David Berlind, in his article, "Intel wags mangy Centrino dog by Boeing's tail" , makes it clear how he feels about Intel's deceptive Centrino marketing campaign. However, I have to wonder why David is so surprised.
How many of us have called tech support at one time or another -- to get help on product XYZ that should "just work" when you plug it in (but doesn't for some reason) and been told, in so many words, "I am sorry but we cannot guarantee performance or our products with other vendor's products."
My most recent personal experience along these lines was when, after an hour or so trying to make myself understood to technical support in New Delhi (or thereabouts), I was told that my brand of broadband modem (recommended by my cable provider) was causing the problem with their brand of wireless router (as it happens, just before he provided the solution which fixed the problem). In reality, I have no idea which box was "not compatible" but the problem has not recurred.
Is it disingenuous? Of course it is. Is it anything more than marketing hype? No, it is not.
Perhaps the most interesting point, which David seems to have overlooked, is that if Intel wants to claim that you have to match a Centrino-brand notebook with a Centrino-certified hot-spot, then they also raise the question of whether or not a Centrino-branded notebook will work reliably in a hotspot which is not Centrino-certified. In short, does Intel expect me to buy a notebook which works reliably in some hotspots but not others?
As I recall, the Centrino standard (which does not include Celeron processors) does not (or at least originally did not) include 802.11g. Are they saying that I should settle for 11Mbps -- and reliability only in some hotspots -- instead of having a reliable 11Mbps in every hotspot and 54Mbps in a growing number of them?
Does Intel secretly implement a "performance tweak" when a Centrino-branded hotspot detects a Centrino-branded notebook? (It is possible since standards always represent trade-offs which can be compensated for when compatibility is not an issue.)
By cutting Celeron out of the Centrino family, Intel is also effectively saying to its price-conscious customers that Intel doesn't value them. Intel is telling its customers that if they are not willing to buy the more costly Pentium processors, that these customers don't "deserve" to have the benefit of Centrino technology. That strategy is just plain dumb!
Of course, customers buy their notebooks from their favorite OEM based upon a number of factors -- not the least of which is price (cost/benefit) -- and "what's inside" is of less importance than it once was. For instance, I'd be hard-pressed to give up my Dell computers just because they decided to get their processors from AMD. Further, if my Dell Centrino notebook didn't work reliably in my favorite hotspot, I would assume it was a problem with the notebook, not the hotspot. And a problem with the notebook would make Dell look bad, not Intel -- leading Dell to reconsider from whom it buys its processors as well as its radios.
Clearly, Intel wants to have it both ways -- to convince OEMs to offer only their radios (because customers demand it) and for hotspots to seek Centrino certification (presumably for a fee, on top of using Intel-branded wireless access points) -- again because their customers will demand it! Well that's just tough!
In the heyday of multi-million dollar IBM System/370 systems and proprietary SNA networks, a vendor could claim that compatibility was an issue and that uniformity of brand selection was a must. But those days are gone. Today, personal computing is accessible to all at commodity prices and compatibility is simply expected. Those that cannot deliver a high-level of compatibility and reliability will not be here next year.
Intel is playing a risky game (especially if their Centrino-to-Centrino marketing scheme includes unpublished "enhancements") and they run the risk of hurting their reputation with their OEMs -- who will drop Intel if their products do not perform as advertised.