Lakers,
re: Proprietary Open Standards Development v. Committee Based Standards Development
<< Keep in mind how the GSM is created. It was because of the need to bring all the fragmented markets together >>
We agree on that one. That was in fact one of the reasons GSM was created. As GSMA would say:
It’s difficult to recall the bad old days. Just imagine a car that will tackle the autobahns of Germany but stops dead when you drive it over the border into France. This pretty much sums up the mobile phone scene in the early 1980s.
Kind of reminds you of CDMA in Y2K01 doesn't it, now that a trip to Paris, London, Johannesburg, or Bogota costs not much more than a trip to Minneapolis for me? Of course as Maurice would say, there is always Globalstar.
As business was becoming increasingly international, the cutting edge of the communications industry focused on exclusively local cellular solutions. And none was remotely compatible with any of the others. NMT 450 in the Nordic and Benelux countries. TACS in the UK and C-Netz in West Germany. Radiocom 2000 in France and RTMI/RTMS in Italy. These were some of the building blocks of a tottering Tower of Babel that might enable you to call the office if you were in your own home, but not if you were with a client in another country.
Mobility and cross-border roaming was the second issue.
Capacity was the third issue. Over-capacity threatened.
Data transmission was fourth. Analog at that time was limited to voice only transmission and telecoms were evolving to ISDN.
<< So when they were working that things together, it was totally different from what we are seeing today in 3G. That is everybody brought their working solutions to the table and talked it over, both technology wise and political wise. That is they at least had something that was run and tested in the real world one way or another. >>
We are back to that "real world" again, and my interpretation of what was real, and what was not real, what was working or not working, and who was talking to who at table and how, is different than yours, so perhaps we should compare notes.
I'm open to your commentary but FWIW, here is my understanding of the evolution of the open comittee-based GSM standard, and if you think I have stated or interpreted anything incorrectly please advise:
GSM was established by CEPT in 1982 to deal with the issues of fragmentation caused by multiple standards, several of which were rather proprietary, spectrum (clearing 900 MHz), capacity, cross border roaming, and the potential of data transmission wirelessly.
The membership of GSM 1 and the original 3 working groups that were established (the 4th was data oriented) were comprised of telecom administrators (no vendors). In 1985 the basic requirements of GSM were set forth and consisted of 18 short line items grouped in 5 categories, and these were distributed to the telecommunications industry.
Cellular was strictly analog at that time. Presumably all the major cellular players including Motorola, Nokia-Mobira, Ericsson, Siemens, and AP-Phillips were evolving various digital technologies in their labs, but none was yet working or "real world".
Eight proposals (four from France and Germany and four from Scandinavian countries) in response to requirements were submitted to GSM3 and it was decided to compare the proposals on the basis of prototypes allowing actual radio transmissions.
Comparative testing of prototypes took place in December 1986 at CNET labs in Paris, and comparative results were announced early in 1987, at the same time additional proposals were being received by 3GSM.
Perhaps this is what you mean by "everybody brought their working solutions to the table and talked it over, both technology wise and political wise. That is they at least had something that was run and tested in the real world one way or another."
Nobody was at table yet and comparative tests of prototypes were confined to laboratory.
Up till this time industry vendors had no "official" involvement in GSM or its working groups.
GSM3 had expanded the original requirements to over 100, grouped in 12 series.
More importantly the PN recognized the "prestige and the advance that would be conferred to a proponent were its solution to be chosen. To circumvent this problem none of the proposed solutions was selected. Only the key features of the transmission method were decided.".
These key features were summarized concisely in what was called "The Broad Avenue" which was designed to provide a level playing field for industry participants who were then invited for the first time to participate in GSM3 working groups. on an ad hoc basis and of course they became directly engaged when ETSI was formed in 1988 to evolve the requirements into the GSM standard in 3 planned stages (specification writing, validation, field tests) which later were deviated from for practical purposes.
The "Key Features" (added to 100 odd requirements) were:
* medium-sized band with 200 KHz carrier separation.
* digital speech transmission at a rate not exceeding 16 kbps
* Time multiplexing order of 8, expandable to order of 16 for future vocoder.
* Slow frequency hopping capability.
That's it. Standard development began without a working solution, internationally, in standardization committee, on a relatively level playing field. Stragic markrting decisions had to be made by vendors on whether or not they wanted to play the comittee-based standards game.
"The understanding of the gain to be obtained by combining resources, and of the potential business opportunity offered by mass-market radiotelephony resulted in substantial man-power and financial commitment and effort by the participants", with plenty of risk and no guarantee of financial reward.
<< The open standard is created out of proprietary ... The open standard is created to bring the different proprietary together. >>
In the case of GSM some players brought "proprietary" in the form of IP to committee, some acquired it from others, some developed it as the standard developed. This is true in the development of any committee based standard. It was certainly true here. To minimize the impact on fellow participants, cross-licensing, whether royalty bearing or at no charge, comes into play.
<< The open standard is the end result, not the starting point. >>
That depends on whether you are dealing with proprietary (open) standards or committee standards when developing technology architecture.
<< That is how the industry works. >>
It is not the way the wireless industry has worked so far.
That is the way you would like it to work and the way Qualcomm would like it to work.
If it was the way it worked cdma technology adoption by carriers would have been considerably higher over the last several years than it has been and you would see this reflected in subscriber growth and market share growth and you would see a lot more carrier members of 3GPP2, and a lot more orders on the books for 1xWhatever..
Your looking at a rather classic battle for control of architecture between a single company (proprietary) or a group of companies. The classic chronicling of this phenomenon in the tech industry was done by Morris & Ferguson in "Computer Wars" and G. Moore in "Inside the Tornado".
In proprietary standards whether open or closed you typically develop then document or standardize. In committee based standards you go from standard to development.
It really is not to difficult to categorize and distinguish between GSM & CDMA in this regard, and it sure isn't to difficult to determine which the carriers favor.
It is not a trivial matter because if you buy off on Morris & Ferguson's or Moore's marketing theories, control of architecture provides a significant competitive advantage for the proprietor that leverages well beyond royalties that accrue from IP. Companies take this into account when determining their strategic marketing plan. Great technology companies are not always great marketeers.
If you think the industry is working that way, go back to gsmworld and the CDG site, and look at carrier membership by country, look at carrier adds for each of the last several years, and look at networks coming on air in each of those years, then tell me its working that way with a straight face..
<< Give me one real example of things as a pure result of a standard by committee without having a practical solution first. >>
The Global System for Mobile Communications
GSM - GSM Phase 1 - GSM Phase 2 - GSM Phase 2 Plus - and soon 3GSM
The end result of this effort is:
* GSM Networks on Air - 413
* Networks Coming on Air - 83
* GSM Countries on Air - 151
* GSM Countries Coming on Air - 20
* GSM Total Subscribers - 560 million
* GSM Association Members - 551
<< Look at ISO OSI 7 layers model, it's a thing of beauty right? Only in the class room my friend. >>
You have a better model?
Best,
- Eric - |
