I haven’t looked at the claims here, but the independent claims are the most broad claims of a patent. The dependent claims are always necessarily narrower. So, even if an independent claim is invalidated, it’s dependent claim may survive if it is novel and non-obvious over the prior art.
This is not true. Dependent claims are often viewed as "backup" claims in case the independent claims are rejected in prosecution by the patent examiner or found invalid by a court.
It's not hard to find online sources to back this up, for example:
You might be confusing infringement with patentability.
If an independent claim isn't infringed, then by extension none of the dependent claims are infringed and they don't have to be considered. So the easiest route to defending a patent lawsuit can be proving the independent claims don't apply.
There are two main defenses: invalidity and non-infringement. You're correct, and you're talking about non-infringement.
Invalidity can be proven outside of a full-blown lawsuit (via re-exam) and so it's cheaper. Asking a jury to judge invalidity is asking for trouble, because they usually figure they're not competent to do it.
I actually watched a video of a mock jury, where they debated this. It's pretty shocking how little they understand the law. One juror actually said, "well, that prior art reference -- it's not fair to expect the PTO to have heard of that."
You have it backwards. If a dependent claim is invalid, then the independent claim from which it depends is also invalid. It doesn't make sense the other way around.
Your Patently Obvious article refers to patent prosecution, which is different from re-examination. Let's keep referring back to my original 1(A, B, C) and 2(D) example.
A dependent claim is "backup" because the examiner may find that (A, B, C) is not patentable, but coupled with D, you get (A, B, C, D) which is.
This happens all the time in prosecution. So let's say (A, B, C, D) is allowed and issued.
Note that you did not win the right to sell a product with (A, B, C, D), because someone else (Call them CompanyX) has patented (A, B, C) and you are infringing it.
All you've got is the right to exclude anyone, including CompanyX, from selling (A, B, C, D).
If someone else infringes your patent and you sue them, and they're able to prove that 'A' was known in the art, then CompanyX's patent and yours are both invalidated.
1. A fly-swatter comprising
(A) a shaft,
(B) a handle attached to one end of the shaft, and
(C) a perforated paddle attached to the other end of the shaft.
2. A fly-swatter according to claim 1, further comprising
(D) a robot attached to the handle and configured to identify and kill a fly by applying force to the handle so as to cause the perforated paddle to strike the fly.
Here, if claim 1 is invalidated (*) by a prior art fly-swatter, then claim 2 may remain valid if the combination with element (D) is novel, non-obvious, and the other requirements are satisfied (written description, enablement, etc). If claim 2 was invalidated by a prior art fly-swatter, however, then claim 1 would also be invalidated.
(*) Note: This would only happen in a trial or IPR because during prosecution if the examiner objects to a claim then it must be withdrawn or amended to be allowable.
A good test question there (I realize I was wrong about the (A, B, C) example. You have to knock out all three.
You have two contradictory hypotheses: "if claim 1 was invalidated by a prior art fly-swatter" and "If claim 2 was invalidated by a prior art fly-swatter"
If fly-swatters are prior art (as they are), then you would claim your robot thingie as :
"An automated fly destroyer, comprising:
a fly swatting device,
and a robot configured to etc. etc."
And you would say in your spec, "the fly-swatting device can include, in one embodiment, a commercial fly-swatter, but could be any device capable of killing a fly, including newspapers, rolled up magazine, board, book, and many more."
However, if your claim enumerated the features of a fly-swatter, then it would certainly be questioned. You should not bother listing things that are known in the art, and furthermore, you'd have failed to claim the book, newspaper, magazine, etc. methods of killing the fly.
On further editing: your idea of knocking out a dependent claim by knocking out all the elements is theoretically correct. However, that's not the way things are usually done.
[1] explains the issues around this. IPRs almost always go after the broadest claims, and the patentee needs to respond "even if claim 1 is invalidated, claim 2 is still valid because blah blah blah."
Patents almost always put the heavy novelty in claim 1, like I did for your robot fly-swatter invention. The dependent claims are narrower because they add additional elements which often aren't very novel. However, you're right that they can still survive and be useful.
> Awaiting a patent lawyer who does this every day
I am a registered patent attorney (*), but I don't practice in patent prosecution.
> a fly swatting device ... And you would say in your spec ...
You're alluding to means-plus-function claiming. "An element in a claim for a combination may be expressed as a means or step for performing a specified function ... and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof." (35 USC 112(f) or pre-AIA 35 USC 112 sixth paragraph).
> You should not have claim elements that are known in the art, and furthermore, you'd have failed to claim the book, newspaper, magazine, etc. methods of killing the fly.
As a strategic matter, I agree because you usually want to claim the broadest, most essential form of an invention, and that's a good reason to use means-plus-function claiming. The presence of prior art elements in a claim, however, do not invalidate the claim because (1) for a prior art reference to anticipate a claim, all elements must be present in the prior art reference (all elements rule), and (2) for non-anticipating prior art references to render a claim obvious, the claim must be considered as a whole, not element-by-element. (35 USC 103, "...such that the claimed invention as a whole would have been obvious...").
OK, you're an attorney, I also passed the Patent Bar (and am a Patent Agent). But it was before AIA. I don't think the 2004 Patent Bar had anything about re-examining claims, but it sure as hell did cover PCT, reissues, death of the inventor, and other crucial points of law :)
Now that we've managed to alienate all the civilians, would you say
"means for killing flies" or
"fly-killing device" where you say in the spec what that includes? Or have dependent claims that name the most common fly-killing devices.
> would you say "means for killing flies" or "fly-killing device"
I don't know. Again, I passed the patent bar, but I don't practice in prosecution. So, by my estimate, you have infinitely more experience than me. The answer probably depends on whatever the Federal Circuit most recently said about 112(f).
