From time to time I encounter a client or foreign colleague that asks whether some soon-to-be-filed US patent application ought to be termed a CIP (continuation-in-part) of a previous patent application by the same applicant. The way I look at it, this used to be a good idea but nowadays is rarely if ever a good idea. I hope that folks will post comments with their thoughts about this.
Those of us whose patent practice preceded June 8, 1995 will recall the days when the term of a US utility patent was seventeen years from the date of issue. In those days, a decision to call something a CIP did not adversely affect the term of the to-be-granted patent. The decision to call something a CIP was in the nature of “can’t hurt and might help”. The usual practice in those days, when we were getting ready to file a new patent application for a client with previous filings, was simply to call any new application a CIP of any and all copending applications filed by that same client. Each of us has at one time or another reviewed the file wrapper of a granted patent where the CIP relationships go back a decade or more, through a dozen or more previous filings by the patent owner.
When I see this in a modern-day patent application, especially in a patent where it looks like the underlying application was poorly or hastily drafted, I imagine the drafter’s decision to call the application a CIP of everything the applicant ever filed previously (and to incorporate it all by reference) was based at least in part on a desperate effort to reduce the risk that the poor or hasty drafting would turn out to be missing something important. The hope of course was that if in a later office action the Examiner said something lacked support, maybe the hapless practitioner could find the missing support somewhere in the list of applications in that long and rambling opening sentence. Or if, in response to an office action, the practitioner needed to add yet another limitation to a claim to try to overcome the prior art, maybe the language of the desperately needed new limitation could be found somewhere in that list of applications.
And maybe some readers have encountered the sort of thing I have run into from time time when a new client transfers in its portfolio of pending US patent applications that were drafted by previous counsel. And there, too, nearly every one of the pending applications calls itself a CIP of most if not all of the other members of the portfolio.
These days when I see a pending application that calls itself a CIP, my reaction is to recommend that the client delete the CIP relationship. The reason, of course, is that the patent term in recent years is twenty years from the oldest US non-provisional filing date. So a decision to call something a CIP is a decision to forgo months or maybe years of patent term.
Let’s turn now to a particular fact pattern. Suppose that the would-be parent of the would-be CIP was itself published more than one year before our soon-to-be-filed filing date. Then the unhappy situation is that the Examiner will be able to cite that publication against the new application.
The alert reader might then ask “well if the publication of the would-be parent can be used to kill our patent application, then what is the point of calling something a CIP”? Which is a very good question. And there are maybe two answers.
The first answer is that sometimes you will get lucky and the first publication of the parent case will happen to be less than a year before the present (new) filing date. In such a case the Examiner will probably not use the publication against the application because of the remnant of the one-year grace period that still remains in Section 102 even after the America Invents Act.
The second answer is that in a CIP, the Examiner proceeds on a claim-by-claim basis to work out what is the effective filing date of each claim. Suppose that (I am making this up) claim 7 of the new case happens to be supported by the parent case. If so then the effective filing date of claim 7 is the same as the filing date of the parent case, and this beats the publication of the parent case so far as the patentability of claim 7 is concerned.
The other thing that might, I suppose, happen is that a particular Examiner might somehow misunderstand what a CIP is and might think (incorrectly) that a CIP relationship somehow trumps the previous publication of the parent case even when the publication was more than a year earlier and even when the claim being examined is supported only by the new filing date. The applicant whose case falls on the desk of such an Examiner might well obtain an allowance of that claim without the Examiner having taken into account the pertinence of the publication. But it cannot be likewise assumed that the adversary in later litigation would overlook this.
Maybe sometimes nowadays when an applicant calls a new application a CIP, it is in the hope that their application might get assigned to such an Examiner. The way it looks to me is that such an applicant is, I guess, scratching off a lottery ticket, and the purchase price of the lottery ticket is the loss of some years of patent term.
Comments welcome.
This is more questions than comments:
Questions 1 and 2 — The only thing that matters is whether the parent has disclosure that gives full § 101 utility and § 112(a) support for at least one claim of the daughter. The facts in these two questions are not relevant and don’t change either the analysis or the answer.
3a is an interesting question. CIP is just a convenient name that we use among us girls, with only little more tangible relevance than other terms like “business method patent” and “patent troll.” It is not statutory. The only statutory outcome that turns on these facts is effective filing date. The only purpose of the regulation that requires “relationship” is to tip off the examiner to the correct effective filing date to use in searching prior art. So on the 3a facts, it doesn’t matter what you call it.
But I would call it a CIP from the get-go, precisely because of the possibility of 3b.
Bostonian, I appreciate your “among us girls” categorization of the CIP nomenclature, and it made me smile. But while it may not be statutory, to my mind the USPTO has clearly defined a CIP as a continuation that includes new matter (regardless of whether the new matter is claimed or not). MPEP 201.08. So I think your advice of calling it a CIP from the get-go is more of a rule than an option.
fred, as many questions as you get answered on this blog and elsewhere, there will be still others that you don’t even know to ask. For your sake, please consult a competent patent attorney, and hopefully they will ask and answer the right questions for your situation. Best of luck!
I’m a “new” patent attorney with a few years of experience. I am trying to learn from those with more experience.
fred, my apologies! I’m glad you’re asking these questions.
Question #2 was directed to the following quote from the article: “Let’s turn now to a particular fact pattern. Suppose that the would-be parent of the would-be CIP was itself published more than one year before our soon-to-be-filed filing date. Then the unhappy situation is that the Examiner will be able to cite that publication against the new application.”
If it wasn’t published, could the Examiner use it against you?
Question 1. Does your analysis change if the CIP is heavily based on the parent application such that the application would not be patentable if the parent was prior art?
If the application was published (unlike my question #2) and you DID claim priority with a CIP, wouldn’t you get the priority from the parent for the original subject matter, so you couldn’t have your own parent application cited against the CIP application at least for the subject matter that was in the parent application?
Question 3a. If you add subject matter to the specification, but do NOT claim the new subject matter, is it a CIP or just a continuation?
Would it make a difference if it was a transition application, so the parent was filed pre-march 2013? By not claiming the new subject matter you wouldn’t trigger AIA.
I know you are well aware, Carl, but I thought I should mention that the expected shelf-life of a patent varies by industry, so the loss of term may be weighted very differently in the strategic calculus depending on what the technology is.
Substantively, another purpose of a CIP is to keep a priority chain alive, offering the ability to later file a continuation with claims fully supported all the way back to the original parent, through the CIP. This later continuation would generally be one not foreseen at the time of filing the CIP or else you would probably file the continuation concurrently with the CIP. Of course, don’t forget that there are additional important considerations if the CIP would be an AIA transitional application.
As a secondary advantage, the use of CIPs (especially the spider-CIP approach you are talking about) makes analysis of a portfolio by a competitor more difficult and expensive. And the resulting business risk is higher because of the additional variables (such as the priority date of each and every claim). This approach, though, often seems to spring from the same nearsighted reasoning as “I draft crappy claims so that no one can ever confidently draft a non-infringement opinion on them.”
You write: “… the use of CIPs (especially the spider-CIP approach you are talking about) makes analysis of a portfolio by a competitor more difficult and expensive.”
What you wrote is a very clever point that I had not fully appreciated. It can backfire, as one client portfolio intake exercise of a few years ago reminded me. The new client transferred the portfolio in, and then we started a process of trying to generate a map that would tell us which applications were related to which other applications. Our chief reason for the mapping was so that we could be smart about cross-citing references in IDSs. Pretty soon the map was so big that, even using small print, it occupied a printout that was two feet by three feet. The mapping process probably cost the client five figures of money.
This question is already answered —
The only thing that matters is whether the parent has disclosure that gives full § 101 utility and § 112(a) support for at least one claim of the daughter. The facts in these two questions are not relevant and don’t change either the analysis or the answer.
The effective filing date for a *claim* turns on the support in the earlier doc, and only that. The examiner’s reference is or is not prior art depending on that support. Whether you characterize an *application* as a “continuation” or “CIP” is irrelevant.
There is something to be said for the theory that you should file a CIP if, and only if, there is reasonable doubt whether the claims you want to file are supported by the putative parent application. If the claims are clearly supported by the parent, file them in a straight continuation, and avoid the cloud over their effective filing date. If the claims are clearly not supported, file them in a new application, to simplify the paperwork and (if your art cares about that) get the later expiry date.
I agree with Henry, though with one additional wrinkle — you might file a CIP when you have a new species that you want to add to a pending genus claim, and you’re close enough in time that you’re willing to trade off patent term for the cost saving of abandoning the parent and prosecuting only the daughter.
This dialog reinforces my general rule to file a non-publication request routinely when the rules permit. Non-publication offers numerous advantages:
1. It allows the client the option to keep the information in the application as a trade secret.
2. It keeps competitors from seeing exactly what is disclosed and claimed in the application, so they do not know what the scope of the publicly-announced ‘patent pending” is.
3. It keeps competitors from seeing how the PTO is reacting to the application.
4. It prevents your earlier application from becoming a printed publication against you later application, whether it is CIP or not.
I have yet to see any benefit to an applicant from publication of his/her application.
I agree 100% with your analysis and usually advise nonpublication UNLESS we are filing for foreign protection or if we have a known infringer and want to go after provisional damages.
BUT, let me play devil’s advocate here and throw out some reasons that a printed publication might be a good idea.
1. Your client can’t afford to produce the product, but wants to reserve the right to make it. If it is published, it can prevent a patent from being filed later. (Abandon, suppress, conceal arguments won’t work.)
2. If you publish, your competitors may send you references while the application is pending. This can be thought of in a positive light as a “free search.”
I don’t buy your reason number 1. Once the client’s patent issues, it is a reference as of its filing date. If it is doomed not to issue and you want a defensive publication, you can do that right before letting it go abandoned.
I hate to keep asking questions (especially the same question), but I think that there is an important issue here. I could be wrong.
This following is a quote from the article: “Let’s turn now to a particular fact pattern. Suppose that the would-be parent of the would-be CIP was itself published more than one year before our soon-to-be-filed filing date. Then the unhappy situation is that the Examiner will be able to cite that publication against the new application.”
In Question 2, I ask: Does your analysis change if the original application has not been published?
But if there was no publication (e.g. the CIP was filed less than 18 months from the filing of the parent and/or there was a nonpublication request), can the Examiner hold an applicant’s own, unpublished, original (parent) application against the CIP application?
And what about the exceptions:
35 usc 102 (in part) is below:
(b) Exceptions.—
(2) Disclosures appearing in applications and patents.— A disclosure shall not be prior art to a claimed invention under subsection (a)(2) if—
(C) the subject matter disclosed and the claimed invention, not later than the effective filing date of the claimed invention, were owned by the same person or subject to an obligation of assignment to the same person.
Does 35 USC 102(b)(2)(c) give an applicant any protection against having his own applications sited against him? The one-year limitation of 35 USC 102(b)(1) does not appear to be in this section.
And does the exception overrule the publication of the application if it is published?
Thanks!
No change.
Never plan your prosecution strategy around the prior art that the examiner does or does not find, or any other assumption that the examiner will act perfectly. Prosecution is about getting money or market exclusivity out of an issued patent, not about being a hole for the client to pour money into during prosecution. After the patent is issued, and you try to get money out of the patent, the recipient of your lovely invitation to license will find prior art that the examiner did not. You choose your effective filing date to deal with all that future art–only bad patent attorneys have planning horizons that end on the issue date.
§ 102(b)(2) only carves out prior art from § 102(a)(2), an issued patent is § 102(a)(1) prior art notwithstanding § 102(b)(2). So a commonly owned patent becomes prior art from its date of issue or publication, not from its date of application. The drafting of this statute is so atrocious, ambiguous, and baroque that I am sure there are unintended consequences buried in there.
OK, I think I understand where you are coming from on CIP applications. But if there was some reason to claim priority to the parent and the parent had NOT been published before it issued, would the relevant date for the parent application to be used as prior art be the issue date? The parent wasn’t a printed publication (issued) before the effective filing date of the claimed invention. (Of course, if there was a public use or offer to sell more than one year before the filing of the CIP, then this would be another problem.)
Is the reason that this is not a problem for continuations and divisionals because they are fully supported by the parent and, therefore, their effective filing date is prior to the publication, . . .?
One more question, does 102(a)(2) carve out issued patents and published applications as not being the regular “publication” of 102(a)(1)?
Carl:
Why do you say “the unhappy situation is that the Examiner will be able to cite that publication against the new application.” Can the basis for this in the rules? Here is the fact pattern that I am facing:
1. Inventor writes and files own application.
2. It gets published.
3. It gets rejected.
4. I need to rewrite claims. In probing inventor for how the invention really works I am discovering all sorts of details that are not in the initial disclosure.
If I the new claims and additional specification details as a CIP could the intially filed application be used to reject the CIP?
Yes if its publication date is more than a year before the filing date of your CIP.
Here is a slightly changed set of facts: A second application (now PATENT 2) is designated as a straight CONTINUATION (or DIVISIONAL) of a first application (now PATENT 1), and (1) the PATENT 2 specification clearly contains new matter, while (2) all claims of PATENT 2 are supported by the original specification of PATENT 1. What to do? Certificate of Correction? Reissue? Leave it alone?
The set of facts might give rise to a “check-the-box” problem. By this I mean the ADS box where the filer tells the Examiner whether to use the “new law” or the “old law” to figure out whether the claims are patentable.
Loss of patent term is discussed here with respect to filing CIP vs filing new application. I am just pointing out that loss of term remains possible in view of obviousness-type double patenting rejection and terminal disclaimer to overcome the rejection. In other words, if you have a clearly non-obvious invention, there is no reason to file a CIP because of the loss of patent term for claiming the earlier priority/filing date. If your invention might be obvious, filing a CIP sacrifices patent term, and filing a new application might also sacrifice the same amount of patent term if the applicant ultimately has to file a TD (where the publication of the earlier application is less than one year prior to the filing of the later application).
Is this obvious to everyone already?