
Digital TV's Slow Acceptance in Germany's Cable Market |
|
By Barry FlynnGermany is potentially Europe's largest digital cable market, with over 20 million subscribers. Yet digitalization has been slow. Premiere World, the Kirch-owned pay-TV operator that is Germany's dominant digital TV player, originally had a target of 3.5 million subscribers by end-2000, and 4.0 million by June 2001. The actual result today is under 2.5 million customers, around 55 per cent of whom are on cable (the rest are on satellite). That equates to just 1.4 million digital cable subs. The reasons for this are complicated. First, conversion to digital TV in Europe has been sold to consumers primarily on the basis of two propositions: premium programming and increased channel choice. Unfortunately, neither of these selling-points are particularly attractive to German cable customers. Analogue multichannel television via cable is both cheap and ubiquitous (accounting for around two-thirds of Germany's 33 million-plus TV homes), and already shows lots of movies and sports (Price Waterhouse Coopers, for instance, estimates that in excess of 12,000 movies are broadcast by Germany's free-to-air channels annually). Because of this, Germany's cable customers have just not been willing to pay extra for digital, being largely happy with what they have. Second, there are serious infrastructural issues. For a start, the incumbent telco, Deutsche Telekom (DT), remains the dominant cable player, and, for a variety of reasons, has been reluctant to upgrade its cable networks. Recently, this has related mainly to the fact that it didn't want to invest in networks that it was being forced to sell off by the European Commission (efforts which are continuing at the time of writing). There is a striking contrast here with DT's approach to upgrading its PSTN exchanges to ADSL. Here, DT's investment has been considerable in recent years, achieving in excess of 2.5 million ADSL connections by this summer. This has turned the country into Europe's biggest ADSL market. But even had DT decided to press ahead with upgrading its cable networks, the move might well have been stymied by Germany's complex four-level cable infrastructure (see diagram below). Although the telco has historically held a dominant position at Levels 1-3 (i.e. all the way down to the dwelling themselves), it owns only a minority of customers at Level 4 (i.e. within the dwellings). Access to these is generally controlled by separate entities - often landlords or owners of apartment blocks. So DT has had no guarantee that even if it upgraded its own part of the cable network, Level 4 operators would allow any of their customers to access its advanced digital services across it. Indeed, US cable investor CAI, which acquired Germany's North-Rhine Westphalia (NRW) network from DT in February 2000 but had to file for bankruptcy in July this year, has run up against precisely this problem. After purchasing the network, it upgraded the wires to a million NRW households to allow them to access Internet or telephony services, but the Level 4 operators actually controlling those households refused to play ball. For this reason, CAI currently has no Internet or telephony customers on its NRW network - or, indeed, on its second franchise in Baden Wittenberg. Such issues are one of the major reasons why, in the latest round of bidding for DT's six regional cable networks, potential investors are understood to have offered only between 2 and 2.3 billion euros ($1.94-$2.23 billion). These include a private-equity funded consortium led by Liberty Media, plus three other investment bank consortia: Goldman Sachs/Providence Equity/Apax Partners; Warburg Pincus/CVC Capital Partners, and Hicks, Muse, Tate & Furst. Liberty originally offered 5.5 billion euros ($5.34 billion) for the DT networks, but was rebuffed by the Bundeskartellamt (Cartel Office) in February, because - ironically - the regulator believed Liberty would not be investing enough in upgrading the cable networks post-acquisition to compete effectively against DT's ADSL offering. This time round, the Liberty consortium has emerged as one of the lowest bidders. Whatever the outcome of this contest, which should be known within a matter of weeks, German cable's rate of digitalization should accelerate substantially in 2003 as new investment begins to kick in. Meanwhile, US cable operators should also be aware that, despite the slow start, Germany's digital cable plant looks likely to become one of the earliest and biggest cable test-beds for the MHP interactive TV middleware standard - much of which has been incorporated into CableLabs' OCAP specification. For - somewhat paradoxically - the ailing Premiere World platform is pledged to launch MHP early next year if current restructuring plans go ahead. The reason Premiere is able to talk about the introduction of MHP as a realistic prospect is something of a historical accident. Back in 1998, Kirch technology division Beta Research was developing a Java-based middleware called "Betanova" for Premiere's second-generation digital set top, the so-called "d-box 2". The spec for this particular box was high for its time - 32 Mbytes of RAM and 8 Mbytes of Flash, running on a 66 MHz PowerPC processor (in part because of the demands Betanova would place upon it). That fact, combined with four years of middleware development work within a Java TV environment, means that today the Kirch-owned company is confident it can squeeze MHP into existing second-generation boxes, over a million of which have been deployed to date on German cable systems. Says its director of business development, Sven-Olof Koopman, "First evidence can be seen in our labs, where we run already our MHP software on the d-boxes - which allows us to conclude today that [it] will be as performant as today's Betanova 2.01. The reason for that lies in the fact that we have developed our own [Virtual Machine] over the last five years and have constantly optimized the software within an operational DVB environment. Therefore we have gained enormous experience in handling Java, in dealing with download mechanisms and in developing high performing applications." Rival middleware providers - such as OpenTV - may be skeptical that MHP will run adequately on a 66 -Hz processor, but Koopman is adamant: "Our MHP middleware will be higher performing than the other products on the market which have been developed just within the last 12 months, often based on today's state-of-the-art processors with more than 100 Mips." Indeed, Koopman believes that his company has advanced so far that its MHP solution has, in effect, opened up the possibility of its adoption in low-end set tops. "It will be an important step in the direction of low-cost MHP STBs which are necessary to push the market and to increase MHP acceptance at cable operators," he declares. If Koopman is right, that bodes well for the CableLabs' OCAP specification, which has been accused by many critics of being too unwieldy to be widely taken up by US MSOs. For, if MHP can be squeezed into a low-end set-top, OCAP probably can be, too. All cable eyes will be on Germany in 2003. Table: Germany's four-tier cable system
Source: Price Waterhouse Coopers Barry Flynn is a freelance technology writer and consultant. He can be reached via e-mail at barryflynn@compuserve.com or at www.barryflynn.co.uk. |
| Copyright © 2002 Cable Television Laboratories, Inc. All Rights Reserved. | ||