Is the CSIRO a patent troll? US debate turns feral


news An extremely harsh war of words between Australian and international technologists has erupted over a controversial new article published in the United States documenting evidence that Australia’s peak research body’s $430 million patent claim over 802.11 Wi-Fi technology might have been constructed on shaky ground.

The article, entitled How the Aussie government invented Wi-Fi and sued its way to $430 million, was published a week ago by established investigative reporter Joe Mullin, a reporting fellow at the UC-Berkeley School of Journalism on Ars Technica, one of the United States’ largest technology sites.

Broadly, it chronicles how Australia’s Commonwealth Scientific and Industrial Research Organisation had, using a carefully crafted legal strategy, won a series of global lawsuits over key patents it holds with relation to the 802.11 Wi-Fi standard used in virtually every mobile device produced at the moment, with a series of high-profile companies ranging from telcos like AT&T, Verizon and T-Mobile, to Microsoft, Hewlett-Packard and Dell.

Total revenue from the cases has now passed $430 million, leading to Australia’s Minister for Tertiary Education, Skills, Science and Research, Chris Evans, to pronounce several weeks ago that it was hard to imagine “an Australian-invented technology” that has had a greater impact on the way people lived and worked.

However, as Mullin documented in his article, opinions differ drastically on to what extent the CSIRO was actually involved in creating the Wi-Fi technology which is in such common use today. The Institute for Electrical and Electronics Engineers, a standards-setting organisation, created a working group for the evolving 802.11 standard in 1990, three years before the CSIRO filed its first patent in the area, and one of the key Australian engineers involved in the CSIRO’s work, Terrance Percival, has acknowledged in court that the team hadn’t invented any of the constituent elements of its wireless strategy.

It’s not the first time that the CSIRO has been criticised over the issue. Many global technology companies have fought the company’s claims on the Wi-Fi issue — with most ultimately losing, and an extensive article published in The Economist (no longer available online — it has been taken down over the past several years) has documented the fact that the history of Wi-Fi is far from simple, and that a common sense definition of where the Wi-Fi standard came from runs counter to the CSIRO’s position on the matter.

However, Mullin’s article has highlighted a key difference between Australian and US views on the issue.

A large number of Australians have strongly criticised the article online over the past week, defending the CSIRO and its patent litigation.

“Wow. Fuck Ars,” wrote one Reddit commenter. “Both articles border on racism as far as I’m concerned. To paint the CSIRO as patent trolls is fucking laughable. The first $205 million netted through the patent lawsuits was donated to the Science and Industry Endowment Fund. The author has not substantiated his claims that the patents were insufficient or not related to any of the 802.11 standards. The reality is that the CSIRO was encouraged to monetise their inventions. Yeah, US government research organisations have given away their stuff for free. Good for them. That doesn’t mean the CSIRO are patent trolls when they monetise their inventions.”

Wrote another: “As usual, the first thing to go missing are the facts … CSIRO is one of the world’s top research institutions. It is loved by Australians, and highly regarded in many fields by other scientists around the world.” The comments posted on Ars Technica by Australians are similarly vitriolic. “Consider it a small amount of payback for all the shitty IP laws you’ve exported to us, and the billions Australia has donated to the companies that bought those laws in your jurisdiction,” wrote one local.

“I’m getting a very strong feeling of sour grapes from the author,” another added. “CSIRO is one of the world’s top research organisations run and staffed by fantastic people. I have no doubt that the guys at CSIRO did come up with something worth patenting, and they’re entitled to go after companies that infringe on their patents, especially given the free trade agreement between Australia and the US. If we have to put up with the US’s bullshit patent system, the US can put up with ours too.”

Australia-based patent attorney Mark Summerfield, in response to Ars Technica’s article, published his own extensive history of the CSIRO’s Wi-Fi patents, concluding that the patents had been subject to a decade of intense scrutiny and had withstood all assaults on them.

“This really has been a David and Goliath(s) battle and – as in the original story, though perhaps less common in real life – David has emerged victorious,” he wrote. “We can now hope that the rewards of all this time and effort are invested in Australian research and innovation, and that some of the results will be equally successful. Congratulations are certainly due to all involved at CSIRO!”

However, a number of commenters predominately based in other countries highlighted wider problems with patents in general, such as the widely held belief amongst many in the technology industry that the broader patent system has the potential to stifle innovation. The extensive patent portfolios held by a number of large US technology giants — with companies such as IBM being among the most prolific holders of patents — have regularly been the subject of media scrutiny over the past few years, due to the legal action which they tend to stimulate. Just several weeks ago, for example, Facebook and Yahoo became enmeshed in a complex patent battle over the pair’s respective patent holdings, in a move which could affect Facebook’s upcoming public listing.

Mullin has published a follow-up article on Ars Technica responding to the criticism of his article.

My opinion on this matter is that there is a great deal of hypocrisy evident in Australia’s technology community when it comes to the issue of the CSIRO’s Wi-Fi patent holdings.

Broadly, I believe that much of Australia’s technology community is against patents in general. The reason for this is that there have been many cases where patents have stifled innovation locally, with giant US corporations often holding a mandate on the development of technologies which Australians would like to innovate around. In addition, Australian society contains a strongly egalitarian streak, and the idea that innovation could be stifled by software patents, in particular, has always attracted a great deal of opposition here.

However, on the specific issue of the CSIRO’s Wi-Fi patents, many people in Australia are quite intractable. Without looking at the specifics of the case, which do demonstrate that the CSIRO very carefully planned its somewhat shaky patent attack for maximum global impact (why else would the CSIRO file its patent test case in a small court in the remote jurisdiction of East Texas against a small Japanese company, Buffalo Technology), so many Australians just assume that the CSIRO’s argument is completely valid, simply because it throws one back in the face of the US multinationals.

In general, I like to see free and open debate on every issue that has relevance to Australia’s technology community. However, in this case I don’t feel that most of those Australians who are debating this issue are debating it legitimately. The level of backbiting and vitriol hurled at those who dare to criticise the CSIRO for its wide-ranging lawsuit relating to a very standardised and widely adopted technology is simply astounding.

It looks to me like the CSIRO has definitely done some stellar work in aiding the development of the 802.11 Wi-Fi standard, and there is quite a lot of evidence that Wi-Fi as we know it today would not have been developed as quickly without the agency’s work. But I remain unconvinced that the organisation did $430 million worth of development work, or that the global tech vendors which implemented the standard unfairly took advantage of the CSIRO’s work to an amount of that value. In addition, it seems relatively logical that the CSIRO’s methods would, in time, have been discovered elsewhere, as an organic progression of the standard.

I guess what I’m saying is: We should remain open to the possibility that the CSIRO has gone a step too far here, and not be too one-sided in our discussion of what is a very complex case. In addition, I think we need to take a step back and question how legitimate the patent system really is, when it allows this kind of litigation to proceed.

Image credit: J Miller, royalty free


    • Yup. This.

      CSIRO didn’t “invent” wi-fi, nor have they ever claimed to. They developed – (and sensibly patented) – the form of RF modulation that is currently in the wi-fi standards.

      Ars Technica can eat a dick… :o)

      • “Ars Technica can eat a dick… :o)”

        This is precisely the kind of ridiculous criticism which I just highlighted in my article, Michael. Ars Technica wrote a legitimate article highlighting legitimate criticisms of the CSIRO’s actions. US courts may have found that the CSIRO is within its rights, but that shouldn’t stop the issue being debated.

        Your response? Saying that the publication which provoked such legitimate debate can “eat a dick”. Wonderful — and it perfectly demonstrates exactly the sort of ridiculous patriotism about the CSIRO’s Wi-Fi patents that the very same Ars Technica article highlighted as taking place in Australia.

        • I meant it purely as a tongue in cheek statement, hence the clown face… :)

          I don’t see this as a parochial issue either. It could be the CSIRO, or DARPA (US equivalent) – it doesn’t matter. Everyone has the right to defend their patent, regardless of the “nationality” of that patent.

          If a reputable technology site like Ars Technica want to turn it into a “battle of the nations”, it is they who are being “parochial”.

          Imagine if it was reversed, and an Australian company was using a patent held by an American company, and making squillions of dollars from it.

          It would barely rate a mention over there. It might cop a five line blurb at the bottom of page 40 of the Wall Street Journal.

        • “Ars Technica wrote a legitimate article highlighting legitimate criticisms of the CSIRO’s actions.”

          But an article wrapped in some of the most laughable jingoism it has been my misfortune to read. Joe Mullin bizarrely casts CSIRO’a WiFi patent as an Australia versus US issue. Any valid criticisms he may have get wholly lost in his pananoid xenophobia. That was the problem most people had with the article and the problem that most people questioned Ars Technica over, a question which received no answer.

    • “Primarily because the CSIRO patent didn’t involved patenting wifi, it involved patenting a multipath transmission method that is currently used as a part of wifi.”

      The Ars Technica article is clearly aware of that, and explains it in the article. Your point?

      • Strange response Renai, but regardless.

        The date of the patent application is what counts, and that was in 1992, prior to the IEEE creating the 802.11b/g standards. The article also says that “When the IEEE adopted the 802.11a standard in 1999—and the more widely-used 802.11g standard years later—the group was choosing CSIRO technology.”, so they agree that the IEEE did incorporate the CSIRO’s patents.

        The bottom line is CSIRO came to the table first with this idea, they tried to take it to market but couldn’t get a partner, the companies they tried to partner with then started creating products based on the idea that the CSIRO presented. I’m by no means an expert on patent law, but this protection of ideas is what I believe to be the reason why patents were created in the first place.

        • In this, as with everything about this case, there is debate. From the Ars Technica article:

          “All of the elements of the “unique combination” CSIRO proffered in court as a breakthrough weren’t merely old by tech standards, they were decades old. “Multicarrier modulation,” used in WiFi as OFDM, was described as early as the 1950s. Papers had been published on interleaving in the 1960s. Forward error correction, Intel’s lawyer told the Texas jury, “was used when NASA sent the Mariner mission to Mars in 1968.” Harris Semiconductor had actual working products incorporating these techniques by the 1980s and the company was selling its modems to the US military. The lead defense attorney for Intel, Robert Van Nest, even showed one of those Harris modems to the Texas jury during the 2009 case.”

          • Sadly if all the above is the case, and the CSIRO patent is effectively: “All of the things from before, but at the same time”, then it is juts as valid as any of the million other patents being used in the US which are: “What we did before, with a computer” or the new rage: “What we have done in the past, but using a Mobile Phone”.

            Obviously I don’t have specific examples, and even if I did many of them may in fact be slightly more than that. Either way, if the USA protects those patents, being abused by actual patent trolls, then the problem is not australias problem, it isn’t our patriotism for the CSIRO that is “at fault”, it is 100% a problem of the US Patent office. Fact is, it sounds like the technology “All at once” was a new idea. Just because mixing old ideas isn’t strictly speaking a “new” idea, according to the american court system it is (even if it is just a small part of the american court-system that believes it).

            The way I see it; at least we have a pretty good idea that the CSIRO will re-invest in future technology. God knows what your typical patent troll does, and it does sound like the CSIRO came up with *some* idea, or it would have been invalidated in court. Lawyers really do make good arguments.

            PS. If it wasn’t 430 million dollars worth of work, where did the 430 million dollars come from? In my opinion as an idea, *no* idea is worth 430 million dollars. Baked into a product, that sells in the trillions of dollars world wide? Could easily be worth 430 million dollars as a component part.

          • Yes, well there is a lot of uninformed comment on what CSIRO did or didn’t do. This is my uninformed take on what they did.

            The problem with indoors WiFi is multipath interference. This means the radio waves take different paths to get the to receiver. It is rather like sitting in a room with a PA system in a room with hard walls. It is often difficult to hear down the back because although the volume is loud the sound is a mush. What is happening is the sound waves are bouncing off the walls and roof, so a single note can arrive by many different paths. As a consequence they arrive at your ear at different times at different levels of loudness and so you don’t hear a single crisp sound. In fact if it is bad enough you won’t hear anything useful at all. WiFi has exactly the same problem.

            With radio there are several ways to encode the signal. Here encode means “vary so a binary 0 looks different to a binary 1”. Examples are varying the signal strength (ie amplitude) which is called Amplitude Modulation, the phase which is called Phase Modulation, and frequency which is called Frequency Modulation. Of these Frequency Modulation is the one that is less susceptible to multipath interference. This is because bouncing off things can change the amplitude and phase, but not the frequency. FM is what radio uses nowadays of course. It was an improvement over AM for the same reason.

            Frequency Modulation has its own problems. It is susceptible to fading, which could be thought of as what happens when you try to pass white light through a blue glass. The blue light passes through real well, but red and green get absorbed. So if you were using a blue light to send your message it would get through, but if you tried to use a different frequency like red it wouldn’t. (A similar thing happens with sound, where you might hear the base but not the treble.) A solution to that is to break up your information and spread it across lots of different frequencies. In fact you can do better and a similar use trick to RAID5 and spread it across more frequencies than needed. Then, just as RAID5 can loose a disk and continue, this means you can loose a few frequencies and the information will still get through. This technique is called Orthogonal frequency-division multiplexing, or OFDM, and it is what WiFi uses.

            As commenters have repeatedly noted, OFDM was a well known technique before CSIRO came along. But there is a problem. Now we have a whole pile of frequencies in use at one time, and you have to pluck each one out and decode it. It’s rather like listening to 100’s on instruments in an orchestra each playing a different tune, and trying to discern the tune the piccolo is playing.

            The technique for doing this was also known long before CSIRO came along. It involved breaking the signal down into its constituent sine waves – the idea being that any regular signal, no matter how complex, can be represented as simple combination of ordinary looking waves like you see on the surface of water. In fact this is what makes rough water rough. What we are looking at is a whole pile of simple waves such as you see when you drop a pebble into a pond adding up to become something complex. So what this technique does is unscramble the egg, revealing the individual simple waves that make up the rough water.

            So what is the problem? Well, unscrambling the egg just sounds complex, doesn’t it? In fact if you show a programmer the equation that describes how do this on wikipedia, he will see it involves a possibly infinite series of cos’s and derivatives using floating point. If you then tell him he has to do this in real time, with a 2.4 GHz signal, across not one but a whole bunch of carriers simultaneously and he would just smile and walk away. Today’s processors only execute a couple of instructions at 2.4GHz at best. So to implement OFDM on in silicon in a relatively straight forward way is impossible today, let alone back in 1990 when processors ran at 1/10th of the speed.

            No one had figured out a way to do it until CSIRO came along. Yes, OFDM and the basic math was all well known, but to actually implement it on silicon so it ran fast enough – no one knew how to do that. As Glen Turner said it is essentially a mathematical problem – using a whole pile of mathematical tricks to make it run faster. The person who did solve it, John O’Sullivan, had a head start. He was a radio astronomer. Turns out what radio astronomers do all day long is point radio receivers at the sky, and break what they see back down into is constituent frequencies. And what sort of organisation would have the breath of scientific disciplines under one roof to combine astronomy with microelectronic engineering? CSIRO is one of the very, very few. Which is probably why they discovered it first.

          • You’re right, I’m afraid. That is an uninformed explanation. There’s a link to a blog post I wrote, which describes in some detail exactly what the CSIRO WLAN patent is about, in the very first comment above (by Tezz). And, as a sidebar in that post explains, while their ‘best’ solution in fact uses OFDM because of its high spectral efficiency and convenience of implementation using the Fast Fourier Transform (FFT), this is expressly not part of what they have claimed as the invention.

            Computing the FFT is actually very efficient. The critical frequency is not the carrier (GHz range), it is the bandwidth (10’s of MHz), and it is now easily done with a general-purpose processor. Back in the 90’s it did have to be done in hardware, or (at a pinch) with a special-purpose DSP processor.

            Nowadays there are applications in optical transmission in which FFTs are being computed in hardware on signals with bit rates in the 10’s of GHz!


          • > That is an uninformed explanation.

            After reading your blog post it appears I used the wrong sources entirely when putting together a picture of what happened. Thanks for pointing it out.

  1. All patents are bad, but the label of patent troll should be reserved for organization with a primarily commercial focus, not government funded scientific organizations.

    It seems like a case of fighting fire with fire, not an expert, but the groups they are suing look like hardware manufactures who play the patent game, and its not they are resorting to the media induistries level of legal enforcment.

    Live by the sword die by the sword.

  2. Tezz hits straight to the point – they didn’t “patent wifi” as so much of the media has been reporting – they patented what became a small but important piece of wifi. The fact that it was based on combining prior knowledge, or the idea that someone else would have figured it out eventually doesn’t make it unpatentable. If that was the case, no patent would hold up to scrutiny.

    I think the incorrectly reported and overblown claim to “the whole of wifi” is one of the reasons so many US commenters are incensed when they learn that some Aussie government body they’ve never heard of is enforcing a patent claim. But “Aussie inventors sue over a little piece of wifi” doesn’t make as good of a headline.

    As for starting their campaign with a “patent test case in a small court in the remote jurisdiction” – I don’t think this is any evidence that their claim was flimsy (surely higher courts would have eventually rejected it if it was). Rather, this is just the type of shrewd legal strategy that pretty much every big company seems to use once the lawyers take charge of the process.

    • hey Andrew,

      your comment is accurate, but it doesn’t address the granularities of the CSIRO’s Wi-Fi patent or its enforcement action as the Ars Technica article does. The devil of the CSIRO’s actions is truly in the detail here.



  3. Also, some further info on East Texas, where the CSIRO filed its patent app:

    “An individual case often begins with a perfunctory infringement complaint,[21] or even a mere threat of suit, which is often enough to encourage settlement for the nuisance or “threat value” of the suit by purchasing a license to the patent. In the United States, suits are often brought in United States District Court for the Eastern District of Texas, known for favoring plaintiffs and for expertise in patent suits.[35]”

    • While we are throwing around Wikipedia links:

      “Between 2004 and 2011 the district presided over TiVo Inc. v. EchoStar Corp., involving the issues of patent infringement and contempt of court.”

      TiVo strikes me as more on the ‘legitimate innovator’ side of the fence – I think it’s a popular district for both trolls and non-trolls alike.

      My point was, trolling or not, the lawyers of patent holders tend to have little shame in increasing their chances by filing in an obscure jurisdiction, irrespective of the legitimacy of their claim.

      • So is your point that CSIRO should be regarded as a patent troll because their single case was filed in such a way that increased their chance of a finding in their favour? What’s their alternative? File where there’s the greatest chance of wasting government funds?

        I think the major flaw in your analysis is that a “patent troll” – by the Wikipedia yardstick you yourself introduced – is a ” person or company who buys and enforces patents against one or more alleged infringers in a manner considered by the target or observers as unduly aggressive or opportunistic, often with no intention to further develop, manufacture or market the patented invention.”

        1. CSIRO is not a company or person – its a government funded research body. If anything they develop even hints at profitability, they commercialise it ( a la things like Funnelback)
        2. It’s not “patents” in the normal troll sense, it’s a suite of patents related to one thing only
        3. As a scientific research body, there entire purpose is dedicated to “further develop”. If they stopped researching and focused on monetizing, I’m pretty sure that’d be picked up

        Having known and worked with a few CSIRO guys over the years, I’m fairly confident that this case was based on resentment over other people making money off their work (which is exactly and exclusively what IP law should be about, and is in fact why IP law was first created). If this was info that was being given away for free, no problem

        • I’m not saying that the CSIRO is a stereotypical patent troll — clearly they’re not, although I have significant doubts about the value of much of the stuff in the tech space in Australia that they’re “researching”. However, I think there also could have been a better way for them to handle this than to sue the entire world.

          • When this issue first came up a week or so ago, there was plenty of evidence being mentioned that the CSIRO had spent years to trying to negotiate licenses for the patent with companies that are being sued over infringement.

            They were fobbed off. Once that avenue was killed off they’ve resorted to “suing the world” as you so gracefully put it.

  4. I think some of the issue comes from patent trolls that exist as pure entities to suck value from third party organisations for profit without any vision or aim towards re-investment in new products or research. There are plenty of cases of organisations that have invented something with overly vague patents in the software arena (the plugin patent is one example) that are really what people associate with troll behaviour. Yahoo is now at the point where they’ve been searching for direction and have devolved into the state of using their extensive patent suite to try and extract some funding out of Facebook. RIM will likely go that way unless someone buys them first. There are plenty examples of this form of behaviour where it feels more likely money grabbing than honest fair go infringement.

    CSIRO’s case is interesting because as an organisation it’s dedicated to research and increasingly commercialisation. The original case resulted in the CSIRO putting money back into do more research as opposed to paying out some lucky shareholders instead of using the payout to further work in the industry more.

    I feel that the strong counter-reaction is that CSIRO did invent something that is novel and unique at the time and while some of the ideas might have been out there, they took the idea and made it in part a reality. The gap is knowing I need a hammer like thing and inventing said hammer. Sure I can use a club as a crude hammer but it isn’t quite the same. Likely without the research the club would have been used but instead we get hammers to apply to our nails which makes everything go a little smoothly. After a while someone else would have found the hammer because they likely would have done similar research but at that point they hadn’t.

    The fact that CSIRO is not only committed to re-investing with more research and developed something more in the hardware space than pure software space is perhaps why there is such a strong reaction. The distinction between hardware patents and software patents is really where the line is drawn and in the eyes of many this is more hardware than software with an outcome that furthers science not someones back pocket.

  5. CSIRO is not a patent troll simply because they use a broken system. Suing in the most favourable place, against a decent sized, but not too big opponent, is a standard strategic move. MPAA picked iiNet for the same reason. CSIRO applied for a range of patents, for an invention that didn’t exist at the time, and overcame a key issue in progressing science. They defended the validity of those patents against numerous attacks (itself an issue with the patent system). Then they cut a deal.

    CSIRO is not a patent troll simply because they didn’t commercialise themselves. The basic concept of a license allows a company to legitimately sell IP they don’t own. CSIRO attempted to licence their invention, but others just ripped off their invention. They took them to court and won, then other companies decided to pay the researcher instead of their lawyers.

    It isn’t even like CSIRO made much money. $430M, over 20 years, for an invention included in pretty much every wireless electronic device on the planet? The cost per device to the inventors can be measured in microcents.

    Imho the patent system is broken. It works fundamentally contrary to its intention by adding to the cost of innovation. But on this one occassion it worked to give a research company a limited right to cash in on their research in order to fund more research. If it is hypocritical to enjoy the rare occassion that the system worked, because I’m Australian (or to summarise opinion on Ars more correctly, not American), then I just don’t care.

    • “CSIRO applied for a range of patents, for an invention that didn’t exist at the time, and overcame a key issue in progressing science.”

      Except this is a substantial mischaracterisation of the situation.

      How about this for a replacement sentence:

      “CSIRO applied for a range of patents, relating to a very small segment of a much wider invention being developed by many different companies simultaneously, using many technologies which were already well established in the market to do so, and overcame a key issue in accelerating a product to market.”

      You can see in this small example how much of the debate about CSIRO’s “invention” is being framed in a way which does not take into account the reality of how this situation evolved. The CSIRO was on of many organisations working on many, sometimes different small segments of a much wider industry standard. To develop its solution, it used many existing technologies. It did not “invent” anything — it innovated, and in a way which others would likely have reached anyway.

      “It isn’t even like CSIRO made much money. $430M, over 20 years, for an invention included in pretty much every wireless electronic device on the planet?”

      For its investment — which was, I believe, only a small handful of staff, it does indeed look like the CSIRO made a substantial return. And sure, its tiny algorithm was used in pretty much every electronic device on the planet. But the CSIRO did not manufacture those products, and had only the tiniest impact on how they were made. I don’t think it should be pulling licensing values from all of those devices.

      I would also be interested to know how reverse engineering could have come into this situation, and whether Intel and the other tech giants could have found their way to a similar solution to the CSIRO’s without having to pay the exhorbitant licensing fees. You know, like tech companies in Silicon Valley have been doing for three decades now.

      • All inventions and innovations are built on pre-existing technology. Otherwise everyone would, literally, re-invent the wheel. Every time. This is irrelevant to the debate.

        If the problem is big enough, there will always be multiple people working to solve it. Basing an argument on the unsupported assumption that they found a solution someone else ‘would likely have reached anyway’ is also too far fetched. Most inventions look obvious in hindsight. Especially after you have used it every day for 5 years (about the average time a patent takes from application to grant) How many people are currently working to cure cancer? It is worth the investment in research, because of the value of the solution. The number of people working on the problem is irrelevant.

        It also doesn’t matter if anyone else finds a different solution. The point is that CSIRO *found* a solution. Others were then taught how it worked, offered to licence it, but chose to implement the invention without paying. For that, CSIRO deserves their win.

        • I agree with much of what you’ve written here, Matt, but suing the world for hundreds of millions of dollars is not the answer … there has to be a middle ground.

          • There is – you cut a deal. CSIRO did exactly this – their recent ‘wins’ have all been settlement agreements, not court imposed awards.

        • Its not that they reinvented the wheel its that they decided to patent what was obviously concepts already discovered..decades old. Then to turn around and claim you invented wif and behave like a patent troll by threatening companies left and right until you get “go away” money, the label simply fits. Exploiting a system based on questionable “discoveries” is the very definition of patent trolling, only nationalistic blindness can make anyone say different.

    • A couple of things:

      Richard Chirgwin’s points in this article are mainly correct. However, he doesn’t address the fundamental issue here — that much of this technology was already known, and that the CSIRO’s role in the development of the Wi-Fi standard appears to have been only a partial one at best. As I wrote above:

      “CSIRO applied for a range of patents, relating to a very small segment of a much wider invention being developed by many different companies simultaneously, using many technologies which were already well established in the market to do so, and overcame a key issue in accelerating a product to market.”

      Secondly, Chirgwin notes in his article that the CSIRO didn’t attempt to licence its technology on fair terms, but actually on very unfair terms — $4 per device. In a world where Wi-Fi, which the CSIRO can only claim sovereignty over a minor part of — is only itself a minor part of a device, $4 per device starts to look crazy. Something in the low cents would have been much more reasonable, and would have allowed the CSIRO, a public organisation, to make a decent return and continue to fund its operations without making the whole tech industry go crazy over its prices.

      But the industry *did* go crazy over the prices, and the CSIRO then concocted a carefully crafted legal route instead of negotiating a fairer rate … fairly nuts.

      Lastly … “nationalistic dog-whistling and misinformation”? I’m sorry, but it is Australians who are making this into a riot of patriotic fervour — not Ars Technica.

      It is staggering to me that so many Australians who comment on this issue automatically take the CSIRO’s side … even though most agree patents hold innovation back … even though they would slam a multinational who took the same approach of suing the world … even though the CSIRO didn’t have to set such a high price to licence its technology … even though the agency has clearly used the law to its utmost to reap the maximum return.


      • Lastly … “nationalistic dog-whistling and misinformation”? I’m sorry, but it is Australians who are making this into a riot of patriotic fervour — not Ars Technica.

        I fundamentally disagree, this has nothing to do with being patriotic, this about the patent process and in this instance how it actually works.

        Go and have a read of this slashdot article comments .. .. note I say comments, you’ll find there is a reoccurring theme saying that as far as Ars Technica is concerned this is the worst technically, and most biased piece every on that site, and a lot of people are shocked the site even published it in the first place.

        To be complete honest Renai I’m little shocked that you’re basing your entire opinion on the Ars Technica piece, and seem to be ignoring every other point of view, very unlike you.

        • I’m not ignoring every other piece but the Ars Technica piece, and as I mentioned in my article I’ve read significant criticism of the CSIRO wins in other places. In particular I remember an article by The Economist which dissected the situation pretty precisely.

          This is an issue which has been going on for years ;)

          I guess what annoys me is the massive difference in the coverage between the Australian press and the global press. Globally, the coverage is pretty balanced — generally stating that the CSIRO has a case, but that blanket patent lawsuits generally don’t help the industry. In Australia, and Chirgwin’s piece is a great example of this, the press just constantly champions the CSIRO victories without asking deeper questions about them.

          I’m interested in objective thinking about patents — and the nationalistic fervour from Australians commenting on this case annoys me ;)

          • I think the nationalistic thing is because the article you are quoting opens by claiming “US consumers will be making a multimillion dollar donation to an Australian government agency in the near future, whether they like it or not” and goes on to say “CSIRO began boasting to the Australian press that WiFi was a homegrown invention”, the whole article smells of American patriotism, bias, and FUD.

            You probably should have picked a better article to use as your example, for example one that claims “CSIRO (commonly pronounced “si-roh”)” should have been a dead giveaway for lack of research.

      • Renai, I haven’t read the Chirgwin piece, but somebody seems confused between ‘fair’ and ‘reasonable’. Although there’s no legal definition, fair is more about the licencing terms not being anti-competitive, whereas reasonable is the price. I agree that $4 per device does sound a little… ambitious, but obviously the court case was in response to people not liking the price. Given that one randomly Googled news article said 500 million 802.11n devices (cbf finding 802.11a/g numbers) had been sold by 2009, and this was expected to increase to 2 billion by 2014, I would suggest that the courts have granted a MUCH lower price for the patent suite.

        As for the East Texas court, its not just the odds of plaintiffs winning (88% vs 68% elsewhere, apparently), the ability to have a relatively fast trial and use lawyers from anywhere in the US are positives for companies on both sides.

  6. Thanks for a brilliant article Renai, and for once, we have in Australia, a proper analysis of this issue.

    I’m quite familiar with how patent trolls work, and CSIRO’s whole strategy reads like it was adopted from the strategy books of Intellectual Ventures.

    I’m sorry, if CSIRO wanted to be involved in the creation of Wifi, they could have been part of the process through IEEE. The fact that they didn’t, and then got a patent on a technology that had been in wide use for decades in the industry, is the definition of a patent troll.

    Yes, the US patent system is fundamentally broken, and the CSIRO has very cleverly exploited it to its benefit, but that doesn’t suddenly make the CSIRO the inventor of Wifi.

  7. I don’t buy the analysis — that ” there is a great deal of hypocrisy evident in Australia’s technology community”.

    First of all, it’s a community, not one person, so it’s hardly surprising that different people hold different views.

    Even if one person does believe that the patent system is abused regularly and needs to be reformed, it doesn’t mean that they are against patenting genuine research — exactly what’s happened here.

    The reality is that if you’re going to patent something — and if you’re a research organisation like CSIRO, that makes sense — you HAVE to do things like “carefully [plan a] somewhat shaky patent attack for maximum global impact.” That’s just how things work these days, and in my experience is extremely MILD behaviour, not troll-like.

    In other words, while CSIRO’s behaviour might look aggressive to you, it’s not compared to the majority of other “respectable” organisations, much less the true trolls.

    • Suing every major technology vendor for hundreds of millions of dollars is not “extremely mild behaviour”, no matter how you define it ;)

      • The CSIRO didn’t sue everyone though, most tech companies involved had already come to an agreed price, the only suing was against a handful of US based tech companies who refused to comply.

  8. Just a thought, and I realise I’ve no backing for this, but could this suit have come about precicely because all other attempts have failed? I can understand corporations suing on a drop of the coin, but perhaps the CSIRO had already tried to negotiate or otherwise settle this issue, and this is the last resort they have.

    Just an idea, thats all.

  9. I have to admit – I played a part in the comments, and now a week on and reading multiple angles, I still stand by my infuriation with the article. It’s not that it stems from a patriotic bias or that he said CSIRO is “commonly pronounced sih-roh”, but rather the “us vs them” tone the article takes. Ars Technica is a well respected site that writes for a global audience, even if it most of its focus is on US-centric ones, and it should act as so.

    You yourself show much much integrity in your work Renai, your new format of seperating opinion from news is so that you can allow your readers to form an opinion, not have your opinion subtly shoved down someone’s throat. And that’s exactly what that Ars article does – it chooses its own facts, disregards others and uses language that paints CSIRO, the Australian Government and Australia as a nation in a negative way to a global audience. If you want to write an angry blog post, at least indicate it as so.

  10. Thanks for the link to my version of the history, Renai. However, if I may be permitted a small correction, this was not written in response to Ars. It was written and published at around the same time, and before I had seen Mullin’s piece.

    Other commenters have provided links to my post in response to the Ars article, and my analysis of the CSIRO patent itself, for which I thank them also.

    There is a further issue with the Ars article (and, I suspect, with the earlier Economist piece as well) which does not seem to have been noted, by me or anyone else. The problem is pretty clear if you look at Mullin’s sources. Quoting directly from his article:

    “Defense lawyers have been able to point to some serious flaws with CSIRO’s version of history…”

    “Forward error correction, Intel’s lawyer told the Texas jury, ‘was used when NASA sent the Mariner mission to Mars in 1968.'”

    “The lead defense attorney for Intel, Robert Van Nest, even showed one of those Harris modems to the Texas jury during the 2009 case…”

    “Hitoshi Takanashi, CEO of NTT, testified that not only did his company not steal CSIRO’s idea back in 1997, he’d never even heard of the organization. Stephen Saltzman, an Intel Capital official testifying at trial, said that outside of Radiata and its ex-CSIRO employees, engineers didn’t take the idea of a brand-new WiFi patent describing CSIRO’s techniques that seriously.”

    None of this is a sound basis for a balanced account. It is material that was presented in court, or provided to the media, for the express purpose of defending against CSIRO’s claims. The legal arguments provided by attorneys, in particular, are not ‘facts’. They are expressly partisan characterisations of alleged facts. The function of an attorney in litigation is to advocate on behalf of his or her client. The US legal system (like Australia’s) is adversarial, i.e. each side is supposed to argue their opposing cases as strongly as possible, and it is for the judge or jury to decide which argument they find the most persuasive. That’s just how the system works, for better or worse.

    While Mullin also cites testimony from the CSIRO researchers, he cherry-picks statements which, pulled out of context and isolated from any considerations of how they would apply to the legal issues in the patent case, appear to support the defendants’ position. And you will note that he does not quote any of the corresponding legal arguments that were presented on CSIRO’s behalf.

    Throughout the entire process — form the early licensing discussions right through to this final settlement — CSIRO has been up against some of the biggest and most powerful tech corporations in the world, such as Intel and NTT quoted by Mullin. Those corporations never set out to play fair, they set out to win, and I think they expected this would be easy against a comparatively small fish like CSIRO. They have never been shy of using the media as a channel to argue their case, and clearly the level of media interest exists in the US because of the involvement of the big corporate names, not because of CSIRO.

    In short, US media coverage of the case has always been biased. The legal and technical issues are complex, and journalists have necessarily relied upon information and guidance from ‘sources close to the case’. All of the evidence suggests that those sources are wholly, or predominantly, on the side of the defendants.

    The big question is not why CSIRO has not spoken up in response. Keeping one’s own counsel about matters which are sub-judice, or which are the subject of confidential settlement agreements, is the advisable course of action. But the other side has been leaking like a sieve, and the question is why the media has lapped this up and regurgitated it seemingly without question.

    The biggest trolls in this case are not, and never have been, of the patent variety!


  11. Reading the comments here Renai, your view seems to be because CSIRO patented one small part of the WiFi puzzle, and then perused that patent for all it was worth they are a patent troll. Go down that road and everybody who invented something, patented it, then vigorously defended it is a patent troll.

    There is a problem with this view – this is how patents are meant to work. So if you don’t like it don’t blame the participants, get rid of patents. This is not in the least because as far as I can tell, CSIRO was doing exactly what the inventors of patents intended. They found a problem, gambled money on solving it, produced a working prototype, and then sought to make money from their invention, got lucky and are making far more money than they invested. (This is a gamble that doesn’t always pay off. As you point out when you say “I have significant doubts about the value of much of the stuff …”, you have to kiss a lot of toads to find a prince. Those princes have to pay for time invested in toads.)

    Encouraging this sort of activity is why the patent system was put in place originally. And because they successful in doing that, we have things like mp3, ARM (who like CSIRO is purely an IP organisation that doesn’t make things), and the succession of h26x codecs (the latest in development being h265 which gets roughly a 50% improvement on h264 – ie 1/2 the size), and just about every new drug produced in the last 50 years.

    Most people recoil at the thought of not having those things, and so most don’t like companies such as ARM and CSIRO who do invest the funds they earn in inventing new things being branded patent trolls. You characterised this reaction as an Australian one. I disagree. The only reason the big outcry came from Australia in this case is the rest of the world doesn’t know what CSIRO does. Had Ars tried to brand NASA or ARM as patent trolls then the outcry would have come from everybody, and the 300 comments attacking them would have been measured in the 1000’s, not 100’s.

    As it is, I think it is obvious Ars has seriously miss judged their audience. Ars editorial stance on patents is black and white, and they stand in firmly in the black corner. (Nevertheless, I am cynical enough to think they would never attack their own sacred cows like NASA – and yet NASA does own and sue over patents. Their mistake was to think they could get away with attacking ours.) But Ars’s readership is fairly technical, and for all the shortcomings of the current patent system most of them the idea that someone should be rewarded for their bright ideas is a good one. The reason is pretty obvious: their readership is one of the groups of people who are paid to come up with those ideas.

    That said, no one disagrees the patent system is currently broken. But I thought it was common knowledge who broke it, and it wasn’t research organisations like CSIRO. It is not even the US troll lawyers or Microsoft and their 1000’s of trivial patents. It was the US politicians. The US Patent Office ran out of money and so there was a huge backlog of patent applications. There were many possible solutions to this. Instead of choosing one of them, the US politicians first came up with a slogan along the lines of “Get out of the way of innovation”, and under than banner instructed the USPTO to save money by not examining patents. The official policy change was called the “accelerated patent examination procedure”. And so opened the flood gates to patent hell.

    With that in mind I’m inclined to agree with the bulk of the posters on Ars. Attributing any of this mess to CSIRO is at best misguided, and they deserve the spirited defence they received when Ars tried to do just that.

  12. It was a click bait article by a writer who rewrote history to suit his own world view and got some well deserved criticism as a result.

    bite me

  13. I responded to the Ars Technica article, and I will respond to this. An Australian company followed US-defined rules to defend its own intellectual property, and Ars responded by writing a clearly racist rant. The author (with whom I also had some personal correspondence) did respond to complaints by writing another article which lost the racism and was better journalism.

    I couldn’t be bothered going into the specific merits of the CSIRO’s case, but my concern at the time and my continuing concern was the reference to kangaroos and “theft” in the original article. It was unprofessional.

    On the subject of patents more generally, I would be happy for them to disappear – the patent system is patently failing. But complaining about a foreign company that takes advantage of your laws is rather rich.

  14. I don’t know where you got the odd idea that a patent holder has to be involved with the use of its technology in order to claim patent royalties. The notion is nonsense — Porsche isn’t involved with the design of every car which uses its patented technologies. Similarly, CSIRO didn’t need to be involved with the use of its technology in 802.11g and 802.11n.

    In my view the CSIRO patent is exactly what a patent is meant to be. A clear description of a large advance in the technology. I do suggest you read the patent — your description of it being “shaky” doesn’t seem to me to be justified. That advance was incorporated into the textbooks describing the field, and eventually into the 802.11 standard. If CSIRO couldn’t win with this patent then the patent system would be failing its original purpose, leaving only the trolls and company-v-company warfare.

    As for CSIRO’s legal tactics, I am not sure why you are complaining that CSIRO made a considered choice of an optimal strategy to force payments for their patent. And yes, there is some irony that the optimal path for patent enforcement has been so well marked by patent trolls. But really, you’re not suggesting they should have made a hurried choice of a non-optimal strategy :-)

    My personal view is that the whole class of patents into which the CSIRO patent falls is nothing other than applied mathematics. But that patent-law horse bolted a generation ago and coaxing that horse back into the stable isn’t going to happen anytime soon. In the meantime, you’ve got to deal with the patent system as it is, not as you want it to be (ie, even Red Hat holds software patents).

    Your closing paragraphs are really a call for reform of the patent system. Your proposed right to independent invention would have lead to licensees choosing the cheapest license for the patent from the various independent inventors, not the removal of patents on that technology together. CSIRO were very ahead of their time in this patent — I’m not sure there would have been any independent invention prior to the topic entering the textbooks.

  15. It did go a bit feral, I think you’re right about that. Still, the tone of the original article was unnecessarily snarky.

    Must be interesting from Ars’ point of view – clearly they didn’t realise they had so many Australian readers. Seems their audience isn’t quite who they thought it was…interesting to see if that realisation has any effect.

  16. I partook in those debates, and I strongly defend the actions of the CSIRO. Just mostly on principle, it’s about time the US saw their shitty laws used against them. I don’t agree with patent litigation, but maybe seeing how the US themselves are vulnerable, may lead them to doing away with the bat-shit insane IP laws they forster at the moment.

  17. The CSIRO is not a patent troll, Kia Silverbrook is. I worked with him long enough to know.

  18. This issue is a typical example of Australian nationalism at its worst. As an American living in Australia I can account for the anti-American sentiment that is running rampant in Australian media and its people. Renai, thank you for explaining this issue in an unbiased and informative manner. Australians are too emotionally involved in their national pride to look at the issue with the proper perspective.

Comments are closed.