At the highest levels of soccer, we may discern two approaches to how to build a winning team. An example of the first approach is provided by Real Madrid and Chelsea. These two very rich clubs typically buy the best players at the prime of their career--Zinadine Zidane, Ronaldo, Michael Ballack, Andrey Shevschenko--and the coach does his best to meld these stars into a cohesive team. The purest, most faithful example of the opposite approach is offered by Arsene Wenger, the coach of Arsenal. He has a philosophy--a very distinct philosophy--of soccer: Constant flow, overlapping runs, keeping the ball on the ground, intricate passes of perfect weight, exquisite timing, speed and precision. He acquires very young players, often in their teens, whose promise is yet to be realized--Cesc Fabregas, Theo Walcott, Gael Clichy, Jutin Hoyte, Robin Van Persie--who can be molded to fit into and carry out the system and philosophy of soccer Wenger has developed. In the first approach, coaches build the system around the players they buy. In the second approach, players are acquired to fit into a system the coach has developed.
Brian Dipert's gloomy discussion of the direction of 802.11n tells us that what we have is a situation resembling the first approach--that is, you buy some players and then build a system around them. But, and this is an important "but," the players in this case are not at all the stars of Chelsea or Real Madrid, but rather serviceable if mediocre players who just happened to be there when the coach arrived at the training grounds.
The trouble began in August 2004, when Belkin released its pre-N product line. Soon, many other vendors followed, offering routers, APs, streaming-media adapters, Ethernet bridges, USB-, PCI-, and PCI Express-based client adapters, and other devices based on pre- and post-draft ver. 1.0 802.11n silicon from multiple suppliers. Talk about a state of confusion: Not only do many if these products, based on different silicon supplier foundations, not interoperate--but also platforms based on different silicon-engine revisions from the same silicon supplier, and even those employing the same silicon chip but using different firmware revisions, fail essential interoperability tests. This confusion reached its absurd climax a month ago when Qualcomm, on the verge of acquiring Airgo, announced that its latest chip was compatible with ver. 2.0 of the 802.11n draft specifications, even though the document does not yet exist.
The point about this approach to 802.11n is not long-term incompatibility. The majority of equipment today is DSP-based and firmware-upgradeable, so we should expect software up-revs to address bugs that currently hinder draft-802.11n compatibility. The trouble with the plethora of draft 802.11n silicon currently on the market is that it will constrain the IEEE standards committee as it works toward the finalization of specifications. Short of an act of heroism on the part of committee participants, it is very likely that market forces will remove from consideration anything which is not supportable by currently shipping hardware in conjunction with a firmware tweak.
In other words, the standards committee will likely build a standard around the players currently playing in the marketplace, regardless of their quality, rather than create the best standard and then have players play up to that standard. Dipert writes that "802.11n, as currently implemented in draft form, is incredible technology. It potentially could have been even more incredible, if the silicon and systems vendors would have exhibited more restraint." And he concludes: "Let's get 802.11n wrapped up as quickly as it can be. But when it is, I'll mourn for at least a moment what it could have been."