A hazy shade of value: Software patents just took a hit

No readers like this yet.
open source button on keyboard


Much of the problem with the current U.S. patent system involves haze. Whether it is ambiguity regarding what the patent covers or murkiness about whether the technology covered by the patent is actually new, patent plaintiffs regularly use this uncertainty to their advantage. This ambiguity–often planned–allows plaintiffs to leverage settlements against businesses that are forced

to either pay their own lawyers or opportunistic plaintiffs, or take their chances in court where judges and juries are often unsympathetic to companies.

This haze is particularly pronounced with software patents. Unlike electrical patents, in which the electrons of an accused circuit either move in the way described by the patent or they don't, software patents can be manipulated to cover "inventions" far removed from the patent's description or any intent of the patent's inventor.

An important area of "haze" frequently exploited by plaintiffs regards the calculation of damages. Most often, damages "models" are arrived at by a damages "expert" (paid by the plaintiff) picking a number (recommended by the plaintiff or its counsel) and then–after the fact–generating support for this number. Such "support" is often hazy in itself, relying on self-serving opinions on market trends, ultimately including only arguments that support a huge damages award.

Damages analysis in such cases is not empirical in any way. Unlike the methods of science that at least some of these patents are based upon, there is no true testing or empirical research. Instead of researching the market and technology and only then calculating a number based on research, the process is normally reversed. The problem with this "black box" approach is that virtually any number can be selected and presented to a jury that may sympathize with an individual inventor over a large corporation.

Earlier this week, the Federal Circuit (appellate court in charge of patent cases) removed a bit of the haze by invalidating the "25% rule" or "rule of thumb" (see opinion at UNILOC USA, INC. AND UNILOC SINGAPORE PRIVATE LIMITED, Plaintiffs-Appellants, v. MICROSOFT CORPORATION, Defendant-Cross Appellant [PDF]).

The gist of the rule was that a patent holder should be entitled to 25% of an infringer's profits. The problem, as the Federal Circuit pointed out, is that this approximation is not tied at all to what a reasonable negotiation between the parties would have produced (a requirement under patent law). At a minimum, the rule ignored many important factors, including consideration of the particular party, product, or industry involved, as well as commercial realities such as the amount of risk taken by the "infringing" party to successfully market the product.

The rule also overlooked the rates paid by the licensee for the use of other similar patents, any relationship between the patent holder and infringer, the portion of the profit or selling price that is customary in the particular business, and the patentee's licensing policies. Importantly, plaintiffs' use of the 25% rule ignored the fact that the same plaintiff usually wanted to apply the 25% figure to all sales of an entire product–even though the alleged patented invention was only a small part of the product–which additionally may have had nonpatented but market-changing elements added by the infringer. The effect of this was that the percentage of profit from the specific patented invention asked for was much greater than 25%.

The rule was merely a shortcut that eliminated much or all of the detailed analysis that should be a part of proper damages analysis. In other words, it was yet another way for patent plaintiffs to hide behind haze to produce unjustifiably large patent damages awards.

With this decision, the value of hazy patents, including software patents, has decreased. While further reform is needed regarding damages and other aspects of all patents, particularly software patents, the death of the "rule of thumb" eliminates an important weapon from patent holding companies and others that attempt to manipulate successful businesses through the patent system. In the end, all we can ask is that plaintiffs be required to prove their case–including their bases for damages. This recent ruling helps.

User profile image.
Erick Robinson is Director of Patents - India and Patent Counsel for Qualcomm, and formerly was Senior Patent Counsel for Red Hat.


About effin' time - score one for the good guys, score one for the consumer, and score one for justice.

That is some of the most encouraging news that I've heard of late.

From what I've seen of the Republican's policies of late, I wouldn't be surprised if they don't find a way to either reverse this or "tweak" it so that it gets real hazy again.

BTW, I'm not a Democrat or Republican or an independent. I'm discouraged.

this is a long time coming. It is sad we get so excited about something that should of been done years ago and is such a common sense item. But nothing about patents has much common sense involved.

That is an important step and it proves that it is acknowledged that there are problems with the system.
For software patents I think that the use on a computer makes part of the haze. Computers did become generally used, building software stays for quit a large part of the population mysterious. People with a legal profession are not the most active free time programmers.
Patents descriptions are apparently trying to limit the scope of the patent. Probably trying to avoid prior art. Using this in that specific platform etc. So every new platform can provoke many new patents. But should the scope even be limited to computers? Should using an obvious or old method for the first time on a computer even be patentable?
I can't interpret that law enough but for me it is not obvious that there is a different patent inside and outside a computer. Software being a procedure that is registered in a special format and a computer being a tool, one way to execute procedures.

I think very harmful patents are those where a company succeeds in setting a standard. A standard wherefore patents are needed. Patents held by the same company.

The company can now get it's research investment payed back many times. Obviously from that moment the company is not that much interested in improving the standard, in innovation, if that could create even a small risk for his market position. That same mechanism could also be used by a group of company's.

Not covering every problem, but for my feeling there is a conflict of interests having the possibility to push a standard and ask money for patents needed by that standard. Having a monopoly on a type of software and better supporting your own patent protected format.

I don't know it there are already laws against it. Perhaps the anti-trust legislation, using one monopoly to create an other?

It is true that much of the problems in the patent system have a lot to do with haze more reason why I would advise that one contact a consulting company like <a href=http://www.fulcrumpatentsolutions.com/>Fulcrum Patent Solutions</a> for any patent related issues rather than go it alone.

<a href=http://www.fulcrumpatentsolutions.com/>Reverse Engineering</a>

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.