Remember the Ala-MIMO

There is a joke that goes: what do you get when you grab a man by the testicles? The answer: his full attention.

This little titbit came to mind after hearing the latest developments in what I have come to think of as The Case of the Missing Wi-Fi Standard.

That standard, of course, is 802.11n, which is claimed to increase the range of wireless LANs, more than double their speed, and grow back hair in places you farewelled it years ago.

It uses MIMO technology with multiple antennae (two in current gear and several more once they become cheaper) to split up the signal into lots of pieces, send them careening around your house or office at high speeds, and reassemble them into a continuous stream of data.

By all accounts, the technology works great, and "pre-N" products conforming to an early draft of the standard have been on the market for months. Nonetheless, it's widely held that 802.11n won't become an official standard until the IEEE's ratification committee sits again next year.

In the meantime, business and government buyers, who aren't generally keen on even scratching their noses unless they have a set of instructions and standards to blame if anything goes wrong, are left tapping ... their ... proverbial ... fingers ... on ... the ... proverbial ... desk.

Unless, of course, they buy into the marketing hype from companies like D-Link and Netgear, which recently began selling products certified to Draft 2.0 of the 802.11n standard.

Certifying products to a draft standard -- which by definition is likely to change at any minute -- is like marrying an identical twin blindfolded: you just don't know if you're getting the right one. It's a strange way to sell a product, but a necessary step for the Wi-Fi Alliance, which faces a new need to stay relevant given that wireless, i.e. cellular, broadband standards like HSDPA are stealing Wi-Fi's thunder.

The real question, however, is why -- after more than three years of work -- we're still no closer to having standards-based, commercially available products. To get some answers, I recently spoke with Jeff Fulton, a senior engineer with Netgear, who knows someone who knows someone on the IEEE's 802.11n working group.

The hold-up, he explained, is due to both infighting from some contributing companies and some lingering technical problems, which saw the standard fail to achieve the 75 percent support necessary for it to pass its initial vote. It was only with the formation of the Enhanced Wireless Consortium -- a splinter faction that now includes 72 vendors -- that a common functional architecture was agreed upon.

Nonetheless, this architecture is crawling through the ratification process -- forcing vendors to promote product certification that really isn't a certification at all, but more of an agreement that their products are all, well, close enough.

"We're in the high 98 to 99 percent certainty that there aren't going to be any changes to the draft standard until it gets ratified," Fulton told me.

"This is a standards committee that's trying to be completely rigorous and trying to finish the standards process, but customers already need this extra speed. We're seeing that commercial imperatives, and the timelines that go into developing this kind of new technology, really put a lot strain on the formal standardisation process."

One more thing may put even more strain on the formal standardisation process: the CSIRO.

Turns out a patent they lodged in November 1993 and were granted in January 1996 -- for some crazy contraption called a "wireless LAN" -- has something to do with the technology used in every piece of wireless LAN equipment sold ever since.

That's a lot of potential back royalties -- a fact that apparently has not been lost on the CSIRO's accountants. Although the CSIRO allowed its patent to be used for earlier incarnations of 802.11 wireless, it turns out the CSIRO has so far declined to sign what is effectively a permission slip for the IEEE to use CSIRO's technology in the 802.11n standard.

Without that permission slip, wireless LAN vendors will need to figure out a way to build wireless LANs without anything resembling a wireless LAN in it.

Back to my joke. If CSIRO's claims hold up -- and it appears they do, as reflected in the recent landmark US court ruling against wireless LAN equipment maker Buffalo Technology -- the organisation will definitely have the full attention of the IT industry. Claims CSIRO is holding out for royalties of several dollars per wireless LAN unit aren't yet substantiated, but if that's what CSIRO wants, it could keep full 802.11n from ratification for even longer as it fights the majority of the IT industry for what could be hundreds of millions of dollars in royalties.

I am reminded of the 1836 Battle of the Alamo, in which a small number of US soldiers held off an overwhelming Mexican force for 13 days. The Mexicans won in the end, slaughtering nearly everyone in the building in the process -- but the delay bought time for revolutionary soldiers that eventually led to the establishment of the state of Texas. To this day, Texans use the expression "remember the Alamo!" when faced with a situation calling for resistance against difficult circumstances.

Could CSIRO's stonewalling accomplish the same thing for Australia, securing a licence revenue stream to rival those from exports of TimTams and Kylie Minogue CDs? Perhaps. Alternatively, it could cause even more delays to 802.11n, swinging global sentiment against the white-coats at CSIRO and making Australia the pariah of the wireless LAN world (and I'm sure that would be bad).

What do you think? Should the CSIRO donate its intellectual property for the betterment of unwired mankind? Or should it keep sticking it to The Man, so to speak, and hold out for dollars on every unit? Let us know in comments below.