Print

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

[avalys]

I am working for a start-up company developing a very complex system composed of multiple innovative components that are in many respects tightly-coupled (meaning it is hard to describe the functionality of one without describing several others, even though they are all distinct components that can exist and be used separately but not optimally with existing technology).  We will eventually be hiring patent lawyers, but I would like to understand the process and considerations involved in dividing our entire system into a set of patents that adequately cover both our entire new system, and the individual components that comprise it.

Can anyone provide any guidance, or does anyone know of a good book/resource that will help with this?  I have bought "Patent it Yourself" by David Pressman and "How to write a patent application" by Jeffrey Sheldon - both are good references, but they discuss only single patents. 

As an example, one question I have is: if I have a patent application for technologies A and B, and it is obvious that embodiments of these can be combined, is there any situation where it would be useful to write a patent application specifically describing a device that combines A and B, or would it be best to just describe that combination as one of the embodiments of A or B?

[JimIvey]

Oh, I don't envy you for the task you have lying ahead....  That's how many of my client relationships start -- a huge ball of inter-related innovations getting ready for release.  The topic has come up much here.  Let me see if I can pull up a few posts/topics.....

I found this one.  That sets out some of the basic considerations.

The bottom line is this:  if a number of innovations are all tightly-coupled as you say, it's often best to write them all into one large patent application.  It all depends, though.  When I begin to cover a large collection of inter-related innovations, the things I'll need to write start taking shape in my head and I get a sense for whether it makes sense to file one large application (to be divied up later, or perhaps at filing) or a number of applications that incorporate each other by reference.  Either way, they all tend to be filed at the same time, so the task is comparable either way.

As an example, one question I have is: if I have a patent application for technologies A and B, and it is obvious that embodiments of these can be combined, is there any situation where it would be useful to write a patent application specifically describing a device that combines A and B, or would it be best to just describe that combination as one of the embodiments of A or B?

Well, be careful about using "obvious" as it's a heavily loaded term in patents. 

There are generally two ways to approach that issue.  One is to write 2 separate applications, one for each of A and B and incorporating the other application by reference.  Each application should make it clear that A and B can be used together.  A second way is to write one application for both A and B.  You can claim A and B separately and even file separate copies of the same application for A and B, with different respective claim sets for each innovation. 

If you hope to claim a combination of A and B, I'd follow the second approach.  If the number of claims gets a bit large, you can separate the application into 3 applications with respective claim sets for A, B, and the combination of A+B.  The descriptive portions of the 3 applications would be identical, more or less (might change the summary and/or the abstract).

There are many ways to go about covering many innovations in a mass filing.  Each has its merits.  And, if you hope to get patent protection outside the US, there are other challenges -- particularly in Europe.

I hope that helps.

Regards.

[MYK]

Good timing.  I've got a bundle to put together too.  Considering doing multiple provisionals and then joining them into one big app, just to be sure the end result gets an early priority date on the central thingamajiggie.

[JimIvey]

Of course, I don't recommend provisional applications for reasons mentioned many times herein.  If it's important that some applications aren't prior art for other applications, it would be good to file them all at the same time.

Regards.

[khazzah]

I get a sense for whether it makes sense to file one large application (to be divied up later, or perhaps at filing) or a number of applications that incorporate each other by reference.  Either way, they all tend to be filed at the same time, so the task is comparable either way.

I typically don't make the "write as a big 1" vs. "split into N" decision until I've spent a significant amount of time on the project.

I think some clients expect significant fee reductions for this sort of situation: "You're not really writing 5 different apps, they're all related"; or worse, "it's just one big spec with 5 claims sets so I'll pay you $X base + $Y per claim set". But I find that the time I spend understanding and managing the complexity outweighs most of the "efficiencies" gained by the similarities.

A couple of other random thoughts. One, if the decision is to go with multiple apps, conventional wisdom says file on the same day so that no one app comes up as prior art against the other. Two, if you wanted to defer filing fees, you could write as one big app, file a single app with claims directed to one of the many inventions, then file continuations to the other inventions as money allows/the market suggests.

[ChrisWhewell]

As an example, one question I have is: if I have a patent application for technologies A and B, and it is obvious that embodiments of these can be combined, is there any situation where it would be useful to write a patent application specifically describing a device that combines A and B, or would it be best to just describe that combination as one of the embodiments of A or B?

I'd look to which types of patents one may obtain would be the most enforceable.  It sounds like you're considering dropping a wad on IP protection and will maybe foreign-file.  If anything is to be foreign-filed, I'd look at enforcement in those foreign countries of the same types of patents, for example, its easier to enforce a process patent in the US than it is in Europe (last time I checked).  I'd decide not to file certain types of patents in certain places, and otherwise whittle down the brainstorm session results to get some focus.    I'd be tempted to approach it from the bottom-up, looking at the simplest sub-combinations likely patentable, and building up from there to my heart's content.  Good luck !!

[avalys]

Thank you for all the insightful replies.

I used 'obvious' intentionally, I am aware of it's special meaning.  Let's say A is a bullet/cartridge design, and B is a type of gun.  You can use A bullets in prior art guns, and B guns for prior art bullets, but the best embodiment of A bullets is with my B gun, and the best embodiment of my B gun is with my A bullets.  Is there any reason to file a patent on A, a patent on B, and a patent on the "weapon system" A+B?

In reality, technologies A and B are actually even more closely related than bullets and guns, but.

[JimIvey]

Is there any reason to file a patent on A, a patent on B, and a patent on the "weapon system" A+B?

Well, assuming you have good reason to use patents for your particular business model, there is probably very good reason to pursue A and B -- probably in two, separate applications.  Whether it makes sense to pursue A+B depends on the amount of inter-relation between the two.  By that I mean the extent to which special features of A interact with special features of B better than with prior alternatives for B and vice versa.

If A is just a great bullet and provides the same advantages for prior art guns that it does for B, and if B is just a great gun and provides the same advantages using prior art bullets that it does using A, then it probably doesn't make much sense to pursue an application for A+B.  However, if A+B provide some synergies, it might make sense to pursue A+B as its own innovation, particularly if your strategy is to have a large portfolio in terms of numbers of patents and applications.

Now, because A and B work particularly well together, you should mention each in the other's application, probably incorporating by reference, to meet the "best mode" requirement.

I hope that helps.

Regards.

[avalys]

So, if I say that A's best mode is A+B, and B's best mode is B+A, I can still get a separate patent on A+B (whose best mode is A+B, with no real invention other than "stick bullet A in gun B")?

A+B is definitely an innovation over the prior art, but in order to meet the "best mode" requirements, it is completely described in the A and B patents separately.

[JimIvey]

If A is non-obvious, or B is non-obvious, then A+B is non-obvious.  You'd want to file an application for A+B at the same time (same day, anyway) as A and B, so neither would be considered prior art.

If A and B were prior art, you'd probably need something more than what you suggest in order to have a reasonable hope of getting a patent to cover A+B.

Like I said, it's a matter of your strategy.  Some like to cover every possible combination of their innovations.  In that case, you might want claims covering A+B.  Others don't care so much for covering every possible combination as long as all the parts are covered.  In that case, you'd probably be fine letting A+B go.

Regards.

[avalys]

Like I said, it's a matter of your strategy.  Some like to cover every possible combination of their innovations.  In that case, you might want claims covering A+B.  Others don't care so much for covering every possible combination as long as all the parts are covered.  In that case, you'd probably be fine letting A+B go.

Regards.

But if I wanted a claim on A+B, I could simply have the A patent incorporate B by reference and describe/claim A+B there, right?  I don't understand why you would want to go through the trouble and expense of a separate patent on A+B, other than to pad your patent count.

[JimIvey]

If you want to claim A+B, it's probably best to write one application with A, B, and A+B described.  You can include claims covering A alone, B alone, and A+B.  Or, you can clone the application into 3 with distinct claim sets for each. 

If you choose the former, the examiner will probably limit you to either A or B, but not both.  You might be able to keep A+B in the same application, but either A or B will have to be pursued in a separate application you can file later.

Regards.

[LaplacesDemon]

Interesting. I have an A + B + C that work best in combination. I'm primarily interested in protecting them in combination. C is pretty much useless without A & B. A could be quite valuable on its own but could have prior art conflicts. B could be made to work without A and / or C but not as well. My intent is to file a single patent with separate claims. A + B + C, A + B, B + C, A, B.

Based on previous comments I'd guess the USPTO may not let me do that. If they don't will they at least come back and tell me of any prior art conflicts so I know I'm not wasting resources filing each separately?

[LaplacesDemon]

Now that you gentlemen have me going down this road I'm starting to figure out what some of the combinations have in common and getting to the root of "What is it?" I can see how some of the variations can be combined into one statement. Going through this process is helping me nail down what my claims should be. Perhaps that will help the OP as well.

I'm still curious to know, however, what the USPTO would come back with as far as prior art conflicts should they tell me my variations are too numerous.

[khazzah]

My intent is to file a single patent with separate claims. A + B + C, A + B, B + C, A, B.
Based on previous comments I'd guess the USPTO may not let me do that.

I'm still curious to know, however, what the USPTO would come back with as far as prior art conflicts should they tell me my variations are too numerous.

I don't see how "prior art conflicts" and "variations are numerous" are related.

There is a > 90% chance the USPTO will use prior art to reject one or more of your claims in a first Office Action. So you prepare a response and negotiate back-and-forth until you get some claims allowed or give up.

What's more problematic is getting a restriction requirement for having different inventions in the same patent app. In that case, you pick one invention (corresponding to particular claims) and file one or more divisional applications for the other inventions that you want to pursue.

And if really want a heads-up on what sort of prior art the PTO may find, you can always pay for a patentability search.