Specious arguments from Datacomm: I noticed this piece via a link at Tech Dirt and it's worth commenting on. Ira Brodsky makes the argument that the Wi-Fi Alliance's statement that they are resistant to products labeling themselves having to do with 802.11n coming to market in a way that might interfere with existing Wi-Fi networks will stifle innovation.
His arguments seem to focus on the benefits of MIMO technology coming to market earlier, and he cites the early release of 802.11g, before the IEEE had ratified a final standard, as an example. But 802.11g's early rollout was almost a disaster. Firmware was changing constantly. Equipment using the same chipsets often didn't work correctly among devices from different vendors, and performance glitches could make networks slower with 802.11g than with just 802.11b. Eventually, when the standard was ratified and the alliance could certify against a testbed, 802.11g settled down.
We've seen a host of extensions to 802.11g that have had some success in increasing speed, but they mostly--not entirely--work in homogeneous environments. Homogeneity benefits the manufacturers, not the consumers, because it requires lock-in. If you want 40 Mbps, then you have to buy all from one maker. For 20 Mbps, you can use any 802.11g. It's in the better interests of those who need the speed to suffer lock-in and lose the commodity pricing benefits that have come from standardization and interoperability. But it's not in the benefit of all consumers.
I should clarify that I don't mean that these devices shouldn't reach the market: they should. But don't come crying to the Wi-Fi Alliance when your non-Wi-Fi branded (or Wi-Fi certification pulled) pre-N, pre-X, pre-Y, and pre-Z equipment interferes with the operation of your Wi-Fi networks. As individual consumers or businesses, you can make the choice for speed or features over compatibility. But that doesn't mean the group trusted by its members to ensure interoperability needs to adjust its view of its own brand.
Brodsky writes: The market is brimming with products that offer proprietary but well-behaved enhancements to the 802.11a/b/g standards. Far from disrupting current Wi-Fi products, MIMO-OFDM products offered by vendors such as Belkin Corp. and SOHOware Inc. are Wi-Fi-certified and boost performance when used with 802.11a/b/g products.
His analysis is a red herring. These very enhancement are only well-behaved by accident. There's still the conflict between Atheros and Broadcom, which I've never seen a resolution of, in which Atheros's products use a dual-channel bonding that Broadcom claims can cause disruption and lower speeds on nearby networks. Atheros said their testing didn't show this, but all of their OEMs seem to now support a dynamic mode for this element of their higher-speed proprietary add-ons that drops out of the dual bonding if non-Turbo mode clients are detected.
It's pure luck and some intent that none of these enhancements is truly disruptive. If devices are FCC approved, even a disruptive add-on might still conform to FCC rules. The Wi-Fi Alliance threatening decertification is the only tool that might convince a manufacturer with FCC licensed equipment to modify its firmware to avoid stepping on Wi-Fi networks that would be affected.
Finally, the Belkin products are certified as Wi-Fi only in their Wi-Fi modes: the fact that they use MIMO is perfectly reasonable, but they're not certified in their pre-802.11n speeds, of course, and that's the point of contention. MIMO can certainly work for 802.11a/b/g, and it apparently provides much better range for existing Wi-Fi network adapters connecting to a pre-N access point, according to early reviews of Belkin's gear, even without using its proprietary high-speed mode.
Here's the part that I find most specious: Some leading vendors are worried the pre-n products spilling onto the market could reshuffle the market share deck. Their fears are well founded. But it would be unfair to make users wait three years just so slower-footed vendors can catch up. But the Wi-Fi Alliance isn't the government: they're not keeping MIMO pre-N products from the marketplace. They're just saying if they cause interference with Wi-Fi, disrupting the alliance's brand promise, then they can't have the Wi-Fi label. That's a market and marketing threat based on a certification and testing standard. Companies can release all the pre-N they want, but they can't necessarily call it Wi-Fi even if it can handle a/b/g.
Brodsky concludes: ...the alliance must be careful not to confuse special interests with common interests. It's just as important to protect vendors' right to innovate. Standards help grow the market to the next level, but innovations like MIMO-OFDM get the ball rolling.
The alliance doesn't exist to protect vendors' right to innovate. It exists to protect its members' ability to use a brand that has a strong promise to consumers and business IT: that anything labeled Wi-Fi interoperates with anything else labeled Wi-Fi. Innovations aren't preventing by adhering to a brand promise: they're preventing by disrupting a clear message in the marketplace that has led to Wi-Fi's total dominance and ease of interoperability.