25 April 2013

EFF Off the Mark With Latest ‘Abuse’ Allegations

Invisible man with glassesThe Electronic Frontier Foundation’s (EFF) Daniel Nazer and Julie Samuels have a tragic tale to tell, which they hope will stir the souls of the online community to action.

In a nutshell, the narrative goes something like this…

Once upon a time, a few innovative, dedicated entrepreneurs started a Little Red Riding Company called Ditto Technologies, Inc, based on the great idea of letting people try on eyeglasses in a virtual environment by using webcam images to build a 3D model of their heads and faces.  Within two years, their hard work in building an appealing product was starting to pay off.  Ditto had received funding to build its business, employ new staff, and continue to improve its product.

But then along came a Big Bad Wolf Company, called 1-800 CONTACTS.  Although the Wolf was already in the business of selling prescription eyewear (i.e. contact lenses) online, it had no product that competed directly with Ditto.  Yet the Wolf had purchased a patent from a ‘defunct company’ which (it alleged) covers Ditto’s virtual try-on service, and was now suing California-based Ditto in the distant Land of Utah.

The cost of litigation (and the possibility of losing) threatens Ditto’s very existence, unless an army of diligent Woodcutters can be rallied to find prior art capable of invalidating the Wolf’s patents!

Poor little Ditto.  But is it possible that there is another side of the story?  One in which the Big Bad Wolf is just misunderstood, and the Little Red Riding Company might be, to some extent, a victim of its own (and its investors’) mistakes?

An Alternative Tale

Once upon a time, in the days when years just barely began with a ‘2’, a group of California-based inventors with a company called Geometrix, Inc, had a great idea.

Geometrix was in the business of 3D modelling, and had developed scanning software which could be used to capture 3D models of objects ranging in size from jewellery to motorcycles.  In 2000, Geometrix had formed an alliance with another company, called Kaidan Incorporated, which was a supplier of photographic virtual reality (VR) solutions for the Internet.

The Geometrix inventors’ idea was to apply the company’s 3D modelling technology to human faces, in particular to enable people to try on eyeglasses in an online virtual environment.  In 2001, Geometrix filed a patent application for this invention.

Sadly, over the following years things did not go so well for Geometrix.  Having no doubt started out, as many Silicon Valley companies do, on the back of the dreams of a few innovative, dedicated entrepreneurs, it failed to survive (alas, also the fate of many Silicon Valley startups).  As for the poor, orphaned, patent application, it had at least managed to stay alive.  It issued as a patent in 2006, but as one of the few potentially valuable assets of the failed Geometrix, and after being briefly fostered by biometric security solutions company ALIVE Tech, Inc (which is also now defunct) it found itself shuttled around between Geometrix’s creditors (including the Silicon Valley Bank) for a number of years.

Eventually, however, the plucky little patent – for an idea which had clearly been ahead of its time – found itself in the hands of Big Stage Entertainment, Inc, described as ‘a media technology company that enables the projection of self into digital life.’  At last it seemed that the patent had found a home where it could be commercially exploited and valued!  But then Big Screen was acquired by automated avatar creation company Image Metrics, Inc, and the patent was once again on the move.

For its whole life, the patent had been passed from hand-to-hand by custodians who obviously saw its worth.  Yet none of them really wanted to put it to use in building value around the patent’s one true calling as an online eyeglass try-on platform.

Eventually, however, a company started up that really did want to do just what the patent was always designed to cover, and whose founders believed that with advances in technology, and the application of their own smarts, they could now make it work as a viable business concern.  That company was Ditto Technologies, Inc.

At around the same time, however, established company 1-800 CONTACTS was looking to grow its online contact lens sales business, by branching into eyeglasses.  In mid-2011 its domain glasses.com stopped redirecting to 1800contacts.com and became a separate site selling prescription glasses and eyeglasses.

It is not a stretch to suppose that 1-800 CONTACTS was looking to develop better ways to sell eyewear online, or that it had quite likely commenced development of its own ‘virtual try-on’ service.  So when 1-800 CONTACTS found out about Ditto’s service in early 2012, this would have come as something of a blow to its business strategy and revenue projections from its nascent eyeglasses business.

Lacking any protection for the intellectual property it was still developing behind closed doors, 1-800 CONTACTS went looking for some means to protect its own technology investment.  Its research led it to discover the now middle-aged Geometrix patent, languishing with Image Metrics.

An approach to Image Metrics, some negotiations, and 1-800 CONTACTS had acquired the patent it needed to consolidate its place in the online eyeglasses market.  And it had poor Ditto in its sights…

Could It Have Turned Out Differently?

Of course, I have engaged in a little speculation in the above version of the tale.  The USPTO assignment records for the patent are a mess – the result of various owners failing, over the years, to properly and promptly record changes of ownership.  Much clearing up appears to have been done to ensure clean recordation of title after the patent was acquired by 1-800 CONTACTS.  There is also no way to know for certain whether 1-800 CONTACTS had already invested any funds or effort in development of a virtual try-on service before discovering Ditto.  But there are, as I shall explain, good reasons to believe that this was a likely scenario.

The really big question, to my mind, is what the entrepreneurs at Ditto and, more importantly, their investors, were thinking in not ensuring that they had freedom-to-operate – at least before going live, if not before expending any significant resources on the venture.

The Geometrix patent was just sitting on Image Metrics’ shelf when Ditto got started.  Its title is Interactive try-on platform for eyeglasses, and its abstract contains the keywords and phrases ‘3D representation’, ‘glasses’, ‘face model’, among a number of other terms clearly relevant to Ditto’s technology.  It is almost inconceivable that even the most cursory professional patent search would not have identified this document.  Indeed, this is the only patent uncovered by the most basic USPTO search imaginable – for the words ‘3D’, ‘glasses’ and ‘face’ in the abstract.

Ditto’s founders really should have known about this patent from the outset.  If they did not, then their first investors should have discovered it in the course of due diligence.  Either way, they should have sought to acquire it before leaving ‘stealth mode’ and going live.  The patent was clearly available for purchase, and while it may be a tragedy for Ditto that it fell into the hands of 1-800 CONTACTS, it is a tragedy that might have been avoided.

The same could be said for 1-800 CONTACTS.  Similar rules apply when investing in a new area of business as when starting a new business from scratch.  You need to do due diligence on the proposed business strategy, to ensure that it is viable, and that there are no obvious show-stoppers – such as a blocking patent, or a potential competitor with a head-start.  It appears that 1-800 CONTACTS failed to do its research properly!

I do think, however, that 1-800 CONTACTS did, and does, have, as it now contends [PDF], an ongoing investment in development which it is seeking to protect by its acquisition of the Geometrix patent.  Had it been the case that the company had developed no technology of its own when it discovered Ditto, the commercially sensible course of action would have been to seek to acquire Ditto itself.  If the price was right, this could have been a great outcome for Ditto’s founders, staff and investors.

Lessons in IP Management

While Geometrix may have failed (as the majority of ventures do), its eyeglasses try-on patent has lived on.  When the company did fail, it would have had few, if any, tangible assets.  The intellectual property developed during its limited lifetime would be all that the company’s investors and creditors had to show for their faith and risk-taking.

It is therefore good that there is a ‘secondary market’ for intellectual assets, such as patent rights.  This market enables intangibles to be bought and sold, to recover costs of development, and to enable new businesses to seek to capitalise on inventions when originating inventors and companies fail – for whatever reason – to execute successfully.  Ideas may be free, but bringing them to market can be very expensive.  Without capital, it will not happen.  And without assets to back capital, there would be less investment to make it happen.

Had Ditto identified and acquired the Geometrix patent, it would be a more valuable company today.  Not only would it have deprived a competitor of the opportunity to acquire the patent, it would be holding all the cards against 1-800 CONTACTS, and anyone else wanting to set up in direct competition.  Investors might well be looking at a very profitable exit strategy, by way of an industry sale either to 1-800 CONTACTS itself, or to another larger prospective competitor.  Alternatively, Ditto might wish to remain independent, but generate new revenue streams through licensing.

As for 1-800 CONTACTS, to my mind it got lucky.  Having failed to establish its own IP position on its new venture, and finding itself with a faster-moving competitor at a critical juncture, it landed on its feet when it found the Geometrix patent available for acquisition.

I have no doubt that Ditto does not deserve the bad luck that has befallen it, and 1-800 CONTACTS probably does not deserve its good luck.  However, the fact is that luck is a factor in the success or failure of any business venture, although to some extent well-managed enterprises can make their own luck.

The story of Geometrix, Ditto and 1-800 CONTACTS is indeed a sad tale, though not for the reasons suggested by the EFF.

Why is the EFF Even Taking Up This ‘Cause’?

On its ‘about’ page, the EFF says:

…technologies are transforming our society and empowering us as speakers, citizens, creators, and consumers. When our freedoms in the networked world come under attack, the Electronic Frontier Foundation (EFF) is the first line of defense.

This is all very well, and there are many areas in which advocacy for electronic privacy and freedom is essential.  There are certainly cases in which the balance between private intellectual property rights, and the public interest in freedom of expression, is under threat. 

However, last time I looked there was no fundamental natural human right to success in business.  And that is really what this case comes down to – a group of inventors at one company came up with an idea in around 2000 which was ahead of its time, and now that its time has (possibly) come, there are multiple companies seeking to exploit it commercially.  None of them can claim to have been the first to invent virtual try-on of eyeglasses.  Indeed, it appears that they are all at least a decade too late for that.

At the same time, this is not obviously a case in which a ‘bad’ patent is hampering innovation, or denying a service to consumers.  Whether it is Ditto, 1-800 CONTACTS, or some other company, which provides the virtual try-on service, it will be there for consumers to use, should they choose to do so.  There is no ‘troll’ here, with no skin the the game itself, trying to extort money from every player in the business.  And the story would be very different if Ditto had acquired the Geometrix patent, rather than 1-800 CONTACTS having done so.

There is also a good argument that this case illustrates how the patent system helps to foster innovation.  Commercialising technology costs money, and capital comes at a price.  A company with no assets or collateral cannot raise the funds necessary to succeed on a large scale.  The Geometrix patent has survived because it is an asset capable of being traded, and thus of practical value to lenders and investors, as well as to operating companies.

At the end of the day, the dispute between Ditto and 1-800 CONTACTS is of a commercial nature.  It is about whether, and how, each company manages to succeed and grow, provide services to consumers, and deliver value to shareholders and investors.  It is hard to see how this is a ‘cause’ worthy of the attention of the EFF.

Crusading Against ‘Software Patents’

However, the EFF is on a crusade against ‘software patents’.  Julie Samuels is a Staff Attorney at EFF who holds something called ‘the Mark Cuban Chair to Eliminate Stupid Patents’.  In this role, it is (apparently) her job to advocate the position that patents are, as a general proposition, impeding innovation within high-tech industries.

On his IP Watchdog blog, Gene Quinn recently expressed the alternative view that ‘Mark Cuban is an Idiot, Patents do NOT Impede Innovation’.  Mark Cuban is a very rich man, so it is possible that he is not an idiot.  However, beneath the hyperbole Quinn’s main point is valid.  He quoted Chief Judge Rader of the Federal Circuit, explaining that:

The patent system has a narrow focus. It is not a consumer affairs program. It is not a manufacturers’ guarantee compliance program. It’s not a competition program.  It has one objective, summarized well by the Constitution: promote the progress of science and the useful arts. It’s there to create more investment and more incentive for innovation and invention. The things that the patent system is criticized for is not its job.

Quinn then went on to say that:

Indeed, there is nothing wrong with the patent system. There is litigation abuse, and if district court judges would look behind the curtain to see the litigation blackmail that is going on and do something about it the problem would disappear. The problem isn’t that the courts don’t have the authority they need to stop abuse. The problem is they don’t use the authority they have to stop litigation shake-downs. This is NOT a patent system problem. It is a litigation problem.

The EFF is in a position to lobby for a solution to this litigation problem.  Instead it is directing its time and energy to a case which has nothing in common with the real abuses being perpetrated by the bad actors widely known as ‘patent trolls’.  The only plausible explanation for this is that it has lost touch with its foundational principles of protecting fundamental human freedoms and rights in a technological world, and is instead being distracted by tilting at ‘software patent’ windmills.


A patent is not automatically ‘stupid’ because it happens to be directed to a computer-implemented invention.  Nor is it ‘stupid’ because it is asserted against a poor little startup entity.  Would the same patent be less ‘stupid’ if it were owned by the startup, and asserted against an incumbent like 1-800 CONTACTS?  I cannot help thinking that, for the EFF at least, maybe it would be.  But if the ‘stupidity’ of a patent depends upon who owns it, or who devised the invention, then it cannot be the patent system that is to blame.  It is always the same patent.  It is just who uses it, and how it is used, that differs.

Is anybody funding a Chair for the Elimination of Stupid Advocacy?  If so, please let me know – I would like to apply for the position!


arachne said...

This case highlights that the currently available term of a patent (uniformly 20 years, regardless of field of technology) might not be best serving the intentions of the patent system. Patents should give an advantage to the originator of the invention if they (or their licensee/assignee) become the 'first mover'. Patents should not merely encourage speculation on what might be practical only 15-20 years in the future (even though currently possible).

A hypothetical for you: If you were designing a patent system from scratch, what would be the term of a patent?

Mark Summerfield said...

Your hypothetical assumes that I agree with your first proposition, however I am not sure that I do.

The patent system currently does not encourage pure speculation. The specification is required to provide an enabling disclosure, i.e. a sufficient description of how to implement a working example of the invention. The fact that the inventor may be unable (for whatever reason) to describe and/or implement a commercially successful embodiment is not the concern of the patent system.

Despite common assertions that the 20-year term of a patent is too long for many fast-moving areas of technology, I actually think that the most valuable ideas probably require on the order of a decade to get to the point at which the technology and the market are ready for widespread commercial success, i.e. viable scaling. There are all sorts of reasons why the originator of an invention may fail commercially.

To give another example, Palm filed a patent application in 2000 disclosing what is now pretty much the standard smartphone architecture, i.e. a baseband processor to handle the communications processing, and an applications processor to handle the UI, apps etc. It was granted in 2005 as US6,976,217. As we now all know, efforts to commercialise this idea struggled until Apple launched the original iPhone in mid-2007.

We know that the failure rate of startups is very high, although whether it is 60%, 70%, 80%, 90%, 95% or whatever depends upon what kind of startups are included in a study, and how you define 'success'. University of Tennessee research confirms generally what we already know anecdotally: the more speculative and 'high-tech' a venture, the more likely it is to fail; the primary reason for failure is incompetence; and higher success rates (measured simply by survival, rather than growth) occur in existing industries with stable and predictable markets.

A 2011 study of 650 internet startups found (among other things) that: founders overestimate the value of IP before product-market fit by 255%; startups need 2-3 times longer to validate their market than most expect; and premature scaling is the most common reason for poor startup performance.

If the patent system is to achieve its objective of providing an incentive for investment in innovation, then the IP needs to have a lifespan which is sufficient to give it some prospect of commercial success to provide a return on the investment, independently of any particular venture which might endeavour to exploit it.

It is not evident to me that 20 years does not strike an appropriate balance. I am unaware of any persuasive evidence that would lead me to choose a substantially different term if I was designing the system from scratch.


arachne said...

You don't have agree with any of my propositions. I'd just be interested to hear what your intuitive (or reasoned) opinion of an optimal patent term would be.

We are approaching the question of optimal patent term from different directions. I am most interested in ensuring the patent system extracts disclosure of inventions that might otherwise not be disclosed. You are focused on ensuring (maximising?) ownership and licensing opportunities.

I think we should be asking what is the minimum patent term that ensures disclosure. Your approach seems to be pushing for the maximum term without the system becoming completely unworkable (actually, I still don't really know what your approach would be, so I'll leave this as a point to rebut).

In my opinion the Palm example supports my approach at least as well as yours. Would the disclosure of smartphone architecture been any later if the patent term was shorter? What about the out-of-patent date of such an invention (i.e. the date of disclosure + term of patent)?

Mark Summerfield said...

Obviously I don't think we should eliminate the disclosure requirement. The most abused patent system in the world (i.e. the US system) continues to allow applicants who do not apply elsewhere in the world to request non-publication prior to grant, which is (to my mind) an excessively pro-applicant policy.

Just because applicants do not weigh up patent term as the quid pro quo for disclosure does not mean that publication is not a public benefit of the system.

You are making a very strong argument for reducing the term of patent.

No, I'm not. In the paragraph you quoted there I am specifically discussing applicant behaviour in relation to the alleged trade-off between term and the disclosure requirement. This does not mean that reducing the patent term would not change decisions as to whether or not to invest in particular development, or to use the patent system as opposed to some other form of protection (such as trade secrecy).

Most applicants I work with accept publication as a 'cost' of using the patent system, whether they like it or not. However, their actual decision as to whether or not to apply for a patent, or adopt some other strategy (including, potentially, abandoning a particular line of development), is generally commercial. As with any other business decision, what matters is whether there is a foreseeable adequate return on investment in a particular commercialisation and IP strategy. Publication and patent term are almost completely uncoupled in the commercial decision-making process.

To slavishly accept the status quo because “don't rock the boat, I'm making good money”, or it “change is hard” are clearly not adequate reasons.

That's a little offensive. I have said neither of those things. And I have backed my arguments with experience and data. I suspect we're done here, and we will just have to agree to disagree.

lorac said...

Thanks for the take-down of EFF. They deserve it, IMHO. In fact, there's a cadre of like organisations that could stand a little heat.

Mark Summerfield said...

Thanks for your comment.

I have nothing against civil liberties organisations per se, but I do get concerned when they get distracted by causes that are not really very fundamental to their core objectives. With all of the genuine injustice in the world, I would have thought the EFF could find better things to do than intervene in a commercial dispute, and that the ACLU might be able to get its head around the idea that 'gene patents' do not actually deprive anyone of liberty, integrity or identity.

Flash said...

But how do the patent examiners know the invention actually works? You don't need to provide a working system to the patent examiners. At least in the software industry, most of the R&D cost is in the actual implementation of such an idea, not the idea itself. It appears perverse to protect the very idea/concept so much.

I am concerned in this case that the very idea of allowing people to try on glasses in a virtual environment is being patented, not a specific implementation of this idea. Certainly if there is only one obvious implementation path from an idea (e.g. **cough** Amazon 1-click **cough**), then I question whether something should be patentable at all.

I am concerned in your suggestion that companies now need to do exhaustive patent searches (these would need to be done at an "arm's length" as well so you don't potentially open yourself up to willful infringement damages in the future) and buy up patents from defunct companies (just in case they may fall into the wrong hands). I think this makes it very hard for new companies to start, especially against strong entrenched competition with deep pockets. I have probably said this before but given the current patent environment I don't see how another Apple or Microsoft could start from such humble beginnings. The "cheap money" at the moment is hiding how big these barriers to entry really are.

By the way, I don't believe arachne was trying to be offensive. I think this thread was a genuine attempt to try to engage you on what could be improved in the patent system. From reading your answers, it appeared like you were evading directly answering his/her questions...so I think his/her final response, whilst not very diplomatic, was understandable.

Mark Summerfield said...

There is nothing new in startups and investors needing to conduct appropriate IP due diligence at key decision-points.

In my previous career as an engineer, I was involved in two startups from the late 90's. In both cases investors expected to see that some level of freedom-to-operate research had been done, as well as wanting to know what the company's own IP position was (i.e. did they know what they owned, did they know whether it was genuinely original, if so had they at least considered the availability of patent protection).

Why would anybody invest in a company without a proper risk assessment, including IP risk? Why would they be prepared to put money into a venture with no saleable assets, in full knowledge that most high-tech startups fail? And while it is true that you need more than just an idea to get a patent, the reality is that patents create property out of ideas, and thus facilitate investment decisions. It was, in part, my direct experience of this process that led to my interest in the patent profession.

In my view, refusing to perform a proper IP risk assessment for fear of 'treble damages' is like crossing the road without looking so that you can argue (if you survive) that the accident was somehow not your fault! I just don't think this is much of an issue these days in properly-managed ventures.

New companies start all the time. The success/failure rate has not changed much in the past four decades. Google and Facebook also had humble beginnings, and I see people wondering where, in the 'current IP environment', the next Google or Facebook is going to come from.

I have no doubt that they are already out there. But they are not doing anything as niche, or so clearly within the interests of big incumbent players, as selling eyeglasses online.

Post a Comment

Copyright © 2014
Creative Commons License
The Patentology Blog by Dr Mark A Summerfield is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Australia License.