Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Chas]

I want to claim a "system" of one or more devices (computers, cell phones, etc.).  If I claimed a "system", would a single device that met all of the claim limitations be infringing?  In other words, I'm wondering if the standard meaning of the word "system" in a claim can include a single device, or whether "system" means multiple devices.

Alternatively, if I claimed "an apparatus" that comprises "a processor" and "a memory storing instructions" that when executed performed various functions, could that apply to multiple devices that in aggregate met the same limitations?

Thanks in advance for your thoughts.

[khazzah]

I want to claim a "system" of one or more devices (computers, cell phones, etc.).  If I claimed a "system", would a single device that met all of the claim limitations be infringing?  In other words, I'm wondering if the standard meaning of the word "system" in a claim can include a single device, or whether "system" means multiple devices.

The default rule is that terms in a preambles are not patentable weight. If so, there is no difference in claim scope between "a device comprising X and Y" and "a system comprising X and Y." That said, if liability turns on this issue, you can bet the infringer will come up with a lot of arguments about why the term should be given patentable weight, and about how his "system" is not covered by the claim term "device."

Alternatively, if I claimed "an apparatus" that comprises "a processor" and "a memory storing instructions" that when executed performed various functions, could that apply to multiple devices that in aggregate met the same limitations?

Now *that* is a great question. I believe some practitioners have boilerplate in the spec in an attempt to achieve that effect.

If I was an infringer that used two separate devices with two separate processors to perform the function, I'd certainly argue there was no literal infringement. 

However, this seems like exactly the kind of situation that Doctrine of Equivalents was meant to cover. Does functionality-split-among-multiple-devices perform the same Function in the same Way with the same Result as same-functionality-in-a-single-device? My knee-jerk answer to this is Absolutely, so that the infringement is captured under DoE.

[JimIvey]

Yeah, that's a good question.  Would a cluster of computers operating in a distributed manner infringe a claim to a single "computer system"?

Of course, a method claim would cover that.

At one time, I had some boilerplate in the Spec that a computer system included multiple individual computers operating in a coordinated fashion to exhibit the behavior described therein.  I'll have to give that some more thought as to how thoroughly I carry that through the Spec and through the claims.

I wonder to what degree I can claim computer readable media in which the logic is distributed across multiple media.  For example, what if an otherwise infringing product is distributed in multiple parts -- e.g., one on CD and the other by download?

As for the reverse, whether a single device can be a "system", I think it's pretty clear that it can.  You can buy an SoC (system on a chip).  If a "system" can be realized in a single chip, it seems clear that a "system" can be a whole device.

As Karen noted, DoE should cover logic that is merely converted from a single computer to a distributed implementation, something that, in and of itself, is not new.  However, there is some language in case law that requires that the equivalent implementation was not known during prosecution of the claims -- the typical example being replacement of vacuum tubes with transistors, the presumption being that the invention of transistors happened when the claims could no longer be readily amended.  I don't remember whether that particular part of Festo survived the SCt.  I believe it's at least a factor, but probably not controlling.

As I've commented numerous times, the law more or less requires patent applicants to be clairvoyant, perhaps omniscient.  It doesn't hurt to be a minor deity.

Regards.

[khazzah]

However, there is some language in case law that requires that the equivalent implementation was not known during prosecution of the claims -- the typical example being replacement of vacuum tubes with transistors, the presumption being that the invention of transistors happened when the claims could no longer be readily amended.  I don't remember whether that particular part of Festo survived the SCt. 

I believe you're referring to Festo's "forseeability" exception for the presumption that the patentee gave up equivalents through amendment.

I haven't studied Festo in a while, but I think the triggering amendment isn't just any amendment, but one related to the element for which you seek equivalents. That is, making the following amendment

A device comprising:
a vacuum tube; and
a widget blodget connected to the vaccum tube.

results in a Festo presumption against obtaining widget as an equivalent for blodget. But the Festo presumption does not arise for equivalents to vacuum tube -- I should be able to get transistor as an equivalent -- even though the claim was as a whole was amended.

So in the current hypo, the only damaging amendment would be

A device comprising:
a plurality of processors;
a processor;
...

I admit that I take this position more as a matter of common sense than from a study of Festo. Though this statement from the conclusion in Festo III seems to support my position:

Under these circumstances, the amended claim elements are entitled to no range of equivalents.

[JimIvey]

My take-away from Festo is that we shouldn't ever consider the DoE to be any sort of safety net, that practitioners should more or less forget the DoE exists.  The DoE is really there for litigators to compensate for the human fallibility and imperfect knowledge of practitioners.

I agree that the DoE ought to cover conversion of a single-machine process to a distributed process, but we'd do well to try to make the DoE unnecessary for that purpose.  That's the point I was trying to make.

Regards.

[khazzah]

My take-away from Festo is that we shouldn't ever consider the DoE to be any sort of safety net, that practitioners should more or less forget the DoE exists. 

Hmm. Disagree. I think DoE coverage still has above-zero value, even after Festo. Even so, I make every effort to ensure my claims will be literally infringed.

OTOH, it seems like capturing some instances of infringement under literal infringement requires a stretch in claim construction, so that you might be more as likely to win on DoE as on a strained claim construction.

Here's one example. Some folks say that "circuit configured to perform bandpass filtering" reads on a processor that does bandpass filtering as well as an analog circuit that does bandpass filtering. After all, a processor is a circuit made of lots of transistors. However, my personal opinion is that depending on what's in your specification (what it says about the meaning of "circuit", types of circuits described, etc.) this interpretation is a stretch. Sure, you can add language which you hope will increase claim scope, but the court may not read "circuit" to cover a processor.

Therefore, I feel better knowing that my claim should cover, through DoE, an infringer who implements the bandpass filter with a DSP.

we'd do well to try to make the DoE unnecessary for that purpose.  That's the point I was trying to make.

Agreed. Literal infringement is always preferable than DoE. If for no other reason than DoE complicates the infringement case and requires additional evidence/testimony about equivalents.

[Chas]

Given the discussion so far, it seems prudent to pursue a strategy that doesn't depend on the DoE, even though, as Karen stated, it has "above zero value".

As for the reverse, whether a single device can be a "system", I think it's pretty clear that it can.  You can buy an SoC (system on a chip).  If a "system" can be realized in a single chip, it seems clear that a "system" can be a whole device.

It seems that "system" can be used to mean either a set of components working together to create a single device (a computer), or a set of computers working together for some purpose.  When I posed the question I was assuming the latter.  The spec uses the term "data processing system", and I'm not sure how that leans.

How about this as an attempt to cover one or more computers working in aggregate to perform particular functions...

apparatus comprising:
  a processing portion
  a memory portion storing instructions that when executed perform the following functions:

The attempt here is to make the language "number neutral".  Does it work?

If so, I imagine it may be possible to apply the same solution to "computer readable media" (which also seems to be number neutral).

[JimIvey]

Therefore, I feel better knowing that my claim should cover, through DoE, an infringer who implements the bandpass filter with a DSP.

we'd do well to try to make the DoE unnecessary for that purpose.  That's the point I was trying to make.

Agreed. Literal infringement is always preferable than DoE. If for no other reason than DoE complicates the infringement case and requires additional evidence/testimony about equivalents.

I'm not sure how to reconcile the first part with the last part.

How about this as an attempt to cover one or more computers working in aggregate to perform particular functions...

apparatus comprising:
  a processing portion
  a memory portion storing instructions that when executed perform the following functions:

The attempt here is to make the language "number neutral".  Does it work?

In essence, you're asking how your claim would be interpreted by a judge who was presiding over some interstate contracts issue or insider trading the week before and jurors such as dentists, school teachers, stay-at-home moms/dads, unemployed professional video gamer wannabes living in their moms' basements, and people who graduated high school last June.  Would the absence of any numbers clue them in?  I don't know.

Regards.

[smgsmc]

Given the discussion so far, it seems prudent to pursue a strategy that doesn't depend on the DoE, even though, as Karen stated, it has "above zero value".

As for the reverse, whether a single device can be a "system", I think it's pretty clear that it can.  You can buy an SoC (system on a chip).  If a "system" can be realized in a single chip, it seems clear that a "system" can be a whole device.

It seems that "system" can be used to mean either a set of components working together to create a single device (a computer), or a set of computers working together for some purpose.  When I posed the question I was assuming the latter.  The spec uses the term "data processing system", and I'm not sure how that leans.

How about this as an attempt to cover one or more computers working in aggregate to perform particular functions...

apparatus comprising:
  a processing portion
  a memory portion storing instructions that when executed perform the following functions:

The attempt here is to make the language "number neutral".  Does it work?

If so, I imagine it may be possible to apply the same solution to "computer readable media" (which also seems to be number neutral).

I would go with:

A system comprising:

a processor; and
a data storage device.

Although the preamble typically has no patentable weight, it could under some circumstances.  Now, one could argue that “apparatus” or “device” means a single unit (whatever that is) and could be given patentable weight.   However, “system” is more nebulous; some Examiners love to issue 112 p 2 rejections once they see “system”.  But in this case, it can be put to good advantage.

In a previous life, I was a systems engineer for over ten years.  A “system”  is entirely context dependent.  If you are designing a cell phone, a cell phone can be a “system”.  If you are designing a base station, a base station can be a “system”.  If you are designing a cellular  network, a “system”  can be the whole shebang:  cell phones, base stations, switching centers, routers, servers, …. I don’t think one could argue that “system” means more than one unit; therefore, it wouldn’t have patentable weight.

We recently had a long thread on “a” = “at least one” = “one or more”.  So I would argue a “system” would literally infringe whether it is configured as a single computer or ten computers.  I would also argue that a “system” would literally infringe whether the data storage device is a CD-ROM built into the chassis of a computer or a stand-alone remote access hard drive connected to a network via Ethernet.

[Monkey1]

There is a related U.S. subclass for:
Dependency based cooperative processing of multiple programs working together to accomplish a larger task
http://www.uspto.gov/web/patents/classification/uspc718/defs718.htm#C718S106000

Cases in this class/subclass may give some examples of some related claim formats/structures.

[Chas]

How about this as an attempt to cover one or more computers working in aggregate to perform particular functions...

apparatus comprising:
  a processing portion
  a memory portion storing instructions that when executed perform the following functions:

The attempt here is to make the language "number neutral".  Does it work?

In essence, you're asking how your claim would be interpreted by a judge who was presiding over some interstate contracts issue or insider trading the week before and jurors such as dentists, school teachers, stay-at-home moms/dads, unemployed professional video gamer wannabes living in their moms' basements, and people who graduated high school last June.  Would the absence of any numbers clue them in?  I don't know.

But don't you have an opinion on the potential effectiveness of the strategy?  Do you see any down side to the approach?

I would go with:

A system comprising:

a processor; and
a data storage device.

... “system” is more nebulous; some Examiners love to issue 112 p 2 rejections once they see “system”.  But in this case, it can be put to good advantage.   ... A “system”  is entirely context dependent.  ...

Thanks for your thoughts and suggestions.  I'm leaning that way also.

But why "data storage device" instead of "memory"?  Isn't "memory" broader in that it could include RAM, etc.?

[khazzah]

Therefore, I feel better knowing that my claim should cover, through DoE, an infringer who implements the bandpass filter with a DSP.
...
Literal infringement is always preferable than DoE. If for no other reason than DoE complicates the infringement case and requires additional evidence/testimony about equivalents.

I'm not sure how to reconcile the first part with the last part.

Hmm. Let me try again.

Literal is better than DoE because a) DoE is estopped under some conditions and b) DoE requires additional evidence. Even so, despite all my efforts to write a spec/claim that captures through literal infringement, the court may interpret my claims in a way that doesn't allow for literal infringement. If that happens, I think I can still use DoE to read a claim to "a bandpass filter circuit" on a DSP implementation of the filter. Because this sort of situation is precisely the reason that DoE exists.

[JimIvey]

... despite all my efforts to write a spec/claim that captures through literal infringement, the court may interpret my claims in a way that doesn't allow for literal infringement. If that happens, I think I can still use DoE to read a claim to "a bandpass filter circuit" on a DSP implementation of the filter. Because this sort of situation is precisely the reason that DoE exists.

Yeah, I get that.  It just sounded like you disagreed with my point and then agreed with it.  Maybe I didn't originally make my point well.

Just like you should drive as if you're not wearing a seatbelt while always wearing your seatbelt (y'know, just in case), you should try to write your claims as if there is no DoE, even though the DoE can save your posterior portions some day. 

Regards.

[Chas]

There is a related U.S. subclass for:
Dependency based cooperative processing of multiple programs working together to accomplish a larger task
http://www.uspto.gov/web/patents/classification/uspc718/defs718.htm#C718S106000

Cases in this class/subclass may give some examples of some related claim formats/structures.

Interesting thought -- I'll take a look.  Thanks.