Leapstone – the “linchpin behind mobility” – a short four months ago. Leapstone’s service delivery platform is designed to move voice, video and data between wired and wireless networks.
• On-demand video – With the significant growth VOD has seen in the last year, Motorola acquired Broadbus at just the right time last fall.

Motorola has already made some significant returns on its investments, including the greatest revenues in telecom TV of any company, the milestone HBO customer win for MPEG-4 encoding, and massive commitments for switched digital video deployments. Not the kind of story you’ll see coming out of CES in two weeks, but a good one for wrapping up 2007.

Happy Holidays, everyone. There will more light posting here this week, and then back to the regular posting schedule next week leading up to on-site CES coverage.

I’ve watched other companies grab the limelight on switched digital video for a while, so it was with great delight that I found out we could go public with Motorola’s own news. Specifically, Motorola has commitments in hand to pass more than 24 million homes with its switched digital platform. About a quarter of those deployments will take place before the end of this year, with the rest coming 2008.

More on Motorola’s switched digital technology here. More on the one announced cable operator using Motorola’s platform here. More official announcements on operator customers coming soon.

Jeff Baumgartner over at Cable Digital News also covered Motorola in an SDV update yesterday.

CableLabs put out a release earlier this week on a new device to give retail CableCARD devices access to switched digital video services. And folks have been clamoring for details ever since. (Dave, Jeff, that means you…)

Here’s what I found out.

First, the cable industry is moving quickly. As much as consumers don’t want to lose access to switched content, cable operators don’t want to lose customers as a result of the move to a switched platform. Motorola started working with CableLabs in July on a technology solution, and as of two weeks ago in an Interop with TiVo, already had a prototype device ready and performing well. That’s a four-month turnaround on a new cable device. Virtually unheard of. Motorola has also committed to general availability of the product, called a tuning resolver, by the end of Q2 2008 – essentially six months from now.

As for the device itself, there are a couple of things worth understanding from a technical perspective. The tuning resolver is likely to come directly from cable operators, much like a CableCARD. Why? Because there are huge interoperability challenges being overcome. Any tuning resolver must work with an operator’s conditional access system, switched digital technology and electronic program guide. As mentioned above, Motorola has already done interop tests, but solving these interoperability issues was no mean feat, and every cable operator is going to have to have a solution tailored for its particular network architecture.

There are a lot of people asking what the tuning resolver will look like and what it will cost. The first part’s easy. Motorola’s tuning resolver will look an awful lot like the famous (in my book) and widely deployed DCT700. As for price, that I can’t really answer. A lot of it will be dependent on volume production.

Lastly, information leaked out yesterday that TiVo is now working with CableLabs on an OpenCable product. In theory this would negate the need for a tuning resolver in TiVo’s case. However, how quickly TiVo could move forward with an OCAP product is unclear, and there are still existing TiVo and other CableCARD devices in consumer homes now that won’t be able to access switched digital services without a tuning resolver.

Last, last note – the tuning resolver does not currently support VOD services, but I’m told it’s not an unreasonable assumption to believe that it might in the future.

So much going on, so little time to cover it all.

Switched Digital Video – CableLabs made it public this morning that several companies working together (including Motorola) have developed a solution to deliver switched digital video (SDV) services to retail CableCARD devices. With cable operators keen to roll out SDV in order to save bandwidth, this development is critical for consumers – who wouldn’t be able to access switched channels on retail CableCARD devices without it.

Hope to have more technical details soon, but in the meantime I want to make a point about Motorola supporting a technology that enables competitive set-tops in retail. With the long-term view in mind, opening up the cable network is good for Motorola. First, since Motorola powers cable networks, the more devices connected to those networks, the better. Second, with the goal of seamless mobility in mind, Motorola wants open networks so that more devices (including cell phones) can connect to each other.

Black Friday Bandwidth – This past weekend marked the official start of the holiday shopping season, both in stores and online. And as Broadband Reports suggested a while back, bandwidth may very well be this year’s holiday Grinch. The Raw Feed reports (via eWeek) that online traffic slowed some retail sites by as much as 400%. Obviously a lot of the issue here is at the retail end, but overall bandwidth availability is also a factor. I wonder how many people saw their Internet connections slow down this weekend during prime online shopping time.

The Killer App – There was a story up this morning out of the UK suggesting that lack of a “killer application” is stalling fiber deployments. How about lots of killer apps? Like video streaming on the Web, HDTV, video telephony, VOD…

From Multichannel News:

“Comcast has selected Motorola as a supplier for switched digital video, Motorola president and chief operating officer Greg Brown said on an earnings conference call with analysts Thursday.”

Resident expert on switched digital video (SDV), Bruce Bradley, gave a briefing last week to talk about the bandwidth implications of SDV, deployment timelines, and how the technology is likely to evolve. Here are my takeaways from the conversation:

On the bandwidth front, cost models are showing that SDV is the most efficient (read cheap) way to free up bandwidth in the 2007-2008 period. No wonder cable operators are so enthusiastic, and willing to find solutions to issues with one-way CableCARDs in retail devices. Operators can reclaim a projected 50% of bandwidth with the technology. That leaves a lot more room for more HDTV.

As far as deployment goes, 2008 is expected to be the big year. (No surprise there.) But, we’re talking about North America, not globally. By the time SDV starts rolling out in Asia and Europe, we may be into the next phase of the technology. Specifically switched unicast, which is designed to provide personalized video streams (and new revenue opportunities via targeted advertising). Switched unicast becomes feasible when the number of set-tops in a service group approaches a 1:1 ratio with the number of video streams being served. Today a lot of service groups contain around 1,000 users. To get closer to the 1:1 ratio, that number would have to be more like 250.

If you’re looking at a timeline of technologies, the order of events will probably go something like this: switched digital video, followed by SDV with a variable bit rate, then MPEG-4 (less immediately urgent for cable operators now), then switched unicast, and finally cable GPON deployments. Not everyone will follow the same timeline in every region, but SDV is the logical first step.

UPDATE #2: Please note, I did not intentionally pull this post. I corrected it, and apparently WordPress automatically made the post private at that point. Corrected post (strike-outs and all) below.

Nielsen ratings measure the most popular television programming. Preparing for switched digital video (SDV) requires exactly the opposite. No cable operator is going to move content to a switched system that is never going to be switched off. The best channels for SDV are the ones at the tail end of the popularity curve.

This has two a major implications. First, c Cable operators need a very accurate system for measuring which channels will give them the most bandwidth return when moved to a switched network. I was on a call last week where Motorola’s Bruce Bradley described the monitoring technology that Motorola uses with operators. It’s wild. The software actually shows a running feed of which channels consumers tune to in any given service group. (I’ll post a screen shot when I have one.) Operators run the software to determine which channels to switch initially, but also continue running the software so they can make changes dynamically based on subscriber viewing habits.

Second, the current kerfuffle going on about whether consumers with retail CableCARD devices will be able to access SDV channels seems to deflate a bit in light of what will actually be available in switched form. I don’t want to minimize the issue at all, and I certainly hope the NCTA and the CEA can come to an agreement soon that will get two-way cable services to retail devices. However, the most popular TV programming will never move to a switched network. Even if you don’t have a set-top from your cable company, you’ll never lose Lost (or American Idol or NFL football) with switched digital video.

UPDATE: Turns out, even though a channel is switched on all the time, there’s still the matter of having an entire network operating on a switched system which makes all channels only accessible to devices with two-way cable access. So much for my theory! Hopefully the NCTA and CEA will come to an agreement soon.

Apparently Comcast has been providing VOD stats in a few of its markets for a while now, including top movies, free programs and premium programs viewed on demand. Today Multichannel News reports that Comcast is expanding the practice to include new markets. Comcast in Oregon even plans to release stats weekly to local media.

This is a brilliant move on Comcast’s part. Just like Google, Comcast gets to be an authority on what people are interested in and gets a little free brand exposure in the process.

What’s next? Releasing switched digital video stats? An interesting new way to track the long tail.

Michael Vizard over at eWeek inspired me with his “10 Things They Believe in Silicon Valley but Nowhere Else.” Please note that the list below is my attempt at humor and does not in any way indicate the real opinions of Motorola, cable companies, industry reporters, or anyone else I’ve offended.

10 Things Only Industry Insiders Believe About CableCARD

1. CableCARD is a conspiracy that was started by TiVo (Humor, people, humor)
2. CableCARD is a conspiracy that was created to keep Todd Spangler and Jeff Baumgartner in business
3. The seven-oh-seven deadline was originally six-six-six
4. CableCARDs come with a line of credit and can be used at select local retail outlets instead of your Visa or Mastercard
5. Two-way CableCARDs mean not only can you get video-on-demand and switched digital video, you can now send messages to Brian Roberts through your cable set-top
6. Multi-stream CableCARDs generate multiple revenue streams per subscriber household and therefore greater ARPU
7. CableCARDs were designed so the government can monitor our VoIP (er, I mean digital voice) phone calls
8. The FCC is going to educate the public about CableCARDs by sponsoring the next World Poker Tour
9. EBay is setting up special auctions for non-CableCARD set-tops that need to be sold after July 1^st
10. CableCARD 2.0 comes with social networking features that let you add VOD channels to your MySpace, Facebook and LinkedIn profiles

Somehow I missed the news yesterday that Intel has signed a licensing agreement with CableLabs to support the Open Cable Application Platform (OCAP). What does this mean? Well, it means that CE companies and CableLabs are slowly resolving the issue of how to make two-way cable services available on CE devices.The July 1^st CableCARD deadline is a mere five days away, and the hyper-focused attention of the media has generated serious discussion about how consumers are going to be able to access certain cable services from their retail CE devices. Specifically, most CE devices today are not certified to support two-way services, including video-on-demand (VOD) and switched digital video (SDV). And the major sticking point around getting certified has been OCAP.

There are a couple of things to keep in mind about the CableCARD/OCAP debate. First, everyone wants the issue resolved. CE companies want to be able to offer more services. Cable operators want more people spending money on their services. And Motorola (which has nothing to do with how the issue actually gets settled) is in favor of anything that gets more people hooked on broadband networks.

Second, while OCAP has gotten something of a bad rap, consumers should be aware that the platform has a lot of potential to benefit mass audiences of cable TV. OCAP means more cable applications on the market faster. Whole-home DVR, Follow Me TV (whatever you want to call it), here we come.