Architectural Control
TM-H,
<< Yes, it is true that a committee, not Qualcomm, controls the overall WCDMA architecture. But, it is not true at this point in time that this committee could define that architecture in such a way as to exclude the use of Qualcomm's IP. >>
The same was true of GSM where "the committee" could not (or actually did not) define the architecture in such a way as to exclude the use of Motorola's FDD IP, and Ericsson whose baseline was used had little or no IP in GSM initially, and neither did Nokia, but having architectural control of a system and having (some) essential IP from which one derives revenue or offsets other IP through cross-licensing are two related but distinct matters. Ericsson became the king of infrastructure. Nokia became the king of handsets and a strong number 2 to Ericsson in infrastructure. Motorola lost considerable marketplace power.
<< my bet is that just about every proprietary open architecture you can think of is less than absolute, some measure of yield to market forces and other players having been required to establish the standard in the first place. >>
I totally agree with you and Moore addresses this in "Inside the Tornado" and in "Living on the Fault Line" as well as "The Gorilla Game." This certainly applies to CDMA2000 which was the undeveloped 3G standard announced in June 1997 by Qualcomm, Lucent, Nortel, and Motorola - not just Qualcomm - which gained immediate backing by Samsung, LGE, and the Korean carriers all of whom have made contributions to the platform and all of whom hold IP for their contributions. Today you have contributions to that standard like the Wideband AMR Codec which was a Nokia contribution. That said, Qualcomm controls the architecture while others influence it and participate in its commercialization, and they have successfully defended attempts to undermine that control. One recent example of this is the attempt by Motorola (subsequently joined by Nokia, TI, Philips and Altera) to control the framework for 1xEV-DV (CDMA2000 Releases C&D).
In addition to deriving revenue from their IP, as measures of their competitive advantage, they today produce 85+ % of the subscriber equipment modems, and 95% of the base station modems, used in implementations of their proprietary open technology. As they move to higher levels of integration increasingly they will derive more revenue from applications processors (integrated with baseband) decreasing the IC revenue that currently flows to TI, Intel, and others, and offset share erosion by Nokia, TI, Via, Samsung, EoNex, et al. A significant number of the carriers deploying their technology utilize their proprietary run time environment, many will probably adopt QChat as their PoC technology, and their value chain will probably adopt their Mobile Display Digital Interface (MDDI) as a camera phone interconnect.
Within the 3GSM domain Qualcomm will compete well in wireless ICs based on operational and marketing excellence which is how all participants compete, but they'll never see market share like they now see in subscriber equipment modems and integrated applications processors for CDMA2000 and may never supply a single base station modem. If BREW (even with J2ME) is ever deployed in the 3GSM world it will never have the prominence it has in CDMAland, and the standards evolved in MIPI (which Qualcomm still hasn't joined) will likely form the base for the dominant implementations of 3GSM PoC, camera interconnects, high speed multipoint links, system power management for handsets, and the physical layer of chips. All this while OMA specifies the mobile service enablers that ensure service interoperability across devices, geographies, service providers, operators, and networks, and several of the most powerful GSM/3GSM carriers of the world attempt to define requirements for an open 2.5G/3G mobile terminal platform (OMTP).
Wannabe mobile wireless gorillas have a challenging task attempting to overcome the barriers imposed by 3GPP, 3GPP2, OMA, MIPI, and OMTP in this ~125+ billion growing industry. These are defensive mechanisms. They are gorilla traps. It was never easy to become a gorilla. It's tougher today.
I might add that Qualcomm is impressing me. When I first started following Qualcomm back in 1994 they marketed on a shoe string (and standards participation is an extension of marketing as well as R&D), and really had no apparent globalization strategy. Until late 1997 they never participated in any of the IMT-2000 3G research programs in Europe or Asia which is the primary reason their technology wasn't even considered by ETSI. A real question mark in my mind was how well they could participate, contribute, and compete, in a comittee-based standards environment. They have gained a lot of maturity, and I've erased that question mark. They participate, contribute, and they will compete successfully.
The battle of princes to watch in the near future, IMO, is the battle between Qualcomm and TI as they compete for dominant market share in wireless ICs where today each has about 12% revenue share with a slight edge to TI. Either one could become a king with ideally 2x the market share of the other, or they could continue to duke it out neck to neck as dominant princes while watching in the rear view mirror to see what other challengers like Intel are up to.
This is an issue that Eric and I have always agreed upon from day #1. We're either both very right or both very wrong. --Mike Buckley
Scary. <g>
Best,
- Eric - |
