With a skillful combination of nationalistic dog-whistling and misinformation, Ars Technica has created a storm of outrage against Australia with the claim that the country’s CSIRO has patent-trolled its way into American WiFi users’ pockets.
At issue is the recent settlement which brings the Commonwealth Scientific and Industrial Research Organisation’s long battle with US vendors over WiFi modulation to a close.
The abbreviated history of the battle is that back in 1992, the CSIRO applied for a patent covering techniques to overcome multipath distortion in wireless LANs, a combination of known radio techniques such as orthogonal frequency division multiplexing (OFDM), forward error correction, and interleaving. The patent was granted in 1996, and was incorporated into the 802.11a and later 802.11g standards.
Having advised the IEEE that it would apply fair, reasonable and non-discriminatory licensing to the technology, the agency sought to get license payments, first directly from vendors, and then through the courts – in the long, drawn-out process now near its end (not long, we should add, before the patent itself expires at the end of next year).
All of this is intolerable, however, to Ars’ John Mullin, who in this article essentially accuses the CSIRO of patent trolling, and takes exception with politicians’ mis-statements that CSIRO “invented WiFi”.
There isn’t space to canvass all of the issues raised by Mullin in the Ars article. For a strong analysis of the validity of the patent – which survived several re-examinations in the lawsuit – there’s a post by Australian patent attorney Dr Mark Summerfield on his Patentology blog which looks at the history of the CSIRO case, and another examining the patent itself.
But a few of Mullins’ complaints are worthy of discussion.
CSIRO hid in the shadows
“CSIRO didn't even participate in the 802.11 committee. The group published the first 802.11 standard in 1997 and CSIRO came forward years after the fact.”
There’s a good reason CSIRO didn’t participate in the committee under its own name: participation was delegated to Dr David Skellen, one of the inventors named on the patent, a long-time member of and contributor to the IEEE, and who founded the spin-off company Radiata which was tasked with turning the technology into a product.
The substance of the work was certainly known to the IEEE, at least as early as 1997, since in that month IEEE’s flagship publication Explore published an article by Dr Skellern and others named on the “069 patent” on WLANs. The article’s submission date isn’t given in the abstract, but was probably some time in 1996.
In 1998, the IEEE asked CSIRO to confirm that it would license the technology if it was adopted, and CSIRO agreed to do so.
Even after being taken to task over this notion, Mullin persisted: “why did CSIRO bypass the only meaningful way to determine the best method—submitting an IEEE proposal that could be voted on?” he asked in his defence.
This is flat-out wrong. CSIRO didn’t “bypass” the IEEE; its participation was at arm’s-length via Dr Skellern and Radiata, and the technology was submitted to the committee, and voted on, and accepted.
Cisco was scammed!
“The closest thing CSIRO ever had to a commercial product was a 'demonstration chipset' produced by Radiata, an Australian wireless company formed by ex-CSIRO employees. Radiata was acquired by Cisco in 2000 for $295 million but turned out to be a dot-com era boondoggle. Its products weren't successful. Cisco ultimately took a large write-down on the value of the Radiata acquisition.”
Regardless of the reasons Cisco decided to drop Radiata, the idea that it was a “boondoggle” – that Cisco was somehow scammed into the acquisition – is hard to credit. Cisco’s reasons at the time seem pretty clear: having invested in Radiata early on, once the company got its product to demonstration, the networking giant grabbed Radiata before it had a commercial-scale device in production.
That decision rested with Cisco – which at the time certainly had the engineering know-how to assess whether or not the Radiata chipset actually functioned.
Mullin also misunderstands the role and remit of CSIRO: it doesn’t exist to create products, because it’s supposed to avoid competing with the private sector. Once a technology is judged ready for prime time, CSIRO’s job is to license it to the private sector. Doing the science, but leaving the commercialization to the private sector, is not patent trolling.
CSIRO didn’t invent 802.11
True enough, because no one organization did so, not even the IEEE.
The IEEE’s role was, and is, to assemble technologies that already exist, brought forward by their inventors, and choose between them for incorporation into the standard by way of a vote. The IEEE doesn't invent any of the core technologies - it is the arbiter of which deserves inclusion in the standard.
When the IEEE is setting a standard, it’s not doing so in a clean room. One criterion the standards committee will apply is whether a proposed technology actually exists; and another is whether it works.
Vendors participate in standard for both selfish and altruistic reasons: the altruistic reason is to end up with a standard that works; the selfish is to get “my” technology (and patents) into the standard rather than someone else’s. Success is valuable, even if the license fee turns out to be nominal, because a successful standard will be bedrock of millions of sales.
As this list shows, there’s a huge number of vendors whose patents cover different aspects of the 802.11 standard. What patent-holders are asked to do – and the CSIRO agreed to do – is license their IP on “fair, reasonable and non-discriminatory” terms. When the CSIRO asked for $4 per device as its initial ambit claim, vendors may have considered this unfair (and they resisted paying it, all the way to the courts), but nobody’s suggested that the CSIRO tried to apply different licenses to different vendors, or to deny access to its technology to particular vendors.
Many of the patents that are incorporated into 802.11 are even older than the CSIRO patent – so anybody asserting that a new WiFi patent wasn’t a serious proposition is ignoring the patents owned by Fujitsu, Nokia, the University of California, Motorola, SMC, Xircom and others.
Nobody “invented WiFi” – and in this, the hyperbolic statements made by Australia’s science minister and various press should be challenged. Various companies invented various parts of what got voted into the standard, and all of them hold and to varying degrees enforce the patents that the IEEE incorporated into the standard.
Sure, trolls try and parlay their ownership of patents into income. But so do legitimate inventors. ®