In order to proceed further, we have to step back for a moment and revisit our exploration of patents; specifically, we need to look in more detail at what a patent actually is.
And I don’t mean that in any philosophical sense; for this exploration is about understanding a patent in the same sense that you want to understand a fish that you’re about to eat: what are its parts; which of those parts taste good; and, of course, can we infer from all this intimate ichthyological information how to find other good fish to eat, so we’re not stuck eating bottomfeeders ad infinitum?
All The Parts
At the most fundamental level a patent consists of two parts: the claims; and, everything else. This dichotomy is easily understood, in that the part of the patent that has direct legal effect is the claims, which define the scope of patent protection; all else is used in interpreting the claims.
The claims of the patent are found at the end; I once heard an amusing story told by an academic researcher who was told to review a number of patents relating to her work before meeting with a number of very rich investors, and how she dutifully spent all night trying to read several hundred pages of patent jargon only to discover that, when they said “read the patents,” what they actually meant was “read the claims” — which are of course at the end of any patent document. So when reading a patent, the meaty part — the claims — is at the end.
So what do claims look like? Here’s a cooked-up example:
1. A fish comprising: a head; one or more gills; a body; one or more fins; and, a tail.
What this claim means is that we have a thing — in this case a fish — which has at least five elements: 1) a “head” (whatever that is); 2) at least one “gill” (whatever that is); 3) a “body” (whatever that is); 4) at least one “fin” (whatever that is); and, 5) a “tail” (whatever that is). In general the fact that the thing is called a “fish” doesn’t mean much of anything, this “preamble” part of the claim (that is, the part before the colon) is usually just a placeholder for the actual definition that follows, although in some unusual cases the preamble gets rolled into the definition of the claim.1
On the other hand, the use of the word “comprising” is very important — a claim using this word is considered an “open” claim, since “comprising” means “the following list of elements, but there could be more elements than are just in the list.” So “A thing comprising elements 1, 2 and 3” could be a thing of just those three elements, or a thing of more than those three elements — that is, an open set. However, a claim like “A thing consisting of elements 1, 2 and 3” can only refer to a thing with just those three elements — this use of the major words “consisting of” results in what’s called a “closed” claim.2
Fine, preamble, open versus closed. Got it? Now on to the delineated elements themselves — in the example claim there are five elements, head, gill, body, fin, tail. These elements are self-evident, right? We all know what a fish’s head, gill, body, fin and tail look like?
Well not necessarily. Blind cave fish may not have eyes, for example, so should “head” automatically mean a head with eyes, without eyes, or both? All the specifically fishy bits are generally clear — gills, fins, tail — but those fishy bits are very different in different fishes,3 so what does the claim mean when it uses each of these terms?
And don’t even get me started on “body.” Clearly the claim is referring to some sort of structural member, but is it just a lump that the head, gills, fins and tail go into, sort of like the potato in Mr. Potato Head? Or does the body do other things? And even if it’s a Mr. Potato Head type of body, what’s the spatial arrangement in which the head, gills, fins and tail go into it?
I could go on (and, obviously, on), but the above is probably enough to give you more of a taste than you wanted of what it’s like to be a patent agent or attorney. And also — I hope — a sense of how the elements of any particular claim work. Specifically, they are words or combinations of words, that usually have some sort of common meaning, but that can be slippery enough in that meaning that there’s room for doubt in interpretation — something that should be avoided at all costs in a patent (unless of course you explicitly want to leave wiggle room, which is a strategy that I won’t discuss now).
The way of clarifying the words is by means of the “specification,” which is everything in the patent that isn’t the claims, and which commonly includes the text and drawings (although drawings aren’t required in a patent document). So for example let’s say that we had the following:
1. A fish comprising: a head; one or more gills; a body; one or more fins; and, a tail.
… “Body,” as used herein, refers to the portion of the fish that is intermediate between the head at the proximal end and the tail at the opposite distal end, and that serves a number of purposes, including acting as the point of attachment/integration of the gill(s) and fin(s). Fish bodies may be rounded or flat (e.g., for the flat-fishes such as the flounder), they may be covered with scales of various types and compositions, mucous, etc. Examples of various bodies explicitly contemplated in the present invention are provided in figures 1-42 …
where the second paragraph is taken from the text of the patent. Now you can see how the “everything else” portion of the patent — in this case the text and drawings — help with the interpretation of the claims.
The Good Bits
So how does this rather long explication of a patent’s two fundamental components — 1) the claim or claims, and 2) everything else — help us understand what’s particularly important about a patent, what part(s) ought be carefully scrutinized and what part(s) can be safely glossed over?
Part of the answer is that it depends upon why you’re looking at the patent in the first place. If you’re looking at it as a way of determining possible infringement, you start with the claim(s). If you’re trying to get a capsule summary of what the patent protects, once again the claims are the usual place to start.
If, on the other hand, you’re trying to get a broader sense of what kind of protection is being granted for a particular technology area, and what the state of the art is in that area,4 then you’ll probably find less value in the claim(s) — which are, after all, in some sense a separate summary language of what’s in the document as a whole — and more value in a review of the entire document.
“De gustibus non disputandum est” as the Romans used to say — there’s no right or wrong in matters of taste.
Finding Other Fish (Patents)
Perhaps surprisingly, “De gustibus …” is less applicable when you’re using a particular patent or patent application as an entry point for patent document searching. In searching you want to use both the claims and the specification; but since each is written in a somewhat different language, how you use each section is different.
This should be somewhat intuitively obvious from the simple fact that the claims represent a condensation down of what’s written in the specification; the claims have a conciseness that the specification doesn’t have. So let’s say you’re doing a keyword search based on keywords that you pull from the claim(s) versus keywords that you take from the rest of the patent document (the specification). The claims may be the perfect place to get useful keywords because they’re so condensed that the words in the claims might tend to be the really important ones, whereas the specification might have enough other stuff in it that it’s hard to select useful words from it.
For example, the specification might talk broadly about a new invention for changing lightbulbs, in which case it would probably use words like “lightbulb,” “incandescent bulb,” “fluorescent bulb” etc. Those are good words, but if you searched them you’d find patents that are more likely to be about lighting itself than about changing the lights.
On the other hand, our hypothetical lightbulb changing patent application (or issued patent) would probably have claims that talk about — at the very least — burned out bulbs, since that’s the essence of a lightbulb changing apparatus right? And so if we used keywords like “burned out lightbulbs” and searched for that in the claims only, we would be much more likely to pull out on-point patents and patent applications relating to changing lightbulbs rather than making them.
On the other hand — there’s always an “on the other hand” isn’t there? — an invention on lightbulb changing might have claims that are very detailed in their description of the changing apparatus itself, and they might say very little if anything about the fact that the apparatus is for changing lightbulbs.5 Such claims would be much less likely to contain “burned out lightbulb” type keywords, and would therefore be harder to find if you were searching those terms or — in terms of keyword selection — would be misleading if you picked keywords from them based on their claims.
So there’s always a balance to be struck between picking keywords from the claim(s) versus from the specification; and searching keywords that occur only in the claims, while useful in terms of limiting down the number of hits, often fails to pick up documents that are relevant.
Clear enough? Perhaps, but once again the best thing to do is to give some examples. Stay tuned for the next post, where I’ll do just that.
- In US patent law this happens when the preamble, for example, “breathes life and meaning” into the claim. [↩]
- This discussion is, as usual, USA-centric. [↩]
- A bit of pedantry: “fish” is either a single fish or multiple fish of the same species, whereas “fishes” refers to more than one fish of different species. So fish have more or less the same bits (apart from natural variation and subtractions of parts by accidents), whereas fishes — being different species — likely have different bits. More than you wanted to know, I’m sure. [↩]
- Note that patent documents don’t become public until at least 18 months after their earliest filing date. Thus no matter how thoroughly you look at issued patents and published patent applications, you’ll never see documents still within this 18 month window. [↩]
- Remember that the preamble doesn’t usually get rolled into the meaning of the claim, so the precise language of the preamble can be rather wiggly (an IP person’s term of art). [↩]