Apparently Sutton’s superior knowledge of the prior art has placed him in a position to challenge many patent attorneys over the past couple of years, enabling him to inform those attorneys that – in his words – ‘you’re wrong, fix it’ when they have sought to defend their short, broad claims. In his concluding paragraphs, Sutton calls on his own experience at Oracle to suggest that a particular source of this problem is laziness on the part of the patent attorneys. ‘Laziness is bad,’ he says, ‘and I am calling you [i.e., the ‘Dear Patent Attorney’ addressed by the article] out on it.’
While Sutton’s piece strikes me as stunningly arrogant in its tone, this is perhaps excusable considering that he works for Oracle – a company that received over 600 US patents last year. Although he is at pains to point out that the opinions expressed are his own, ‘and do not necessarily represent those of Oracle Corporation, its subsidiaries or affiliates’, there is no question that he is well-placed to know what he is talking about. And there is no disputing the facts established by his analysis of the thousands of patents/applications of two large (but anonymous) software-related product/service companies:
- many applications in the data set were filed with remarkably short independent claims;
- examination (‘prosecution’) of patent applications generally (and not just those with ‘extremely short’ claims) results in amendments that lengthen the independent claims; and
- on average, applications filed with shorter independent claims are subject to lengthier and more costly prosecution, and to more extensive claim-lengthening amendments.
Based on my own experience, I would suggest that the answer to both of these questions could well be ‘no’. More data is needed if we are going to understand what is truly going on here. I doubt very much that it is as clear-cut as Sutton proposes.
Big Clients, Big Discounts, Broad Claims?I have been fortunate as a patent attorney in having only rarely worked directly for the kind of large, big-patenting, companies that have been used as the basis for Sutton’s analysis. He does not name his subjects, but my guess is that if you were to name four obvious suspects off the top of your head, Sutton’s two would most likely be in there.
While big companies can mean big business for attorney firms, the reality is that for the individual attorneys who do the actual work, they are not always great clients. A company that files 1000 patent applications in a year wants bang for its buck. Such companies drive a hard bargain (and rightly so), and firms are willing to offer very attractive pricing arrangements in exchange for volume of work.
But it is truism of almost any project that out of ‘fast’, ‘good’ and ‘cheap’ you can only pick two. At a firm level, churning through the volume of work to justify the cut pricing means deploying (cheaper) junior attorneys and/or pressuring more senior attorneys to work faster. At the individual attorney level, there is no incentive to spend the time to do a better job, because it will not contribute to those all-important billable hours. You are not going to get any thanks from the big client, either – that patent application that you toiled over is just one of 50 they are going to file this week.
This can be stressful, unrewarding work, which is why I say that I have been fortunate to have largely avoided it. As an attorney, you are just part of these clients’ patent production line. You do not really get to know them, or their business, or feel as though you are anything more to them than a gun-for-hire. You know that you are not doing your best work, because you are not being given the time and resources to do so. And you also know that the client does not really care – competent work that covers what they consider to be the invention, and provides adequate disclosure to deal with whatever arises during prosecution, is all they are looking for. So you give them that, and move on to the next job.
Big Portfolios, Big SavingsThe thing is, however, that these big-patenting companies are absolutely right to operate this way! Much as I personally do not want to be the one doing this kind of work, I understand the motivation.
For a startup or a small company with one or two innovative products, each invention they devise, and the patents that protect them, may be vitally important to their competitive edge, and to the value of the business. However, for a big company with a large R&D budget and a huge patent portfolio, each individual patent is of relatively little importance in itself. What matters more is the value and structure of the entire portfolio.
This is not to say that some patents are not more important and valuable than others – clearly, they are. But it will rarely be the case that the ultimate value of each individual invention is known when the original patent application is prepared and filed. If there are 1000 filings in a year, it is simply not practical to treat all of them as if they might turn out to be the rare diamond amid the zircons, and spend the time and money necessary to produce a gold standard application for each and every invention.
It is also a certainty that not every application will survive examination, even if the inventors think they know the prior art pretty well. In prosecution, as in nature, the weaker inventions will be picked off by their predators – which, in this case, are subject-matter eligibility, novelty and inventive step. By definition, Sutton’s data only tells us what happened in those cases where a patent was actually granted, which implies that the applicant considered it worthwhile to spend the time and effort overcoming rejections. We do not know how many applications were abandoned. And even if we did have this information (which should not be hard to obtain), we would not know why each one was abandoned. Some may be completely hopeless cases (e.g. knock-out prior art found by the examiner). Some may be cases in which the rejections might have been overcome, with sufficient effort, but the benefit would not have justified the cost. And still others might have been abandoned regardless, because they were simply reassessed as being of insufficient value to the business.
Considered from this perspective, Sutton’s estimate of an ‘additional’ $10,000 cost in prosecution of these ‘overly broad’ applications is not an additional cost at all. Rather, it is a deferred cost. Less is spent on the preparation of every single application, thus massively reducing the client’s up-front costs, perhaps by many millions of dollars each year. As a consequence, some of those applications – if they run into trouble during examination, and if they are, by that time, still worth the cost of continuing – may cost more in prosecution. Setting aside the additional disadvantage of prosecution history estoppel (which is very hard to quantify, particularly on a ‘whole of portfolio’ basis) it is difficult to argue that this is not a perfectly rational and reasonable trade-off.
Conclusion – It Takes a Village to Raise a Patent!I do not set out to draft ‘extremely short’ or ‘overly broad’ claims for clients with computer-implemented inventions. By the same token, I do not allow clients to labour under a misapprehension that drafting patent specifications with decent prospects of standing up to scrutiny in this technical field is a simple or cheap exercise. If a prospective client wants a patent application at a bargain price, they are welcome to shop around!
On the other hand, I can fully understand why big companies with large portfolios might adopt a different approach. If in doubt, it is always preferable to err on the side of breadth. It is generally easier and less contentious to narrow the scope of claims during prosecution than to try to broaden them.
Furthermore, in all of this we are not even looking at how or why the short independent claims are being amended. Are entirely new limitations being imported from the specification, or are the applicants simply falling back upon narrower dependent claims that were quite sensibly included in the original application? Are the amendments required to overcome the citation of prior art (and thus almost certainly substantively narrowing), or are they addressing subject-matter issues (which may involve ‘fleshing out’ the language of the claims with additional technical detail, but without significant narrowing of the practical scope of the patent)?
The process of obtaining a patent – or, indeed, in some cases not obtaining a patent – from the initial invention disclosure through to a final disposition of some sort, involves a complex interplay of technical, legal, business and financial considerations, and involves inventors, managers, and in-house counsel, as well as the outside service providers. Laying the primary blame on patent attorneys for the way the process plays out is manifestly absurd and unreasonable. But surely senior patent counsel at one of the world’s largest and best-known software companies must know this already?