The time has come for Cisco to make a huge play in the home.” – John Chambers, Jan 2009
Users want quality and they want it now. Today, the Internet experience is clearly completely different from 10 years – accesssing the Net over dial-up access was a trying experience, waiting for even the graphics on a Web page to fully load… and downloading one file might slow down surfing to a crawl. Today, the experience is much better courtesy of video services like YouTube (which uses Adobe’s Flash technology for the streaming), but real-time video remains very much a hit or miss phenomenon, and largely low quality through services like Skype Video. Want to try a video chat session over a Net Cafe’s Wi-Fi connection? Good luck.
In the enterprise, a high-end two-video solution is possible these days, but it can be prohibitively expensive, with each video room costing well north of $300k, not to mention dedicated bandwidth from service providers. Yet, the experience of Cisco Telepresence is something to behold (I’ve tried it personally twice thus far) and the exciting value proposition of very good two-way video is that you quickly forget about the technology, and it “feels” like you’re communicating in person. A smart thing that Cisco requires is that installations of Telepresence need to have precise room specs, including the room backdrop and table heights. Why? The circular table extends into the video conference wall, simulating talking with someone across the room.
What does this have to do with Pure Digital, the maker of the popular Flip series of portable video recording devices? After all, it’s a big premium on a company which may do around $200M in sales this year in a relatively low-margin consumer business. In my view, this deal is all about streaming video and more importantly two-video in the long-term, and for Cisco, it appears to be very-well positioned for this transition given that it dominates in a number of areas where intelligence is required. It all starts in the core of service provider networks, and Cisco’s share among service providers is very strong, around 50-55%, starting with its largest router, the CRS-1 core router. As IP routing is being built out throughout service provider networks, Cisco is winning more than its fair share of these equipment sales, and its acquisition of Scientific Atlanta in 2005 for $6.9B brought a huge video play for the company, and a strong presence at the cable companies. The purchase of Linksys cemented Cisco desire to go “end-to-end” with a strong profile in the consumer’s home (combined with S-A’s strong share in cable digital boxes and PVRs). The end game is moving an increasing amount of video content around the home and over the Internet, and two-way video seems like it could be the next “killer app”.
The big problem in the consumer arena? There are no standards to speak of… well, that’s not quite true. There’s just a lot of them. As we’ve seen with the Blu-ray vs. HD-DVD battle of a few years ago, major CE companies really want their protocol to be the way that machines communicate because it means significant $$$ in royalties. In the convergence of CE devices such as televisions, set-top boxes, packaged media players, and computers, figuring out a way of moving information between them has been challenging, in part due to the concerns of the content companies. Wi-Fi appears to be one of the basic building blocks of most consumer networks, providing basic connectivity between devices, but beyond that, figuring out how to share files between Sony’s PS3 and a PC for example is… rocket science. (I’ve used a tool called TVersity which can provide “translate” between the two devices). Ultimately, getting devices more intereconnected needs to be completely idiot-proof. In this light, is there a role for companies that can be glue that holds in all together? Absolutely.
Turning back to why Cisco wants to own the end devices, ensuring video quality is absolutely about having knowledge about the end-to-end connection. A good example of this is in AT&T IPTV installation, where only the information between watched is sent between the AT&T’s video servers and residential neighbourhoods, helping to reduce the load on AT&T’s network. With AT&T’s network barely able to deliver 2 HD channels to each home, there’s a complex handshaking going on between AT&T’s network and the set top boxes in the consumer’s home to get channel changes working fast. The concept of the channel change might be the closest thing we have today to requiring a very fast response time for high speed video (HD video rentals are not as demanding because if a video starts a minute later, that’s ok but clearly this isn’t good enough for a video chat). Now in bringing two-way video to the consumer arena, it will be critical that Cisco help to not only develop the home networking market, but also have end devices that effectively communicate all the way through the end-to-end path. For example, will service provider networks want to know if there’s congestion taking place within a consumer’s home network? Absolutely. At the end of the day, how fast video is streamed is limited by the slowest element in the path. Likewise, if an end-device is transferring information that isn’t time sensitive, it would be helpful to flag this information as “best effort”, helping not only the service provider but also the consumer in terms of getting a better overall quality of service by not trampling over time-sensitive services like VoIP… and two-way video calls in the future.
The next step for Cisco with Pure Digital could be to gear the video recorder towards two-way video over home networks, especially since the Flip Video camera already supports a video standard called H.264, which is currently in use in a modified form in Cisco’s high-end Telepresence solution. Longer-term, it’s possible that we could see a Cisco Pure Digital Home Telepresence solution, which would allow high-quality video calls in a handheld form factor. Where this gets really interesting is if this mini-Telepresence solution could then be placed on top of a user’s flatscreen TV, and then the video could be outputted through a cable from the Flip to the TV. While this sounds like a far-off prospects, the output specs of the existing MinoHD Flip camera are pretty impressive: 16:9 widescreen, HD 720p (1280 x 720) at 30 frames. If this device is paired up with a Scientific Atlanta set-top cable box, the outgoing video could be handled by the Flip video camera, and the incoming video could be handled by the set top box. When Cisco first started talking about the idea of Telepresence in the home, the key question was always how they were going to get the outgoing video integrated into the box (cable companies want this box to be as cheap as possible because they’re subsidizing the box to a great extent). The answer might be in developing the Flip to be a high-quality “outbound video” source, able to send its video content in real-time to real-time video conferencing destinations like Skype, and even Cisco Telepresence call centers! The iPhone will even play a part in all of this, as Cisco could require that Apple allow Cisco to sell an iPhone app which enables full integration into Cisco’s consumer device initiative, perhaps even enabling the iPhone to be used as the outbound video source for home Telepresence solutions.
Networking in the home network is going to get a lot more crowded, as I wrote about in articles on the subject of femtocells (providing cellular access within the home) – with service providers wanting to put their 3G (and 4G down the line) devices into user homes to help off-land smartphone data traffic onto fixed broadband connections, we could see much better performance from smartphones within the home. Does Cisco need to be a player within this emerging market? Cisco does have an investment in UK-based ip.access, which secured an order for 7M units from AT&T and was ranked by ABI Research as #1 in the 2G picocell market (larger-ranged femtocells for areas like businesses and subways) and #2 in the femtocell market (behind Google, T-Mobile VC, and Advent VC-funded Ubiquisys), while also owning WiMAX player Navini. Over time, a cable company could want an integrated device with DOCSIS 3.0 for high-speed Internet and a WiMAX/CDMA femtocell (Comcast is an investor in Clearwire), and Cisco will want to make sure it takes advantage of this move to DOCSIS 3.0 at the cable companies add support for easier home networking (the $120M acquisition of Pure Networking in 2008 was made with this in mind).
In summary, Cisco’s long-term home strategy is still geared towards them selling more intelligent routers to service providers, but with this acquisition, I do see the wheels starting to turn in John Chambers’ head about how Cisco is going to get two-way video actually moving in the consumer arena. Looking ahead, the pieces are starting to fit together through a combination of faster home networking (802.11n) and faster pipes into the home (DOCSIS 3.0, Fiber, VDSL2, WiMAX). The more challenging part is getting the devices to talk together, and in this respect, Cisco really has the most to gain from a LOT more video being generated by the consumer just in terms of more router sales. Cellular wireless might be the Achilles’ heel for Cisco in this entire scenario, and picking up Nortel’s CDMA and LTE assets in a firesale might be the more important next step for Cisco. Now trading at around 12.8x earnings, Cisco is down on the day but has had a good mini-run to the $16 level from under $14 earlier in the month. In my view, the positive for Cisco is that they do appear to see the bigger picture, and have the pieces to not only effect a fundamental shift towards video telephony, but also materially benefit from the paradigm shift…
