Posts Tagged “CAFC Decision”

As Dennis Crouch of the Patently-O blog wrote “The CAFC once again reaffirmed patent law’s parallel to Chicago democracy: File early and often.” With the easy filing opportunity of provisional applications, there was no good reason for commercial sewing machine manufacturer Atlanta Attachment to lose its patent for failure to file before the “on-sale” bar date. But it did.

Atlanta was developing an automatic “gusset ruffler” machine for Sealy, the mattress company, under a confidentiality agreement, with the intent that, if successful, Sealy would patent the machine and Atlanta would only sell the machine to Sealy. In effect, Atlanta was performing paid R&D work for Sealy.

Atlanta built four sequential prototype machines. Sealy tested them as they were ready and provided feedback to Atlanta. Atlanta was paid for the prototypes and included offers to sell production models with their invoices. Ultimately Sealy decided not to move to the production stage and allowed Atlanta to retain the rights to the machine and to sell them to other companies, so Atlanta filed for and received patent US 6,834,603.

When Atlanta tried to sue Leggett & Platt for infringement, L&P successfully defended itself in the Court of Appeals (See CAFC decision) by pointing out how Atlanta had made a legitimate offer to sell their third prototype more than a year before filing the first patent application. That is, Atlanta ran right into the on-sale bar to obtaining a patent.

The “on-sale bar” is intended to keep inventors from obtain the commercial advantage of a patent for longer than the statutory period. In order to be barred from getting a patent, you must make a real offer to sell your invention AND your invention must be ready to patent.

In Atlanta’s case it appears they might have thought that their ruffler was exempted from these rules under the “experimental use” loophole. Perhaps they figured that as long as Sealy was paying them to develop the ruffler and was testing it and giving feedback they were still in “experimental mode”. It was not until Sealy ended the project that Atlanta even started to think about applying for a patent.

The “experimental use” loophole allows an inventor to sell copies of his invention and not be barred from a patent IF the purpose of the sale was to determine (viz., experiment to see) if the invention works as intended. Unfortunately for Atlanta, the judges ruled that Sealy’s testing and feedback, at least after the third prototype, was not experimental use; instead it was simply a customer trying to determine if the invention, as then embodied, met their requirements. That their customer felt the invention needed changes or improvements to make it commercially viable was not experimentation to see if the invention worked as claimed.

It is worth a note that this lesson is the corollary to my dictum that to be patentable your invention doesn’t have to work well. As the CAFC said:

The district court erred in concluding that these facts showed the third prototype was not suitable for its intended purpose. This is because an invention can be considered reduced to practice “even though it may later be refined or improved.” Emphasis added

Had Atlanta not been lulled into thinking they didn’t have to file as long as they were working with a development partner they could have easily filed a series of provisional applications, each one adding the improvements they had made since the previous filing. They then would have had a year to decide if the then described invention worked as intended and was worth the investment in a utility patent.

Comments Comments Off on A Stitch In Time…Saves Your Patent

In drafting your claims it is so tempting to take that extra bite out of the apple. You know you want to get a patent issued to stake out market territory and you know what your product goals are. With a public release date looming your patent attorney says you’ve got to file now. Surely it will all be working soon. So what do you file: an application for what you have or an application for what you plan to have?

A Court of Appeals for the Federal Circuit decision from last summer confirmed what you already knew: an invention without enablement is not a [patentable] invention. Align Technologies, the company behind Invisalign® braces, was sued by Ormco Corp. for allegedly infringing a series of patents relating to computer-aided design and manufacture of custom orthodontic appliances.

Ormco’s patent claims seemed to cover the Invisalign system. However, during claim construction (a determination by the judge of exactly what the claims mean) Ormco was told that their claims only covered a process in which final treatment positions are automatically determined (viz., by computer software) whereas the Invisalign system uses “skilled operators” to determine the final positions of the teeth. The judge issued a summary judgement of non-infringement. If that were not bad enough for Ormco, the patent was also declared invalid for lack of enablement.

 How did Ormco wind up with no infringer to sue and no valid patent? Trying to reconstruct the back story, I think this is what happened. Ormco was trying to develop a fully automated system but had a market-ready orthodontic system even without that final computerized step. Anticipating success, they applied for their patent.

At the same time their patent attorneys were doing what is generally accepted to be the best strategy – they filed the broadest claims possible, claims that did not include the limitation of the computerized step. Claims that were ultimately allowed.

Once Ormco was successfully in the marketplace I suspect they gave up on, or at least greatly reduced, their efforts to get the computerized aspect of the system working; after all, dentists were probably just as happy to see a system that required their skilled intervention.

All was well until the Court looked at the specification and the prosecution history and determined that the broad allowed claims must actually be construed as narrowed by requiring a fully automatic process. Apparently, in the specification and history, it was clear that the inventors’ primary basis for distinguishing their invention from the prior art was its high level of automation.

Then, when one of the inventors testified that:

  • Ormco had never attempted to create a computerized system that automatically determined tooth positions without human decision making;
  • The manual override had been used on all of the approximately forty cases treated using the Insignia product;
  • It was a goal to have the software generate final tooth positions that would not require use of the override, [but] variations in human anatomy had prevented the attainment of that goal; and
  • [He] was also unsure if the problems due to variations in human anatomy could be overcome,

the invalidity for lack of enablement case was sealed. As the court concluded:

If an inventor attempts but fails to enable his invention in a commercial product that purports to be an embodiment of the patented invention, that is strong evidence that the patent specification lacks enablement.

So if you are still making adjustments in your invention when your patent gets filed, be aware that your lack of enablement may let your infringers take a bite out of your market.

Comments Comments Off on Put Teeth in Your Patent: Only Claim What You’ve Enabled

 Everybody wants broad patent claims; have you ever pushed your patent attorney to include something in a claim that you really haven’t quite figured out how to implement just to “cover all the bases”? If so, you may be treated like a crash test dummy when you try to enforce your patent.

Back in the late 1980’s Automotive Technologies International had an important idea – how to make an accelerometer to trigger side-impact airbags (front impact sensors don’t respond quickly enough) – and obtain a patent to protect it. Twelve years later it sued a laundry list of automobile companies for infringement. During the “claims construction” portion of the trail, ATI pushed the court to include both mechanical and electronic sensors in the scope of all the claims. [BAH NOTE: presumably this was done because the alleged infringing sensors were electronic]. Having won that battle, ATI now proceeded to lose the war with a flawed specification.

ATI fell into a familiar trap. Having invented a fast acting mechanical accelerometer (in which a hinged plate swings to close the electrical contact), they recognized that certainly there must be an electronic way to sense the movement of the mass and didn’t want to lose out on protecting their market to someone with an electronic solution. So they described the concept (viz,. electronic instead of mechanical sensing of mass movement) and hoped (probably) that they could make a doctrine-of-equivalents argument against whoever came up with an electronic sensor, by pointing to the section of the specification in which they clearly recognized that an electronic sensor was an equivalent approach.

Unfortunately for ATI, when your whole invention is involved, when you spend a page of text and 5 figures describing the mechanical approach, when your “equivalent” approach is described in only one short paragraph and one “conceptual” figure (ATI Electrical Sensor Figure) you are not going to convince anyone that you were doing anything but speculating that the electronic version could be made to work.

The specific legal point being argued was whether ATI had fully “enabled” its claims. Enablement – the requirement that you describe your invention in enough detail so that the rest of us can make your invention with “undue” experimentation – is closely related to, but separate from, reduction-to-practice. RTP is what defines a completed invention (inspiration -> perspiration -> RTP) while enablement is the description of the completed invention that you include in your patent specification. Enablement is what you share with society in exchange for getting your limited monopoly.

With hindsight we see that ATI tried to obtain protection for a concept, not an invention. They should not have tossed in the description and figure for their concept of an electronic side impact sensor. Instead, they should have focused their R&D efforts to complete this next generation sensor and filed an application on it in its own right.

How about you? Have you tried to stretch a patent application to cover something you only believe will work? It’s a shortcut that might get you a patent that fails you when the legal hit comes along – side or front impact.

Comments Comments Off on CAFC Hits Side-Impact Sensor: The Importance of Enablement