Patent Nonsense?

Tomorrow the Supreme Court will consider a case involving a diagnostic test patent that covers not just the testing method but the biological fact on which the test relies: the correlation between high levels of the amino acid homocysteine and specific vitamin deficiencies. Among other head-scratchers, the case raises the question of whether a doctor can violate the patent simply by performing a homocysteine test and thinking about its nutritional relevance. In yesterday's New York Times, Michael Crichton mocked the patent as an example of intellectual property overreach.

Comments

[Ghost]

Presumably the doctor is charging somebody to do that thinking. So whatever the problem is here, it would not seem to be the mere fact that thoughts can trigger an economic obligation.

Would the doctor be thinking about the nutritional relevance of this test here in the year 2006 if the patent inventors had never been born?

[Hakluyt]

Ghost,

In part the issue is whether one can patent natural processes (e.g., photosynthesis), or acknowledging those processes.

[Jersey McJones]

This is like patenting genes. Complete and utter nonsense (and creepy psuedo-capitalism to boot!)

JMJ

[Warren]

Would the doctor be thinking about the nutritional relevance of this test here in the year 2006 if the patent inventors had never been born?

Right, because the patent holders were born thinking about the nutritional relevance of the homocysteine test. This may be shocking to some, but I've been crying about intellectual property overreach for years. I've seen many examples of equally out of control patents.

There may be a place for patents in my libertarian utopia. But patent law as it exists now is nothing more than a tool the establishment uses to keep itself in power at the expense of everyone else.

[R C Dean]

As a big fan of property rights in general, I tend to be less of a patent-hater than many in these parts.

Still, this is ridiculous.

[Timothy]

I'm with RC and JMJ: this is like patenting a gene. Hak's analogy to patenting photosynthesis* is also apt. This is totally ridiculous.

* Can you imagine? "You must give up your ficus because the Kreb's cycle is a patented and copyrighted product of Phillip Morris USA."

[Modern Day PJ Federico]

In part the issue is whether one can patent natural processes (e.g., photosynthesis), or acknowledging those processes.

Calling a process "natural" means nothing. Photosynthesis in a farmer's corn field is a commercial process. E equaling em cee squared in a nuclear reactor is a commercial process. Ditto all the penicillin that does not get made by accident (basically everything after the 1st batch).

I think people get bogged in these patentable subject matter question because it is an easy subject to discuss and everybody gets to be an expert. Natural or non-natural. Software versus hardware. Business method versus industrial method. Baah. There are more important questions:

how can one tell if an "invention" is obvious (and therefore not patentable)?

What proportion of patents out there now would be considered as obvious (under whatever standard you prefer)?

How hard is it to do a decent obviousness determination? What kinds of work is required and how much of it?

How I feel about the patent criticized in this thread comes down largely to how obvious or not-obvious it may be. How I feel about that depends upon what the doctors would be doing if the patent inventors had never been born. That is a question few think to ask. It is a HARD question to answer.

[Number 6]

Attn: My company has patented the Weak Nuclear Force(tm). Anyone making use of the Weak Nuclear Force (tm) is required to remit royalties to Newtonian Physical Properties, Inc.

[R C Dean]

PJ - something has to be "novel" to be patented.

Photosynthesis and other "natural" processes which predate the evolution of homo sapiens are pretty much the definition of not novel.

The same with "the correlation between high levels of the amino acid homocysteine and specific vitamin deficiencies."

[MP]

Photosynthesis in a farmer's corn field is a commercial process.

No. Don't confuse how a process functions with how someone controls that process.

[MP]

(dammit...I should have checked the link underlying "Modern Day PJ Federico". I don't really want to get into an argument with Dave W over this. sigh...)

[anon2]

R C Dean,

The "property" in "intellectual property" is the same "property" in tobacco quotas.

Patents do not allow for independent discovery, so even if you've never heard of a particular patent, you're prohibited from using your physical property in a manner that violates that patent. As such, intellectual property rights conflict with physical property rights, just like tobacco quota rights conflict with the physical right to grow what you want, in whatever quantity you want on your farm.

People who are opposed to patent rights, are not necessarily opposed to property rights, in fact, they may be people who think physical property rights are more important than government quota systems that are designed to rig a system in favor of a particular outcome.

[andy]

Well, I suppose that even a broken clock is right twice a day, but I wouldn't get into the habit of citing that intellectually dishonest piece of shit, Michael "Denies Extremely Broad Scientific Consensus on Global Warming for Personal Gain" Crichton, too often, Jacob.

Or is he some sort of hero of yours?

[To RCD and MP]

yes, something has to ne novel. It also has to be enabled. And meet the written description requirement. And meet the patentable subject requirements under 35 usc 101, such as they are. And it must meet foreign filing requrements if foreign filed. Must also avoid the prohibition on printed matter (that one may not be around too much longer). and meet the rarely-discussed "inventor believes herself . . ." requirement of 112 first paragraph. Must meet the best mode requirement. The no single means claim requirement. Must not be subject to double patenting. If we lump in ununforceability with invalidity then we need to mention inequitable conduct, too. There are probably more, but:

1. RCD, I know more about patent law than you ever will. It is not close. Never will be.

2. I picked out "obviousness" because, after reading hundreds and hundreds of patent cases and articles over the past decade or so, I have become convinced that this jurisprudentially neglected piece of patent law is what is causing the most problems, especially most problems that laypeople luv to recast as patentable subject matter problems. Case in point: the patent on this thread.

3. side note to MP: isn't all commerce and industry about controlling natural processes? Whether the control renders the process Non-natural or natural-but-controlled seems like a meaningless and semantic issue to me. Once its commerce, the patent system does and should apply (until such time as we decide to get rid of it).

4. Sophisticated side note that I fear will go over MP's head, but I don't know MP at all so here goes: If the patent is written broadly enuf to cover the process in the exact way it occurs in nature (nature not controlled by man), then the patent would be invalid for (RCD's favorite grounds!) want of novelty.

[Evan]

I was about to disagree with PJ here, but then I thought about the Library of Babel---a library with a vanishingly vast amount of books. In this library, say, the limit of the number of pages in a book is 500; the typeface, page size, and language remain constant. Now, imagine how many possible books there would be. Nearly infinitesimal (sp?). But finite, no doubt, because of the limits. Eventually, just by chance, various masterpieces would appear. But you would also have books in which you were 1 word away from that masterpiece, or a comma was in the wrong place. To use Dennett's example, imagine if Moby Dick opened: "Call me, Ishmael!". Completely changes things.

Anyway, my point is that Federico is right. In this universe of laws and limits (of which there are plenty), there is only a finite amount of possible outcomes. Even the most complex inventions, or even masterpieces of art & music, are selected from a vast library of possibilities. The question then becomes, is it reasonable to patent something that is so obvious, instead of being vanishingly possible. Sure, a painting by Monet should be "patented" or copyright-protected, because, given our lifespans, and given the rather vast sea of possibilities in terms of the medium used, it is nearly impossible that one could come up with the same thing again on his own. However, discovering something like gravity or the correlation between a disease and a vitamin deficiency, well, it's a pretty obvious fact, in terms of patentable roads to facts.

[ha-ha]

Does anyone else find it amusing that someone making a "sophisticated" point misspells "enough"?

Ha-ha!

[P.J. Federico's Copy Editor]

Question mark goes inside the quotation mark.

[ha-ha]

Why would it? I'm not quoting the question mark.

[PJ's Clerk, Giles]

Forgot the definiteness requirement on my laundry list of anti-patents. D'oh -- that is another one that, like obviousness, is due for some expanding.

[MP]

Whether the control renders the process Non-natural or natural-but-controlled seems like a meaningless and semantic issue to me.

The issue is if the patent doesn't just cover the control, but the underlying facts of the process which is being controlled. I wasn't trying to play the natural/non-natural angle.

I don't know MP at all so here goes: If the patent is written broadly enuf to cover the process in the exact way it occurs in nature (nature not controlled by man), then the patent would be invalid for (RCD's favorite grounds!) want of novelty.

"would be" assumes that the USPO and the courts are doing their job. I'm not so optimistic.

[Timothy]

Dave: Depends on whether it's a partial quote or not, at least according to my handy AP syle guide. The point above made by ha-ha is a partial quote at the end of a sentence, therefore the question mark goes outside the quote.

I know more about punctuation than you ever will. It is not close. Never will be.

Evan: I believe you mean "infinite" not "infinitesimal". And your point may be invalidated by the latest findingings in cosmology.

[Grounds for invalidity besides]

Misjoinder of inventors (but I think sections 116 and 256 make this a hard one to violate)

[fyodor]

Question mark goes inside the quotation mark.

Why would it? I'm not quoting the question mark.

I was taught in school that punctuation always goes inside the quotation marks. But I always hated that rule and follow it discriminately.

[Timothy]

style guide rather. Learn to spell, lad. Learn to spell.

[ha-ha]

Punctuation and spelling are gay. Fags.

[Mr. F. Le Mur]

"Microsoft Patents Ones, Zeroes" - The Onion

I recall reading, about 40 years ago in Popular Mechanics, about a guy who got a patent on some hinged wheelbarrow handles and also ended up with a/the patent on the wheelbarrow itself. "Prior use" struck, though.

Question mark goes inside the quotation mark.
"Quotation mark" should be plural. "Ha ha" isn't hyphenated.

[smacky]

While we're on the subject of patents, the US Patent Office is trying to recruit me. Should I apply? Anyone have any opinion on this?

[Timothy]

Smacky: that depends. Do you love yourself? And do you think you can maintain your libertarian street cred while working for the government?

[anon2]

With regards to obviousness, there's a critical question of "to whom?" In U.S. law, it's "person having ordinary skill in the art" (PHOSITA). There are lots of things that are obvious to people in the top tail of the bell curve for any art/science are not going to be obvious to PHOSITAs. However, there are enough people in that top tail for independent discovery of things that are obvious to the discoverers, but not to PHOSITAs to cause serious problems.

It's frustrating to be prohibited from doing obvious things when someone else has patented them first. Even if the patent office were doing their job absolutely correctly, as long as the test is whether or not an invention is obvious to PHOSITAs, highly skilled people will still run into huge problems.

[Peege and His Pal Sponge-Bob]

I will defer to Timothy on the punctuation issues.

On the issue of obviousness, I think MP is beginning to see why I get concerned. Yes, the patent system is guilty of a lot of overreach, at least in the sense of companies collecting a lot of patents for things there shouldn't be patents for.

People like simple solutions for this problem, and they usually don't want to get their hands dirty with a lot of patent reading b4 coming to an opinion. For these people, usually the first rhetorical position is: we should do anyway with patents. No patents, no copyrights.

For many (myself included), this idea of wiping out the patent system seems a bit radical. Wheter it is ust lucky timing or not, the patent and copyright systems have been timed to occur during a veritable explosion of inventing and writing in the times and places where such legal regimes have been implemented (we are basically talking about the West during the Age of Reason here). So if we can't wipe the patent system away, then what can we do to reign it in?

I believe that this is the point in the internal dialogue, where people try to reframe patents as a civil rights question. I want my doctor to have the right to think. I want my grandmother to have the right to make sandwiches how she sees fit! (see the link on that one). I don't want patents on golf swings. On yoga. On business methods. No patents on anything I may have occasion to do or hire soemone to do.

This is where ol'Peege and his d'sastrass spelling comes in. Patents aren't a civil rights issue. Smucker's ain't going to come in and bust your grandmother unless she has started up factory level production, like the Grandma on Future-ama. If your doctor is charging your insurance $500 an hour to think about your tests: (1) that ain't personal; (2) that is commerce; and (3) if the inventors really made that thinking better, then, yeah, it is good incentivization to give them a piece.

Of course, the thing about the doctor having his $500 per hour thinking improved really begs the question. Did the inventors really improve the thinking? Really? Really really really?

[Peege]

Forgot the pbnj link:

http://en.wikipedia.org/wiki/Peanut_butter_and_jelly_sandwich

[Questioner of Anon2]

How would you rewrite obviousness law then, in light of your critique?

Would you use the basic same test, but rewrite the applicable legal standards to replace the PHOSITAs with better experts?

Or would you do something else entirely?

[Eddy]

I'm going to patent the correlation between bears shitting and the shit being in the woods. Damn, now I've gone and disclosed it! Curse you public domain!

[anon2]

Dave W,

One obvious thing that could be done would be to 100% categorically deny a certain class of patents for a set period of time and then try to estimate the change thereafter. Say a ban on software patents for five years. There's a (to me) tiny chance that the results would be so horrific that the test would need to be curtailed. There is also a tiny chance that software development will increase so suddenly that the average observer will chalk it up to the ban. Most likely (to me) is that software development, including new algorithms that might have been patentable anyway, would proceed more or less at the pace that would be expected and that the experiment could then be moved on to business method patents.

There are ways to make the experiment even more useful, like allowing people to file software patents during the ban and then doing analysis afterward to see how much duplication crops up, but realistically the patent office could not do such a study.

[anon2]

Questioner,

My 12:57 post was made before I read your 12:41 post.

The short answer is "I don't know."

If I had the authority to make sweeping changes, I'd set up the infrastructure to do an experiment and then eliminate software patents and business patents and see what happens. That wouldn't change the PHOSITA problem in other areas, but if we find out that knocking out entire classes of patents has no ill effects, there's always the chance that we would indeed, over time, knock out all of them, in which case the PHOSITA problem is moot.

If there were a cost-effective way to allow for independent discovery to affect the process, there are a few different rules that could be put in place. Upon independent discovery, the independent discoverer 1) shares in the royalties, or 2) is able to use his methodology royalty free or 3) causes the patent to be voided. Each of the three have good points and bad points, but the biggest problem is that although independent discovery clearly exists, determining when it has actually occurred and when its a matter of copying is not something that can be done inexpensively in all cases, and taking into consideration the possibility of spying, perhaps no cases.

However, just because I don't have a solution to the PHOSITA problem doesn't mean it's not significant. If I were looking at a bridge that had car-sized holes in it and every once in a while a car plummeted through the hole, I would know there is a problem with the bridge, even though-not being an engineer-I wouldn't know how to fix it.

[Further to Anon2]

anon2,

I like those thoughts, although there are some practical problems.

1. Why ban software only? How about everybody's patents for the ban period. I don't see anything special about developing software that would apply to developing drugs or electric cars or less collapsible skyscrapers or whatever. You don't have any special connection to the software industry do you? Maybe different fields of technology, due to their individual circumstances, call for different legal treatments, but we wouldn't really know that unless we do a braoder ban and then compare the relative wreckage.

2. OVERGENERALIZAZATION ALERT: Software patents were basically banned in the US, until the US courts gradually changed their minds circa 1990-95 and unbanned them. Many software companies, like Microsoft, proceeded without patents an built successful software businesses. Other companies continued to fight for patents. Some got patents. Others fought long enuf until the law changed in their favor. Would be interesting to see whether software ppl think their technological progress moved faster in 1980-90 ("pre-patent") or in 1995-2005 ("post-patent"). Certainly there were some good results of the anti-patent atmosphere, not the least of which was Benners -Lee throwing his invention into the public domain. On the other hand, a technology-hostile company called Microsoft managed to get more marketshare than a technology-hostile company should in a technology-driven sector.

3. The US software patent experience 1975-2005 is probably not clear enough to draw great conclusions, but it does show how difficult it would be to really impose the strategic, experimental ban you are talking suggesting. Short answer: applicants would it ignore and fight the ban until it went away.

[Pro Libertate]

smacky, what position are they recruiting you for?

[quasibill]

Anon is hitting all of my main points, but I'd like to emphasize the biggie - patents are not "property" in any sense other than government fiat. The knowledge subject to patent rights does not fit any of the classical requirements for property.

And it's a big mistake to conflate copyright with patent, as some here are doing - they are very different arenas with very different rules and underlying rationales. I'll repeat it again - you can make a solid argument from first principles for copyright rights (but note, they are more a function of contract rights than property rights), but not patent rights. Patent rights are anti-property and anti-market.

[thoreau]

I don't claim any deep expertise in patents, but as people talk about whether and how to change the system, there's one thing we should remember:

Patents weren't just put in place to create/protect/recognize a property right and incentivize invention. Patents were put in place to incentivize disclosure of an invention, so that the rest of the world could learn the juicy details and find ways to further innovate.

As people debate about whether and how to change the system, don't just ask whether or not somebody deserves to have a particular invention protected so he can make money. Also ask whether a proposed change would have inhibited or facilitated the public dissemination of that idea.

I have no strong views on the matter, but I think it's good to remember the big picture motivation.

[Still Further to Anon2]

The history of business methods patents is a lot closer to the clean version of your proposed banning experiment. What I mean is that the US law of business methods patents was considered clear until 1998 (business methods patents illegal) and then clear again after the State Street case (business methods patents suddenly legal). Although some people (like the State Street plaintiff) did ignore the ban, I remember that the ban on biusiness methods patents was, indded, taken pretty seriously when companies made actual patent prosecution decisions.

Has there been some grave, quantifiable harm in the business method sector since 1998 (we've had 7 year now)? (I say no.)

Since we don't see any obvious harm in the business method sector relative to pre-1998, does that suggest patents (in all sectors) are less harmful than you thought when you woke up this morning? I mean, I am basically telling you that we ran your experiment already! Now that's service!

[anon2]

Why initially only ban software patents? Look to your own #2. The history of software development is an open book. There are huge parallels between algorithms and pure math. We know that math was developed quite nicely without patents. We know software was developed quite nicely without patents. Developing software has an incredibly low barrier to entry, so low that the PHOSITA problem is huge. At the other end would be pharmaceuticals, with a high barrier to entry, and is also a topic I know very little about.

Here's one of the early software patents. Back then they played the game that they were patenting a machine ... a machine that incorporated a particularly algorithm / data structure. However, they made it clear that if you had your own computer with bit-mapped graphics, that using their algorithm would be an infringement of that patent, regardless of the origin of the hardware. I'm pretty familiar with some aspects and certain histories of software patents.

If we ever meet in person, I'll be happy to flesh out my ideas, but H&R isn't a good venue for a larger discussion. I just wanted to make a few points, which I've already made.

Thoreau,

Disclosure is not much of an issue for some classes of patents, such as (I'll bet you're surprised I use these as examples) software or business methods. People reverse-engineered the data structure in the region patent, for example. Also, the U.S. Constitution specifically says "To promote the progress of science and useful arts..." which is neutral with respect to disclosure vs. remuneration. Again, if we meet in person, I can provide a lot more info than I'm going willing to type.

[smacky]

Pro Libertate,

"Patent examiner". I guess I shouldn't say that they are recruiting me in particular, now that I look at the email again, it was probably sent by a spambot. Meh.

[anon2]

Dave W.,

I'm aware of the history of software patents and business patents. I can't find a single reason to believe that business method patents are needed. I don't think they promote the progress of science and useful arts. I don't think we'll have a shortage of innovation without them. I don't believe that there will be crucial business methods developed and used but that will be hidden from society.

This is a libertarian blog. If you are advocating such government intrusion into how people can run their own businesses, suggesting that you haven't yet seen any personal harm is not going to be particularly convincing.

I mean, I am basically telling you that we ran your experiment already! Now that's service!

No. That's condescension and ignorance.

[R C Dean]

I'd like to emphasize the biggie - patents are not "property" in any sense other than government fiat.

Not so fast, big fella. Intellectual property in general, and patents in particular, are strongly analogous to real property.

Both are essentially abstractions defined (and limited) by government fiat, after all. Both are arbitrary grants of limited monopoly rights.

IP and patents also share a number of traits with intangible property such as undivided partnership interests, common stock, and the like.

[Einstein, Frustrated, Wifebeat]

We know that math was developed quite nicely without patents.

Compared to what. They got up to the cusp of calculus in ancient Greek times and then basically lost all of that knowledge for 100s and 100s of years.

I can imagine math moving along much more consistently when it did. VCertainly if the ancient Greeks had a good patent database of their math, then Renaissance mathematicians would not have had to wait so long. The intervening Barbarians might have even done some good work in the interim if the early work had been memorialized better.

Math has developed just fine, compared to what?

Furthermore, the famous mathematicians of history did not exactly work in what was considered commerce or industry. As an HnR reader, that should trouble you. We here know that commerce and industry provides the bestest most rational incentives -- not the Ivory Tower, not government research, not the media. Once you understand that, there is another way to look at the history:

industry and commerce effectively excluded the mathematicians by failing to recognize the great ones with patents -- therefore the great matematicians of history didn't work for the part of the private sector that we wub here -- therefore, math has moved a lot slower than it could or should have.

[Peege]

This is a libertarian blog. If you are advocating such government intrusion into how people can run their own businesses

I have my own proposals about how the patent system should be cut back. One reason I think that patents aren't as harmful as made out is that I saw the law of patentability in software and business method change drastically during my career without much of anything in the way of clear, easily-observable effects (good or bad) happening. For an alleged empiricist, this seems to make suspiciously little impression on you.

As far as private(!!!) patent intrusions into how you run your business, the answer is simple: just going back to doing it how you did it b4 the patentee got her patent. Intrusion gone. Works everytime.

Finally: my own agenda is that I think there should be many fewer patents and have my own ways to de-incentivize at the appropriate margins without trashing the whole system and without making a political spoils game out of the reform. Its not like I am some kind of patent humper, ya know.

[john smith]

'Gosh' means 'cash' says the modern scientist. Screw him, I'll patent his asshole gene and charge him for being an asshole.

[Wild Pegasus]

What I mean is that the US law of business methods patents was considered clear until 1998 (business methods patents illegal) and then clear again after the State Street case (business methods patents suddenly legal).

This is wrong. Most of the commentary in the patent community prior to 1998 had said that there basically was no "business method exception". The cases generally cited as showing the "business method exception" were always decided on some other ground. However bad that decision might turn out to be, the Federal Circuit Court of Appeals got the law right. Now it's Congress's job to step in and ban it.

For anyone who's curious, google "State Street Bank decision".

- Josh

[john smith]

'Gosh' means 'cash' says the modern scientist. Screw him, I'll patent his asshole gene and charge him for being an asshole.

[bubba]

Many things are "obvious" after they are disclosed.

I am an author on several gene patents. This in no way prevents you from using them in your daily routine as a human being. You are even free to extract the native protein from your own liver (or those of donors) for commercial use. However, if you wish to use the sequence that I discovered, as part of your method to produce this protein in a vat, then you should expect to pay my company a royalty, as we were the first to disclose the composition of this particular enzyme, receptor, or whatever.

The fact that the US Government later paid to repeat my work is of no consequence.

My experience is that the visceral objection to "gene patents" is based on a lack of familiarity with how biotech works, and the enablement one receives from starting with the correct protein or nucleotide sequence.

The correlation between homocysteine levels and whatever disease state is an important observation with predictive power. The person or company that discovered this should be compensated when this is used for diagnostic work.

This is no different from pharmaceuticals. Pfizer didn't invent the chemistry behind a given drug. They were simply the first to observe that compound A binds enzyme B and yields medical benefit C.

[john smith]

'Gosh' means 'cash' says the modern scientist. Screw him, I'll patent his asshole gene and charge him for being an asshole.

[Pro Libertate]

smacky,

Hmmmm. Did the PTO, in this e-mail, ask you for bank account information?

Just checking 🙂

[Modern Day Peege]

Wild Pegagsus,

Disagree. Most of the pre-1998 commentary said that there was a business method patent exception.

As a young IP law firm associate, I used to get in trouble for suggesting that the business methods and printed matter exceptions were non-statutory and therefore vulnerable. Going back and reading State Street, it is clear that the Fed. Cir. was running from their prcedents here and not embracing them. Then again, I predicted the same thing about the doctrine of equivalents (that is, non-statutory, going away), but the Warner-Jenkinson courts showed me that I can't always predict the patent law.

[Questioner of Bubba]

Bubba,
Would we have the subject of your gene patents now if you had never been born? Would other scientists have figured out your contributions by now or no, do you think? How much longer would all that have taken in a Bubba-less world?

[smacky]

Pro Libertate,

Actually, no. They asked me for my measurements and a few naked pictures. They said they're going to make me a star.

[calliope]

bubba,

there's an important distinction to be made in the examples you assert are identical. pfizer didn't patent the observation that about A, B, C, but rather the specific (or class) of compounds that effect the observed phenomena. there is a physical invention at work here. in the case of the correlation between homocysteine and disease state there is no invention, only observation. an observation/process without a physical invention is patentable if

"a claimed process applies the [X] principle to produce a useful concrete tangible result without pre-empting other uses of the [X] principle"

AT&T v Excel

The case involved whether mathematical algorithyms were patentable subject matter

certainly tying a diseases state to a particular lab value could have useful, concrete, tangilble results. so perhaps according to precedent the instant invention IS patentable.

I haven't followed this current case so carefully, but i suspect it is that precedent that SCOTUS wants to look over

[MP]

However, if you wish to use the sequence that I discovered, as part of your method to produce this protein in a vat, then you should expect to pay my company a royalty, as we were the first to disclose the composition of this particular enzyme

Scientific discoveries about the nature of the universe should not be patentable, as it puts up severe barriers to open research efforts (universities, public bodies, etc.). If a company discovers a gene sequence and wants to capitalize on it before anyone else does, they should keep it secret. The drug they create that capitalizes on it should be patentable, but the sequence should not.

[Comment for Both Pegasus and A]

After the business method patent law changed, there was a rash of articles about how the number of business methods patent application filings were increasing. This shows that some people took the business method patent "ban" seriously. It also means that if some horrible thing happened, then Anon2 should have the empirical evidence he is looking for (unless its just not there!).

[Pro Libertate]

smacky,

In that case, I'd accept their offer. What the heck, it beats running in the rat race.

[anon2]

Dave W,

You're woefully ignorant of the effects of software patents, and you wear your ignorance like a badge of pride. People regularly write software and inadvertantly infringe on patents, often patents that were applied for before they did their own work, but that weren't granted until after they did it. They can not simply go back to doing it how they did it before the "patentee got her patent."

Some will say that it's because the patent office didn't do its job and it allowed something obvious to be patented, but the standard is: obvious to PHOSITA. As long as the patent office is screwing up enough by granting patents to things that are obvious to PHOSITA, it's easy to put the blame on on the patent office, but even if they were to do their job 100% accurately, the problems with people who are particularly skilled in a particular art will remain.

You may have credentials, but if you stand by your claim:

As far as private(!!!) patent intrusions into how you run your business, the answer is simple: just going back to doing it how you did it b4 the patentee got her patent. Intrusion gone. Works everytime.

then you simply don't understand the issues. There is no exception for independent discovery. The requirement that it be non-obvious to PHOSITA is too low of a bar in the software industry. If you are very talented, it's very likely that you'll infringe patents, even without knowing about them. Even if you do patent searches before you begin your project, you can still easily hit a submerged patent.

[quasibill]

"Not so fast, big fella. Intellectual property in general, and patents in particular, are strongly analogous to real property. "

Saying it does not make it true.

"Both are essentially abstractions defined (and limited) by government fiat, after all. Both are arbitrary grants of limited monopoly rights."

No, wrong. Go back to 1L property. Real property evolved naturally, and was a recognition of natural rights. Several things created property rights, with a biggie being "scarcity" - the fact that if I'm using my baseball bat, you cannot be. Patents do not fit this requirement at all - the fact that I independently discover a process does not in any way prevent you from using the same process. In fact, there is a better argument for property rights in air (and hence all sorts of pollution regulation) than there is for ideas, but I don't see many people racing to establish a property regime in that arena.

"IP and patents also share a number of traits with intangible property such as undivided partnership interests, common stock, and the like"

Putting the cart before the horse. How we treat something that we have declared is property does not then establish that it is property in the first place.

So, no, patents are not analogous to real property in any way, shape or form. Sorry.

[quasibill]

Thoreau,

I've had the disclosure argument with Ron previously, but all you need to do is look at the Blackberry dispute - how much "disclosure" was there, really? Is it useful to page through the millions of documents in the reporter to prepare for scientific research, or do you page through your relevant journals?

Also, as anon is pointing out, the issue of disclosure hinges on what those having ordinary skill in the art can be said to know. And if you want to know a little trick about the law, it's that the words "ordinary" and "reasonable" are green flags to litigation attornies. You can find an expert with tremendous credentials to testify that screwing your cat is "ordinary" or "reasonable" under the circumstances. Then you have a non-scientist (either a judge or a jury) making a black/white decision on the issue.

Finally, you need to actually read the disclosure segments of patents, especially in areas like pharmaceuticals. The last thing in the world that they actually do is tell someone else how to replicate the process. What, you think they're gonna pay patent attornies $500/hour for nothing?

[Ghost]

Anon2,

1. Now you are really describing some of the pressing problems with the patent system we have now. That last critique of the patent system is the good one. the critique of my motivations, on the other hand, is pretty far off base.

2. whether you know it or not, most of those problems are not unique to software patents at all. Those problems have potential solutions that are not unique to the software area.

[calliope]

Screwing your cat is perfectly reasonable and ordinary....

in the 11th circuit!

ba-dum-da-DUM

[thoreau]

Is it useful to page through the millions of documents in the reporter to prepare for scientific research, or do you page through your relevant journals?

The question isn't whether the patent application is a useful source of information. The question is whether having the patent makes people more willing to put more information (not necessarily everything, but more) into the relevant professional journals so that others can pursue fruitful lines of inquiry. I've been in conversations and attended seminars where people said that they were holding back on some aspects of their work until the patent was filed. Granted, this has been mostly in physical sciences, not pharmaceuticals.

I have no strong opinions on the patent system, having never been through it myself. I might have something patentable, but it's kind of in a gray area. And in all likelihood it wouldn't be that valuable even if it were patented, so we'll probably just put it in the public domain. At this stage of my career, getting my name on a widely disseminated tool is what matters most.

So, I have no strong opinion on what, if anything, should be done with the patent process. But people tend to talk a lot about property rights and profit as a spur to innovation, and neglect to discuss the disclosure aspect that can produce just as much innovation, as scientists strive to build on pre-existing work.

I don't know enough to say whether patents are the best way to achieve that disclosure objective. I merely toss it out there as something to keep in mind.

[Tom Arnett]

quasibill,

In fact, there is a better argument for property rights in air (and hence all sorts of pollution regulation) than there is for ideas, but I don't see many people racing to establish a property regime in that arena.

Air seems subject to the same general regime as anything else in nature: whatever you can sequester is more or less yours, and whatever you can't more or less isn't. Compressed air, for example, is a saleable product. This isn't meant to detract from your broader argument; in fact, I think it underscores the distinction between real property and artificial monopoly.

[bubba]

Bubba,
Would we have the subject of your gene patents now if you had never been born? Would other scientists have figured out your contributions by now or no, do you think? How much longer would all that have taken in a Bubba-less world?

They would have been discovered later. This is true of all invention. Recall that patents are not valid forever, but only for a finite period of time. Of course in this particular case, our big competitor was a giant Government program. Not exactly typical.

If a company discovers a gene sequence and wants to capitalize on it before anyone else does, they should keep it secret.

We do a lot of that, too. However, the whole point of the patent system is to encourage disclosure, in exchange for a finite period of ownership. In this case, the gene sequence is analagous to a reagent in pharmaceutical manufacture. No, it's not as valuable as the final product, but if it has added value, then I should get something.

With respect to Universities, they are the biggest infringers on the planet. The vast majority of their work is non-commercial, so it's very difficult for a company to justify suing one of them for infringement. A recent example was when a State U patented an item which itself was made by infringing our own patents. Patents we had purchased from another University, as specifically encouraged by Congress. One of our projects is relatively academic, and we thought their item would be useful. They had the balls to ask us for an outrageous amount of money.

My experience with Universities is that they have a callous disregard for patents, unless it happens to be theirs.

The drug they create that capitalizes on it should be patentable, but the sequence should not.

But if I can't patent it, then I won't disclose it, and then Pharma B can't use it. As mentioned elsewhere, Patents are explicitly documented in the Constitution. How many other public policies can say that?

in the case of the correlation between homocysteine and disease state there is no invention, only observation. an observation/process without a physical invention is patentable if

Without a patent, what is the incentive for a company to perform the study that discovers the link between homocysteine and health? Especially when this is a very difficult discovery to make, but a very easy one to reproduce. Just like drugs.

There seems to be a common misperception that all patents are created equal. This isn't true. Companies know that sequence patents are worth less than compound patents, and they are perfectly capable of negotiating some sort of settlement.

Meanwhile, Universities ignore all of them (illegally) and science marches on. There has been at least one case of an individual collecting against a University, but that was only because the University (Duke?) had really crappy IP counsel, and refused to pay him a reasonable fee early.

Again, companies don't find it profitable to sue non-profit entities.

[biologist]

bubba:

if a DNA sequence that contains information to produce a protein can be patented, does the company simultaneously patent all the possible DNA sequences that can produce a protein product with an identical amino acid sequence, or is the amino acid sequence itself patented?

do you have to patent mRNA sequences as well?

sincere questions, I swear

[Kevin Carson]

R.C. Dean,

There's an analogy between IP and legitimate property rights in (as quasibill says) how they're treated AFTER they're defined. But that's where the analogy begins and ends.

Legitimate (tangible) property rights evolved naturally from the fact of possession, and to be enforced require the owner only to maintain physical possession of his own property (with help if necessary). IP, on the other hand, requires calling on the state to invade someone else's property and stop him from using it in ways that a patent holder has been granted a monopoly on.

A legitimate property right gives you a monopoly on the use of specific items of tangible property. A fake (IP) property right gives you a monopoly on using ANY property in certain ways.

[bubba]

bubba:

if a DNA sequence that contains information to produce a protein can be patented, does the company simultaneously patent all the possible DNA sequences that can produce a protein product with an identical amino acid sequence, or is the amino acid sequence itself patented?

do you have to patent mRNA sequences as well?

Our patent lawyers use some boilerplate (and case law) to block the use of trivial variants. "Anything that hybridizes to the claimed sequence with the following stringency..."

We typically file for all the variants we observe in nature, both mRNA and protein. This includes splice variants with significant changes in amino acids, and simple polymorphisms.

With the human genome having largely been published, and a few years experience in what is actually of commercial value, we confine ourselves to proteins where we think the actual protein is a potential therapeutic, to novel splice forms or polymorphisms that have not been published, and where we have some novel biological information about the protein to suggest that we know something that others do not about the therapeutic use of the protein.

When we begin pursuit of a novel therapeutic protein, we actively search for unpublished variants. If we find something that is both unpublished and more useful in our assays, then we feel that we have a stronger IP position.

What value does this really have? I dunno, but Pharma takes it seriously when crafting alliances, so it must have some value, by definition. And, if you hold the sequence patents, you don't have to worry about some other ankle biter trying to extort a cut later.

Genentech is a major player. In sequence patents, not ankle biting.

[bubba]

Phooey. My comments start after: "do you have to patent mRNA sequences as well?"

[science]

A tangentail one here, but what would this

http://www.context.org/ICLIB/IC32/Gilman.htm

model of economic activity mean for the debate?

IP seems to challenge the normal property rights definition sufficiently to warrant other ways to look at the underlying concepts.

[SY]

Some real naivete about the reality of the US Patent system.

An example:

something has to be "novel" to be patented.

No. Something has to be novel to be granted a patent which will withstand a court challenge. Maybe. But under law, an issued patent is presumed valid, and unless you've got a million or two to spare for litigation, you're screwed if you're infringing a clearly invalid patent. Trust me, I know, the hard way. Our system is set up principally to benefit large organizations and IP attorneys.

In Europe and most of the rest of the world, there's at least a chance for outsiders to help the examination process along. Not here. That would impact the principla beneficiaries of the fucked-up system we have.

FWIW, I'm not an attorney, but I've got a few dozen US patents and have been on both sides of the litigation table. Our system is badly, badly broken. The patent at question here may indeed be invalid, but someone is going to have to kick 7 or 8 figures woth of money to a law firm before that can be determined. And with no chance of recovery if they win.

[Robert]

With the human genome having largely been published, and a few years experience in what is actually of commercial value, we confine ourselves to proteins where we think the actual protein is a potential therapeutic, to novel splice forms or polymorphisms that have not been published, and where we have some novel biological information about the protein to suggest that we know something that others do not about the therapeutic use of the protein.

Is the therapeutic use reduced to practice? If not, what showing do you have of utility for the translation products of those genes?

[Still waiting on an answer]

Many things are "obvious" after they are disclosed.

Yes, Bubba. But assume that you had never been born so that you could never disclose the subject matter of your gene patents. How long would we, as society, have had to wait for your breakthru's to come out of another lab? Calling for speculation, but just pick your best rough guess:

a. 2 years?
b. 5 years?
c. 10 years?
d. 20 years or more?

(I am guessing answer a because there are so many people doing gene research, but Bubba would have a better idea than me.)

[calliope]

Your commment above, Still Waiting, is unclear to me. Are you saying:

(1) an invention is (legally) obvious if someone else would have independantly discovered it as well at some arbitrary point in the future

or

(2) the legal standard of obviousness should be changed so that anything meeeting the criteria in (1) is declared obvious?

if you're saying (1), that's not how the law works. the system rewards the first to reduce to practice an invention and file an application. this encourages people to aggressively advance the useful arts.

if you're saying (2) then there will be no patents - as you point out, how long would an application sit in the examiner's office waiting to see if someone else discloses the same invention?

[Still Waiting]

I am saying that I consider the value of patented any invention to society comes from the value in getting that value (early) from the first inventor, instead of waiting for the next inventor. As far as figuring out which inventions are worthy of giving monopolies to (something that should never be done lightly!), the ones that are worthy are ones where society got the invention years (not months) earlier than it otherwise would have. A couple observations on this point:

1. Often we never have a good solid answer to this question because the inventor publishes a paper at the same time the patent application is filed. After the thing is published, it is sensible to presume that later inventors just read the paper, so we don't know how quickly the next inventor would come around.

2. Despite the evidentiary issues identified in item #1, there are inventions that stay secret after patent filing. In these cases, we often do know how long the next inventor would have come around. These facts, when they exist, should be considered as a lot more important than the law currently considers them.

3. When I was in private practivce, most patentees are no more than 2 years ahead of the competition at best. To exchange a one or 2 year "headstart" for a ~17 year monopoly seems like a bad deal for society. Yet the patent law has us making this deal over and over again, and not just in the fields of software or business methods.

4. It would be interesting to see Bubba come round and say that he was 10 years ahead of the next lab. My feeling is that the gene work is pretty well funded in today's world and that nobody is very far ahead of the pack.

5. This question of how far the first inventor is ahead of the second inventor is not currently the touchstone of obviousness law, but it should be. If an inventor is really 10 years ahead of his rival, then you better bet he desrves a big fat monopoly. On the other hand, if she is only a year ahead, then we are better off just waiting for the second team to bring us the new drug (or whatever) unencumbered by 17 years of premium pricing.

6. All this stuff about software patents versus gene patents versus business method patents versus drug patents versus software patents remains the big red herring for people who want to reform patent law without first understanding it. I don't care if your invention is an improved way to make a fire by striking rocks -- if we weren't going to have that knowledge, absent the inventor, then it is a good patent. Conversely, I don't care if your work is on the most complex gene molecule known to nature -- if the next closest team is only 6 months behind, then the inventor has only given us something pretty marginal -- a 6 month lead.

7. Patents are supposed to only incentivize work that isn't being done already. It is not supposed to superincentivize research work that is well-capitalized and duplicative. The current patent system doesn't understand this. the current system should be changed by people who's knowledge runs deeper than harum-scarum articles about whether patents are well-regarded in this business sector or that one.

[Answerer for calliope]

(2) the legal standard of obviousness should be changed so that anything meeeting the criteria in (1) is declared obvious?

I am saying that obvious should depend upon our best guess as to how soon the technology would have come round absent the applicant. If we want to keep the monopoly term at 17 years, then we should probably somehow make ourselves confident that the invention would not have materialized for 5 years. Sometimes we will know how long the "second inventor" would have taken because the patent applicant keeps her application a secret and the "second inventor" actually does invent duplicatively within a year or so of the first inventor's patent application filing.

In other cases, the disclosures of the patent applicant herself infect the market with knowledge, such that there can be no independent invention. However, even here, what we should be asking the experts is how long the technology would have come around with reference to parallel research projects, similar inventions and the like.

In other works, the current law asks how "smart" the inventor is.

My law would only ask how far ahead the inventor is.

Both questions are generally impossible to answer determinatively, but they are different questions and my question is preferable for being: (1) more relevant from an economics standpoint; and (2) less susceptible to subjective, fudgy answers.

More answers to follow, hopefully.

[Answerer for calliope]

if you're saying (2) then there will be no patents - as you point out, how long would an application sit in the examiner's office waiting to see if someone else discloses the same invention?

I suggest that we give applicant's the option of having their thing published as soon as it clears the security dept (right at filing, basically).

Under the current system, applicants would not excercise this option because there is no point. However, under my system, using this early publication option would incentivize the applicant to publish early so that it would not be quite so clear that the "second inventor" was coming up really quick and building up negative evidence on the obviousness question as my proposed law sees it.

I think that is all your questions.

[calliope]

You've argued (if i understand properly) that if an invention is 6 months "ahead of the curve" then the monopoly protection should exist only for those six months...in my opinion, this will encourage even more (!!!) litigation than already exists

as the system stands now, if A is granted a patent for X in front of B, B must litigate the patent, that is, prove that the patent was improperly awarded for whatever reason. Under your system, as I understand it, all B would have to do is prove that it was close to inventing X as well. As far as I can imagine, this would be an even fuzzier question to answer than the current standards for obviousness. This lessens the burden B has to remove A's monopoly, and the lower burden will further intrench patents in the court system, which I view as a negative

[Favorite Lawyer]

You've argued (if i understand properly) that if an invention is 6 months "ahead of the curve" then the monopoly protection should exist only for those six months...in my opinion, this will encourage even more (!!!) litigation than already exists

I haven't been clear. I am saying there should be no monoloply protection at all for inventions that are insufficiently ahead of the "2d inventor" (whether she be real or hypothetical).

If you are 6 months ahead of the 2d inventor, then result (under my proposed system) is no patent: obvious.

If you are only 1 or 2 years ahead of the "2d inventor," then the result under my proposed system is no patent: obvious.

At some point, under my proposed system, an inventor would be far enuf ahead to get and sustain a patent: non-obvious. I threw out 5 years as a number for discussion. This is negotiable. Could have said 10 years if we wanted to scale back patents further. 20 years if we wanted people to stop applying altogether.

You point out that my proposed law is essentially indeterminate. I agree. You are saying that it is less determinate than the law of obviousness that now exists. I agree only to the extent that current obviousness law has no teeth. It is determinate in the sense that if you are defend against a patent based on obviousness, you are probably going to lose, regardless of the facts. Good for patents. Bad for society.

We need to get serious about sorting the useful patents from the bad ones, instead of just throwing our hands in the air and saying, "too many patents, let's scrap the whole system." My special genius is I figured out how to tell the good ones from the bad ones. A way that does not involve a smartness contest.

[calliope]

i only have exp in pharma/biotech, and can not speak to other industries, but in my opinion, the standard of obviousness favors the patent holder rather than the patent infringer. you have to have 1-3 references, which explicitly teach the invention, and in the case of multiple references, there has to be motivation to combine them. this is not terribly common, it's far easier to defeat a patent based on anticipation/novelty

rather than revise 150+ years of patent law, i'd rather strengthen the USPTO. as it stands now, there are not nearly enough examiners, and their performance is often evaluated by how many applications they process rather than the quality of their work. often, this results in applications sailing through without so much as a single rejection.

the standard an examiner must apply to reject a patent is lower than the standard a 3rd party must use to invalidate an issued patent. give the examiners more teeth and perhaps some of these problems will go away

[favorite lawyer]

1. Yeah, the motivation to combine law / motivation to modify law makes no sense at all. It is not a good representation of how inventors actually invent things, so it is not a good model of how to separate the smart inventions from the not-particularly-smart inventions. It is basically a fiction that is used because nobody has come up with anything better (til now).

2. I wish better examiners could do a better job using only the prior art available at the time the patent application is filed, but there are 2 huge problems with that plan: (a) the published prior art available at the time the patent application is not really a good indication of what is coming next in the great, wide world; and (b) since you are having the examiner basically judging how smart the putative inventors are, you would need examiners somewhat smarter than the applicants to make fair decisions about this.

3. I am saying that if you make it relevant what other people are working of independently of the invention before, and immediately after it is filed, then one can glean a better idea of how far ahead the 1st inventor really is. This is not appropriate for the examination stage. It is appropriate later when we are deciding who gets to keep the money for the small percentage of patents that actually became commercially operative.

4. Although a big part of my work is arguing with patent examiners, I do not think this arguing helps society or makes the patent system any more rational than if we just had a no-examination system.