When – yet another life! - I was working in "research", i.e. waiting to be posted to some unhealthy part of the world for real work, we used to be preoccupied with technical issues of weight such as how the heck seismic inversion worked (we are going back a while now). Avid reading of Geophysics often brought illumination (just add up the samples!) and the published algorithm was passed on to the programmers, who did something completely different, but that's another story. At the time, it seemed that scientists just published their results for the greater good, and that everyone could pick up on the fruits of their labors, it was the natural way of the world.
Client from hell
Of course it was never really like that, and has never been. The seismic software houses were playing a fairly open, but delicate game. If they "discovered" something, like a novel decon, or a new way of migrating, then they would weigh up the merits of publication – a kind of marketing after all, or of use – in a proprietary sense. The disadvantage of the latter was that, unless the algorithm was published, the "client from hell" with an arm full of math degrees would be convinced that what was really happening was, well just adding up the samples for instance. So publishing was in reality a route to credibility and hence sales. Now this did mean that others could copy, that investment in R&D led to a time window when the discoverer could "get ahead" of the market, and move on to higher ground while others began to cash in on the "old technology". Only in exceptional circumstances would a patent be applied for, this is after all a costly and time consuming process, with no guarantee that the patent obtained will actually serve any useful purpose.
Old as the hills
Patents and patent litigation in geophysics are actually as old as the hills, or at least as old as Geophysics. In 1936, in Volume 1, there was a paper by a certain C. R. Hrdlicka summarizing pending patent litigation relating seismic exploration. Since then, although Geophysics has been regularly publishing patents, there have been no further "case histories" of litigation in this field. Of course back in those days, software did not exist, but in addition to patents for actual physical devices, many of the earlier patents were taken out on methods. Noteworthy among these was CDP (patent filed in 1950) and Vibroseis (1953). Today, while most geophysical patents are still applied for on the basis of "inventions", software algorithms are increasingly the subjects of claims. Hitherto, these have served as territorial marks, to be negotiated if necessary, but rarely have they either made their inventors much money, or even been challenged in a serious way. To our knowledge, geophysical software patents have not before been the subjects of high profile litigation although rumors abound as to behind the scenes arm-twisting and settlements. We asked James D. Ivey, a California based attorney with E&P experience who now works in the field of computer systems and software, to provide some background on the state of the art in software patents. His contribution (see side box) shows that the goalposts have shifted somewhat in recent years and explains why patenting is often a necessity.
Controversy
In the software world at large, patents are the subject of considerable controversy. The arguments for patents are fairly obvious, if an individual or a company has invested a lot in developing an original algorithm, then it is reasonable for them to expect to be protected from others copying the algorithm and having a free ride. The counter argument goes that programming grows incrementally, and that in many cases it is almost impossible to write a program, which is entirely novel. This has not stopped many software developers from patenting just about anything, from the look and feel of a GUI, to sorting algorithms that a high-school student could well be expected to "invent" for homework. The League for Programming Freedom (http://www.lpf.org) has submitted a lengthy document to the US Bureau of Patents, which addresses these issues. They argue that the current legislation on patents is out of kilter with the modern world of software development, and cite examples of patents of such trivial ideas as "the use of different colors to distinguish the nesting level of nested expressions" (US patent no. 4,965,765) or "use of a host independent byte ordering" (4,956,809). The field of text data compression is cited as becoming an area where software development is "out of bounds. There are now so many patents in this field that it is almost impossible to create a data compression algorithm that does not infringe at least one of these patents". The costs of defending or proving patent infringement are also cited as potentially crippling for a small company. Even large software companies such as Oracle have opposed the patentability of software, arguing that that copyright and trade secrecy legislation are better suited to protecting software development.
White flag?
In the Geophysical context, Rutt Bridges of Landmark has questioned the overall economics of patenting calling for a "truce on patents. Companies should re-examine the role of patents and litigation within their organizations. Key questions are: How much money does your company spend acquiring patents? How much time and research effort? How much money have patents made for your company? In these times of tight budgets, isn't it wise to rethink the patent policies of the 1970s and 1980s? The proposal is to create consortia of companies with unlimited cross licensing of seismic processing patents. This would speed up software development dramatically, reduce cost, and benefit all members of the group." (The Leading Edge, January 1994). So what should the industry do? Protect any invention however trivial at any cost. Or open up a free for all where even an algorithm, which has cost years of original R&D can be implemented by anyone. Clearly neither path is satisfactory, but neither is the status quo satisfactory, except for patent lawyers!
Click here to comment on this article
If your browser does not work with the MailTo button, send mail to pdm@the-data-room.com with PDM_V_2.0_199707_3 as the subject.
© Oil IT Journal - all rights reserved.