journal
Misery loves prior art
Last month, I had to make up an exercise for patent agent trainees (that’s the “misery” part — not making up exercises, but training for the patent agent exams). The exercise itself was the preparation of a response to a examiner’s report, in which a patent examiner raised a number of objections to a patent application. The exercise response required amendments to the patent application, and argument why the prior art located by the examiner did not render the claimed invention obvious.
Naturally it’s fictitious — that is, the examiner’s report is fictitious, and the patent application at issue is adapted from a real patent or application, usually tweaked a bit in order to introduce more errors or uncertainty. The subject matter is usually something simple and mechanical, because the patent agent exams deal with relatively simple mechanical inventions which, in theory, do not require specialized knowledge to understand. (In practice, everybody but the mechanical engineers seems to claim a disadvantage.) So of course, I decided to use a patent for a knitting-related device. The subject matter is simple and mechanical, knitting skill is not actually necessary to be able to understand the invention, and there are few other opportunities to introduce knitting into a patent practice.
The patent I picked on was the one for Knit Klips clips, which were designed for seaming knitted pieces. Unlike the knitting needles with measurement markings, the Knit Klips patent claims generally seem to be fairly true to the product that is actually on the market. (Technically, the patent wasn’t a patent yet when I used it for the exercise — as you can see from the first link, the issue date was in March. But it was on the verge of issuance.)
If you look at a diagram of the clip (especially the ones on the company’s website), it isn’t hard to figure out what how it works.
And, to amuse yourself, before you click the “more” link, ask yourself just what sets these clips apart from any other clips you might have seen in the past. What the inventor (through her agent) and the patent examiner thought were inventive is discussed below the jump.
Comments (6)One skein?
Here’s a recent Japanese patent application (JP2005042246), filed in 2003 and published in 2005. I don’t know the status now, but it is not necessarily patented yet. It’s probably undergoing examination now.
I won’t pretend to understand Japanese, so here’s the English translation of the abstract:
PROBLEM TO BE SOLVED: To obtain a jacket keeping the shoulder warm at a low cost for keeping the whole shoulder including the neck warm at bedtime through improving productivity.
SOLUTION: This jacket 10 keeping the shoulder warm is obtained by knitting and weaving a surface with regular texture and a rear face with pile stitch while forming a bottom 12c and a neckline 13 of a body from one cuff 11a through the body 12 toward the other cuff 11b, and connecting a turtleneck 14 to the neckline 13. The jacket is set, if necessary, with a knitted texture of polyurethane threads as an intermediate layer formed between the knitted textures of the surface and the rear face of both the sleeves 11 and the body 12, and a pocket 16 to hold a scent bag 15.
And here you thought that shrugs and boobholders were all about keeping glamorously warm. It’s not about your looks; it’s about your productivity, people!
I still feel the same about the aesthetic merits of such garments. But when you present it to me as a technical solution to the problem of maintaining productivity at bedtime, that’s different. And note the pocket for holding a scent bag!
(What you say? Oh, sorry, that should have read, “keeping warm… through improving productivity.” All your shoulder-warmers are belong to us.)
Don’t leap to the conclusion that a grant of this patent, should it be found patentable, would result in an embargo on all shrugs and shoulder-covering delicacies. First of all, patents are only supposed to be granted for new inventions. We’ve seen before that sometimes it’s not so easy to find prior art in a field like knitting. It’s easier for machine knitting-related development, since they’re more likely to be documented. But a patent should not prohibit others from doing something that really and truly was publicly known by others before the patent application was filed. (Sometimes it happens. And when it does, corrective measures can be taken, either by the patentee or by a prospective infringer.) Thus, for example, I should still be able to knit and wear any pre-2003 cropped sweater pattern in Japan, even if a patent did issue on this patent application (I didn’t see any foreign patent applications in my cursory search, but I wasn’t looking very hard).
Secondly, I assume from the abstract that the patent claims, which define the exclusive rights in the invention, do not simply cover a cropped, short-sleeved garment. That’s a big assumption, considering I can’t read the claims.
Instead, I’m guessing that if there’s anything patentable here, it would involve at the very least either the scent pocket or the intermediate layer of polyurethane fabric between the inner and outer faces of the fabric making up the garment, or perhaps the actual method of construction. The abstract itself describes that the garment is made of a fabric formed “by knitting and weaving a surface with regular texture and a rear face with pile stitch” which, in the absence of further detail in the application that I can’t read, sounds like a lot of machine-made fabric.
I do wonder, though: where does the scent pocket go in the women’s version? Right under the neck?
Worn socks: the mother of invention
You would have noticed in the double-or-even-quadruple-knit stockings patent that the description skipped over certain realities of sock knitting: namely, the heels and toes. In response to Lola’s comment that an explanation of heel shaping was missing, I flippantly answered that one could add an afterthought heel, and that in fact there was a patent on it.
In the meantime, Kate wrote to tell me about some 40s-era Patons publications she had obtained that described the “Beehive Aladdin Heel” and the “Beehive Innovation Sock”; the latter she described as having an instep and three sides of the ankle worked flat, stitches picked up for the heel, and the sole and back of the ankle knitted and seamed to the first place; in short, a replaceable heel.
She asked how one might go about finding out whether a patent had actually issued. And thus begins the lecture. So, instead of writing only about afterthought heels, first I’ll write about the type that requires forethought.
Comments (3)Inadvertent publication
Today’s intellectual property-fest is brought to you by the letter ®.
If you’re a designer who is planning to file and prosecute your own trademark application, especially in the US, you really should think about what you supply as a specimen of use in support of your application. Not only should you ask yourself, “will this specimen prove that I’m using the trademark?” — which is something your specimen needs to do — but you should also ask yourself “do I want to put this on the public record?”
When you seek registered intellectual property rights — patent or trademark — what you submit to an intellectual property office becomes part of the public record, at some time or other, and it becomes accessible to the public, somehow. In the case of a US trademark application submitted electronically, that’s immediately and online.
(In the case of a patent application, with a few exceptions your application is published eighteen months after filing. Many patent offices publish the patent application online; the US and the European Patent Office now make the entire up-to-date patent file history, including the applicant’s submissions, easily accessible online. This is a terrific tool. Not only can you track your own patent applications as they wend their way through the system, but you can keep very close tabs on competitors, too.)
So, if your example of trademark use is a knitting pattern bearing your trademark, and you choose to provide the USPTO with a full copy of that pattern bearing your trademark as a specimen, then your complete pattern will be uploaded to the USPTO website and made available for download, for free, to anybody who knows where to look. Perhaps a knitting pattern might be your only appropriate specimen, but if that’s the case and it’s a multi-page publication, consider carefully whether you actually need to supply all pages as a specimen. It may well be that the first page or the cover page will be sufficient. And that way, you would avoid turning a document that you normally offer for sale into a free download.
If it turns out that you did do just that, I’m not certain there is a way to withdraw the specimen of use so that it no longer appears online. Conceivably parts of a file could be sealed by a confidentiality order, but a specimen of use cannot, and in a case like this a confidentiality order would have to be after the fact and if you’ve let the cat out of the bag yourself, it’s often hard to lock the barn door. (Please don’t cry over spilt milk. Just deplore my mixed metaphors.)
By engaging the patent or trademark system, you are at least implicitly giving the government permission to deal with your submitted work, even if protected by copyright, in a manner consistent with its legal obligations to the public.
Standard warning: just in case you start getting ideas, the fact that you, a user, manage to download a file with the apparent blessing of a government entity doesn’t mean that you have the right to start distributing the file to whomever you please. Copyright doesn’t explode into nothingness just because something’s on the net. Thank you.
Comments (2)Anna Makarovna’s patented stockings
Monday seems like a good day to start off with a healthy dose of patentese. I’m going to try to post something knitting-patent related once a week — maybe something new, like Knit Klips knitting clips, or stretchy circulars brand needles. We’ll see how faithfully I keep to that schedule.
Today, though, a historical tidbit that’s been sitting on my hard drive for a few years. It has become especially relevant now, though, since Kory Stamper’s article on knitting two socks at once on dpns was published in Knitty. That knitting parlour trick, briefly described in Leo Tolstoy’s War and Peace, is a perfect example for a discussion of prior art and enabling disclosure:
The melancholy silence that followed was broken by the sounds of the children’s voices and laughter from the next room. Evidently some jolly excitement was going on there.
“Finished, finished!” little Natasha’s gleeful yell rose above them all.
Pierre exchanged glances with Countess Mary and Nicholas (Natasha he never lost sight of) and smiled happily.
“That’s delightful music!” said he.
“It means that Anna Makarovna has finished her stocking,” said Countess Mary.
“Oh, I’ll go and see,” said Pierre, jumping up. “You know,” he added, stopping at the door, “why I’m especially fond of that music? It is always the first thing that tells me all is well. When I was driving here today, the nearer I got to the house the more anxious I grew. As I entered the anteroom I heard Andrusha’s peals of laughter and that meant that all was well.”
“I know! I know that feeling,” said Nicholas. “But I mustn’t go there- those stockings are to be a surprise for me.”
Pierre went to the children, and the shouting and laughter grew still louder.“Come, Anna Makarovna,” Pierre’s voice was heard saying, “come here into the middle of the room and at the word of command, ‘One, two,’ and when I say ‘three’… You stand here, and you in my arms- well now! One, two!…” said Pierre, and a silence followed: “three!” and a rapturously breathless cry of children’s voices filled the room. “Two, two!” they shouted.
This meant two stockings, which by a secret process known only to herself Anna Makarovna used to knit at the same time on the same needles, and which, when they were ready, she always triumphantly drew, one out of the other, in the children’s presence.
War and Peace was first published in Russian in the 1860s. The fact that Tolstoy wrote about two stockings being knit at once suggests that this was a technique handed down in Russia (unless we are to suppose that Tolstoy developed the idea himself), if not Europe, that therefore must have antedated the publication of War and Peace. And
But a patent application for accomplishing Anna’s “secret process” was filed in the United States by Frederich Polle, Louisa Keisker, and Sallie Polle on October 24, 1874; their patent issued as U.S. Patent No. 167,563 on September 7, 1875. If you’d like to read the text and see all the figures of this patent, they’re reproduced below the jump.
Polle et al. describe the rudiments of knitting two layers at once using a technique that we’d call “double knitting” today — the variant with two separate ends of yarn, without exchanging the ends, so that the fabric layers always remain separate. Interestingly, Polle et al. only describe the easier part of working the stockings; if you read through the description, you’ll see that there is no discussion about how to shape toes or heels. The only reference made to shaping stitches is this:
By means of this improvement… uniformity of size is secured without the usual comparing of one article with the other, when dropping or taking up stitches in the act of reducing or increasing the diameter of the knit article.
… yet there is no explanation provided as to how to reduce or increase, although this is the trickiest part of the double-knit sock trick (read the Knitty article to find out how to do this).
On the other hand, Polle et al. go to different extremes: they don’t stop at double knitting; they go on to describe knitting four layers at once!
Thus, in simultaneously knitting four stockings, we cast onto the needle C a stitch from the first yarn, A, then one from the second, B, one from the third, C, one from the fourth, D, until the required number of stitches is set up on the needles. Then, after knitting the first yarn, A, we pass it outside of the needles, where it will always remain during the work. Next, pull up the second yarn, B, from underneath the first, knit it, and pull it down. This yarn will continue to be between the first and second garments throughout the work. Then knit the third yarn, C, reversing the stitch to make the right side of this garment face toward the right sides of the other two, keeping this yarn inside of the needles, where it will always remain, the nearest yarn to the knitter. Next, pull up the fourth yarn, D, from underneath the third, and knit it, reversing the stitch. This yarn should always be kept between the third and fourth garments.
The one claim of Polle et al.’s patent is to “[t]he improvement in the art of knitting two or more garments at the same time upon the same set of needles, the same consisting in setting up on the needles the loops of each garment, so that they alternate one with the other, and in removing them in the same order, substantially as described [in the patent description].” (Today, one would not rely on a single patent claim as breezily worded as this; the rules of claim construction have evolved over time.)
If we assume that it was known to knit two garments, such as socks, at once in the manner described by Polle et al., prior to 1874, and if we assume that this knowledge was not commonly known, but was known by a select bunch of knitters who passed on the technique orally — both of which are assumptions that we are making based only on a passing reference by Tolstoy — then how is it possible for a patent to be granted?
Easy. Early in U.S. patent law history, applications for patent were not examined for novelty; by the time this trio applied for their patent, there was an examination process in place. But even so, there are limits to the patent examiner’s ability and resources. In a field such as knitting, where so much was passed on through oral tradition rather than in writing, it is highly possible that a technique could be publicly known, and publicly shared, yet completely miss the examiner’s notice–particularly because the patent office examination for novelty would be restricted to only printed publications. And even today, although so much is recorded in writing, it’s still entirely possible for a publicly available reference to remain “hidden”, only to be unearthed when a defendant or potential defendant in a patent infringement action starts hunting around for prior art (whether printed or practical) to invalidate the patent.
On the other hand, if we were to assume that there was an actual, printed reference that the examiner could have found in 1874 that disclosed this technique of knitting two articles at once, this would not necessarily leave the applicants without a patent: they did, after all, teach how to knit more than two articles, and arguably figuring out how to do this, and where to put all the working ends of yarn, would not have been anticipated (rendered not novel) by a printed publication restricted to two articles.
War and Peace, by itself, could not have been a prior art reference, even if it had been available in the United States in English to the examiner in 1874: while it described the desired result — two stockings knit at once — it was not an enabling disclosure, because that mere description would not have told the reader, even one skilled in the art of knitting socks, how to carry out the method. It doesn’t lead the reader to the method described by Polle et al.; it’s the difference between Poincaré’s conjecture and Perelman’s work. (No, I don’t use hyperbole in my analogies. They’re both mind-blowing when you see them, right?)
Polle et al. does teach a skilled reader how to knit multiple layers at once, with sufficient detail so that those who came after (like us) can read it and repeat their steps to arrive at the same result. That is the purpose of the description in a patent: the patentee must provide enough detail so that their invention can be replicated by others, once the patent monopoly expires. You’ll note that Polle et al. assumes a certain level of ability in the reader; there are no instructions about how to cast on, or how to form knit stitches. That assumption of a certain level of skill is fine, because patents aren’t directed to the general public, but rather to a notional addressee who is assumed to have the requisite skill set to understand the description.
Thus, when patents are litigated, expert witnesses are commonly called on to explain to the court just what would be understood by the terminology of the patent. I couldn’t say what the pay would be like in the knitting field because, well, this just doesn’t happen; but in other fields, expert remuneration is far more lucrative than, say, designing and publishing patterns.
Comments (10)Knit Stix, patented
Well. Kudos to Helen Jost. Not only did a design patent issue last month for a species of knitting needle with a measuring scheme marked on it, but earlier this week the USPTO saw fit to issue a patent for a craft needle with measuring capabilities and method of use of same.
Here are the broadest claims that issued. Yes, they’re long. Read them anyway. You may find that some figures will help:
1. A method for measuring the dimensions of a knitted work-piece during the fabrication thereof, said method comprising the steps of:
obtaining a pair of identical knitting needles, a first needle and a second needle, each of said needles comprising
an elongated member substantially circular in cross section and having a proximal end and a distal end, the proximal end being substantially pointed;
head means permanently disposed at the distal end of the elongated member, said head means being larger in circumference than the elongated member for preventing stitches held on the elongated member from sliding off the elongated member at its distal end;
at least one measuring scale permanently applied longitudinally to the surface of the elongated member, said at least one measuring scale having as its zero point the distal end of the head means and being marked in equidistant intervals numbered consecutively toward the proximal end;
obtaining necessary yarn and instructions to construct the work-piece;
casting on the requisite number of stitches and proceeding to knit a portion of the work-piece;
completing a row so that the first needle holds the entire work-piece and the second needle does not hold any stitches; spreading the work-piece evenly along the first needle;
maintaining the first needle in a horizontal orientation;
holding the second needle in a vertical orientation with the proximal end pointed downward and bringing the distal end of the second needle upward to the first needle until the distal end of the head means touches the first needle and the second needle is adjacent to and parallel with a vertical edge the work-piece; and
measuring the length of the work-piece using the at least one measuring scale on the second needle.
I broke up this first claim with carriage returns to try to make it a little easier to read. This first claim appears to be directed to the method of measuring the length of the work as it hangs from the needles: the first needle, holding the “entire work-piece”, is held in a “horizontal orientation”, and the second needle, which “does not hold any stitches”, is held in a “vertical orientation with the proximal end pointed downward and bringing the distal end of the second needle upward to the first needle” until the distal end of the head of the second needle touches the first needle, and the second needle is “parallel with a vertical edge of the work-piece”; then a measuring scale on the second needle is used to measure the length of the work, since the “zero point” of the measuring scale is at the distal end of the head.
4. A craft needle for use in the fabrication of a work-piece from a continuous filament, said needle comprising: an elongated member substantially circular in cross section and having a proximal end and a distal end; and at least one measuring scale, taken from the group consisting of the metric scale and the English scale, permanently applied longitudinally to the surface of the elongated member, said at least one measuring scale having as its zero point one end of said elongated member and being marked in equidistant intervals numbered consecutively toward the opposing end, head means permanently disposed at the distal end of the elongated member, said bead means being larger in circumference than the elongated member for preventing stitches held on the craft needle from sliding off the needle at the distal end; a depression in the distal surface of the head means, said depression conforming to the shape of the elongated member and capable of receiving the elongated member, and the low point of said depression representing the zero point of the at least one measuring scale, whereby the craft needle may be used to measure the dimensions of the work-piece as it is being fabricated.
6. A pair of identical knitting needles for use in the fabrication of a work-piece from a continuous filament, each needle comprising: an elongated member substantially circular in cross section and having a proximal end and a distal end; head means permanently disposed at the distal end of the elongated member, said head means being larger in circumference than the elongated member for preventing stitches held on the elongated member from sliding off the elongated member at the distal end; and at least one measuring scale permanently applied longitudinally to the surface of the elongated member, said at least one measuring scale having as its zero point the distal end of the head means and being marked in equidistant intervals numbered consecutively toward the proximal end; whereby the knitting needles may be used to measure the length and width of the work-piece as it is being fabricated.
Two things spring to mind when reading this: first — and please don’t think that I’m construing these patent claims or providing a legal opinion, here — it seems that these claims specifically exclude circular needles, which of course have no head means, and secondly, the zero point has to be aligned with a specific position on the needle, such as that shown here:
or in Figure 10, at that “low point” of the depression.
For interest’s sake (because of course I am interested by this sort of thing), I took a peek at the claims that had been rejected by the Examiner. Here are some examples:
The original claim 1, as filed, read:
1. A craft needle for use in the fabrication of a work-piece from a continuous filament, said needle comprising:
an elongated member substantially circular in cross-section and having a proximal end and a distal end; and
at least one measuring scale permanently applied longitudinally to the surface of the elongated member, said at least one measuring scale having as its zero point one end of said elongated member and being marked in equidistant intervals numbered consecutively toward the opposing end;
whereby the craft needle may be used to measure the dimensions of the work-piece as it is being fabricated.
This seems a little more familiar, doesn’t it? In this claim, the zero point is “one end” of the needle.
This claim was rejected by the Examiner, who cited Kohler, US Patent No. 2,187,039. This is basically what the invention in that patent looked like:
In response, claim 1 was amended by the inventor to read:
1. A craft needle for use in the fabrication of a work-piece from a continuous filament, said needle comprising:
an elongated member substantially circular in cross section and having a proximal end and a distal end; and
at least one measuring scale, taken from the group consisting of the metric scale and the English scale, permanently applied longitudinally to the surface of the elongated member, said at least one measuring scale having at its zero point one end of said elongated member and being marked in equidistant intervals numbered consecutively toward the opposing end;
whereby the craft needle may be used to measure the dimensions of the work-piece as it is being fabricated.
As you can see, the metric/English scale limitation was added, and it was argued that Kohler didn’t teach that the markings on the needle were numbered consecutively, or with any numbered scale. Kohler does actually state that the markings could be “spaced one inch apart” with “half-inch markings” in between, and even with “intermediate quarter-inch markings”, but Kohler doesn’t actually describe a needle with numbers on it. But the Examiner didn’t buy that argument, and issued a final rejection against that claim, and against some others that were dependent on it. However, the Examiner did allow the claim that I reproduced at the beginning as issued claim 1, as well as a dependent claim based on that. So, in response to the final rejection, the set of claims based on this rejected claim 1 was cancelled by the inventor, and she stuck with the two claims that the Examiner allowed (and added four more, including the issued claims 4 and 6 that I also reproduced).
So there you have it, a brief, albeit incomplete, examination of a patent file history. You can wake up now.
Among the other things I had said earlier when I put up that patent from the 50s, if you recall, was that “for all I know the Knit Stix patent application claims exclusivity over some other aspect that isn’t readily apparent from their product description, although I can’t imagine what that could be…” And see? I didn’t imagine what that could be. And in my defence, it wasn’t in their product description, either; the product they’re offering for sale doesn’t have the features recited in the claims that were allowed. If the pictures on the order page are accurate, then you can see that the zero point is located below the needle head, and there is no depression or other shape in the head that is shaped to receive the “elongated member” of the other needle. The needles that Knit Stix is currently selling seem to have more of a resemblance to Hadler, except the scale runs from head to tip, not from tip to head. Helen Jost has apparently devised an improvement on an old device to improve the accuracy of measurement.
I do wonder about the feasibility of manufacturing needles with a precisely positioned zero point; if you look at the examples on the ordering page, the zero point on the lavendar pair is closer to the head than on the golden yellow pair; what if the heads of the needles, which here are made from a different material, were mounted at varying depths on various needles? Could this imprecision be avoided if the needles were manufactured from a single material, say, molded from plastic?
Patents don’t depend on whether the invention is actually made or used; it can’t be invalidated on that basis (as long as it’s possible to reproduce the invention described and claimed in the patent; there’s a case in Canada in which an applicant was required to submit an actual, working model of his so-called invention, a death ray gun, because the Examiner didn’t believe it would actually work. He didn’t, so he didn’t get a patent). But unless someone actually makes the article that is described by the claims, then there wouldn’t be any infringement. So, I guess it remains to be seen if someone will actually practise the claimed invention.
If you want to look up the history of this patent application, it’s available online. Go here, and type in 11/217,874 as the application number, then select the tab “Image File Wrapper”.
At this point, we don’t even know if a corresponding application has been filed in any other countries because if it had been, it would still be confidential until March 2007: it’s possible to find out if she filed an application elsewhere claiming priority to this US application, but likely not without spending money, and I’m not going to do that.
And for the record, I am still totally against blindly measuring your work with a measuring guide of any sort while it’s still on the needles. If all you need is an estimated length, fine; but if you’re knitting to match a specific length, I’m still firmly on the know-your-gauge-and-count-row side.
Comments (6)Calibrated knitting needle followup
If you read the comments to this post, you might have seen two interesting ones:
First, a comment from one of the women behind Knit Stix, Aleta Frost, stating that they had a design patent, that they expect to receive a utility patent, that their product was patented, and inviting me to contact her for more details.
Secondly, a comment from Charlotte, who used to have some of these old calibrated needles in her possession, so they were produced and the markings were sufficiently durable, contrary to my speculation.
In response to the first comment, I sent an e-mail to Ms. Frost asking for details and explaining where I found the prior art patent. She did drop me a note this morning to thank me for my continued interest, but I didn’t get any more details — she just said they’ve contacted their lawyers about what I wrote. (Understandable — if the applications were filed this year, they’d most likely still be unpublished, and you’d want to consult your lawyer before divulging information about an unpublished application.)
Comments (0)To save time, take the time to yadda, yadda, yadda
A product review I read in the most recent issue of INKnitters (the Fall 2005 issue — the website’s a bit out of date) sparked a vague memory.
The product in question was Knit Stix: straight knitting needles in colourful shades of anodized aluminum with 12 inches’ worth of ruler markings. The product review gushed about its why-didn’t-I-think-of-that-ness, but I thought, yes, it’s a good idea (although I probably wouldn’t avail myself of their benefits because I’m strictly a circular needle type, with the odd recourse to double points), but haven’t I seen that before? (Not in Kim’s recollection of photographs of striped DPNs circa WWII, but rather in one of my patent forages.)
And indeed, I had…
Comments (5)A really, really brief lecture
Shannon asked about what’s involved in challenging a patent. Here’s a really, really brief dissertation on how patents get issued, and how they get busted.
(more…)
