Interesting that this is coming up again. Over a decade ago, we designed and built a synthetic, refactored phage [1]. At the time, we (Drew Endy, Leon Chan, and I) had discussions with lawyers about copyrighting and open-sourcing it. The lawyers we spoke with (Jamie Boyle & Arti Rai amongst others) seemed to think it might be possible, but in the end we all agreed we didn't really want to be the ones that opened up that can of worms [2-3]. Later, we (George Church and I) encoded digital information in DNA, and made billions of copies of the work [4]. One of the reasons we chose George's book, is that it was clear that he owned the copyright, he cleared it with his publishers, and so we didn't have to worry about being sued by someone.
Anyways, I think the larger point is that patents can be bad, especially for synthetic biology if they cover very enabling technologies (they can be good as well in many cases). The same argument could be made for copyright - however, having both forms of IP protection would probably make things much worse. I'm still of the opinion that it's probably better leaving things be the way they are now.
> I'm still of the opinion that it's probably better leaving things be the way they are now.
First off, I really appreciate your contribution to the topic!
In response though, I'm curious why you think things should be left as they are. Laws around patents and copyrights allow these sorts of things, and those laws are different or non-existent, depending on where you are in the world. It would seem to be anything but something that needs to be unchanged. While I think that status quo can help avoid further problems in many cases with law, patent and copyright reform needs to be happen and is inevitable, and this is purely a case of people working within existing laws, whether or not we agree with their course of action.
IMO & IANAL: IP of one form or another is probably necessary and beneficial for bio. If you make an invention, and it's trivially copied, then you need some protection to commercialize.
There are a few options:
1. patents -> they are expensive to get, and are medium term, but can be bad if they are trivial patents -> the biggest issue here is the non-obviousness part seems broken (as it often is in tech).
2. copyright -> easy to get, last forever , but don't cover ideas, which is very good. That said, copyrights could have much worse consequences if they are taken advantage of (b/c of ease of obtainment, and length), and of course they will.
3. MTA -> material transfer agreements probably are not viable for much longer... DNA and cells are too easy to write and edit now.
4. Secrecy -> This seems like a bad idea.
5. Something brand new -> I could think of individual policies that seem like good ideas, but I worry that I haven't thought about the ways that those policies could be captured and exploited... it seems like we are more likely to be at a better place with less risk if we can work on making the existing system better.
To begin, it's a patent dispute, not copyright. It's a dispute over the process of using mRNA targeting with Cas9 proteins, not the Cas9 genetic sequence itself (which is what copyright would be).
Secondly, their assumption "Copyrights [or Patents] are intended for artistic works." is only partly true - their main purpose is to promote the progress of science and useful arts, e.g. knowledge.
Whenever you have a technique that provides value like this, there is typically an amount of work that is invested into ensuring that the process works. Expensive work. Lots of highly trained people with expensive tools and expensive supplies that you don't see unless you're in the industry.
The groups are disputing over who came up with the process first because they've invested a ton of resources into it already, and they want to distribute that cost to people who are using this new, novel process today. A better article would have been about the new first-to-patent rule, which is what the court case is really about.
Yes, the CRISPR dispute is about patents. But, as the article says in the third paragraph, "But in the midst of this debate, perhaps a more fundamental one is taking place", which clearly indicates it's talking about a different debate than just the CRISPR patent.
The issue here is that CRISPR is like a printing press when scribes were state of the art. Yes, the patent on the printing press is important, but arguably more important is what's the status of all the stuff you can write with it. Previously, synthesizing long stretches or precisely editing DNA was hard enough that there weren't many people doing it (and synthesizing long pieces of DNA de novo is still pretty niche, though there are some pushing for a massive investment in making this tech a lot easier). But as it becomes easier and people get more experienced, it's worthwhile to have already thought through and discussed these issues.
>"But in the midst of this debate, perhaps a more fundamental one is taking place - whether some of CRISPR’s future products, like customized DNA, qualify as copyrightable entities."
That's the full sentence in context, and no, the CRISPER-cas9 patent dispute isn't over copyrighting the products of the technique. This author is trying to generate outrage over copyrighting DNA which this issue isn't about. The issue is over first-to-file.
Hold the phone; these groups are compensated by their own universities, the capital they raised by-and-large public agencies... Science is not supposed to progress only in the direction where there is a market or profit to be made.
Not true - they're also funded with royalties from previous patents. If UC wins this case (for example), then 35% of royalties will go to the inventors and the rest towards further research/university budgets.
True. Some of the first copyrighted items after copyright was established in 1790 were charts and maps which are very much like software, ie. follow a procedure to obtain a useful result, (ex. using a map, find out where you are).
The copyright-like protection for IC masks became obsolete once IC mask artwork was drawn by EDA systems, not humans. If you want to copy an IC, you work from the logic, not the masks, regenerate the layout and masks, and usually have a different physical layout. That's why no one worries about mask protection any more.
There's one other category that has its own special kind of intellectual property - vessel hulls. There's the Vessel Hull Protection Act, which prohibits making exact copies of registered vessel hull designs. (There's a cheap way to make copies of fiberglas hulls.) This only lasts for 10 years.
The fashion industry and the auto parts industry have lobbied for similar protection in the US, without success.
Copyrighting/patenting Life is for me the penultimate example of corporate greed. Luckily we won the first round here in the US [0] but we all know a loss is merely a setback in the constant march of unchecked Capitalism.
Locking life behind lock and key serves only to slow innovations and relief from suffering.
Corporate ownership of people (or The People), not just a copyright on their DNA. You might argue it's already happening in a Corporatocracy, and I'd have a hard time arguing with you.
When corporations not only own production and the right to prevent others from producing, but the governing of those rights (see every copyright extension in the last hundred years) there isn't a whole lot left.
CRISPR-9cas isn't a patent on life or a naturally occurring process, it's a new novel technique for cutting DNA.
The court case isn't disputing over whether or not life can be patented, but whether the first-to-patent rule can be upheld when someone already filed a patent. It's completely unrelated to that article you linked to.
Patenting dna is also a bad idea, especially for obvious transgenics, "express gene X from organism Y in organism Z". It's the equivalent of a software patent that says, "program a computer to sell X... over the internet".
> The strongest case that could be made for it, which a couple of people resurrected recently, is that Congress made a decision around 1978 that we should put software in the copyright system. But that was a complete disaster.
Why was it a complete disaster? I've heard lots of criticism of software patents or even closed source software but copyright isn't to blame for either as far as I can tell.
But DNA is also hardware (its structure encodes data, it can catalyze reactions)! So the rules that apply to hardware... no, let's not trivialize this by making weak analogies.
But still you can view a reaction of substance X with DNAzyme Y to be the hardware X taking apart the storage of software Y in order to read it. A destructive read operation, so to speak (or non-destructive in some cases).
This article is pretty far below Nautilus's usual standards. Though I don't think it quite makes the basic mistake of failing to distinguish between copyright and patent law, it does fail to (correctly) explain what copyright law is all about, and why a person might think that genes are a natural fit for copyright law. (And I do want to get this out of the way up front: Dan Burk is just wrong if he thinks that, to be copyrightable, something needs to have "artistic" or "aesthetic" value. It may be that this was the original purpose of copyright long ago, but it has thoroughly been left behind by actual copyright law, under which one clearly can copyright most collections of data, instruction manuals, etc.)
Copyright law is used to prevent literal copying. If I embody certain ideas in a particular way, copyright law just says that you can't come along and copy that expression. (Note the distinction here between copying the ideas, which is OK, and copying the expression, which is not.)
Patents, in some ways, are very different: to simplify things a bit, they allow an inventor to protect the idea of accomplishing a certain task in a certain way.
Software is interesting because it exists in both realms. Code embodies a set of ideas in a particular way, and also embodies patentable methods for achieving a certain task. The article frames this legal development as though it was somehow unnatural--something that Congress "decided" to do back in the seventies. But they decided to do it because both types of legal protections really do seem to apply. Yes, working through the details of how to apply copyright protection to software presented some interesting legal challenges (and I know there are many here who think this was a bad idea for other reasons), but the mere fact that this was a difficult process does not strike me as a good argument for declining to extend copyright protection to other novel forms of expression.
In principle, it seems to me that the burden really ought to be on those who want to argue against copyrighting genes. Like source code, and like the written word, they are a means for embodying ideas and information. In this way, the newfound ability to 'easily' edit genomes is akin to the invention of a new storage medium. If we were talking about a new type of optical storage, there would be no debate about whether data encoded in that medium and format was copyrightable. That should also make genes copyright-able by default. (Similarly, like software, genes can also serve as embodiment of patentable ideas.)
Of course, I don't mean to discount the fact that genes are special by virtue with their connection to the development of life, and their origins in nature. Both of these should give us a lot of interesting things to talk about in deciding whether, despite everything above, we really should have different intellectual property rules for genes. But this article doesn't discuss any of those interesting things. Instead, it sets up a legally inaccurate conception of the "purpose" of copyright law and uses that, along with the "difficulties" of applying copyright law to software, to make it seem as though copyright is just not conceptually the right fit for modified genomes.
EDIT: Previously, I had listed phone books as an example of a copyrightable collection of data. White-pages are actually not copyrightable, though yellow pages may be. The problem with white pages is that so little original thought goes into compiling, sorting, etc. that it is not thought to actually embody any original idea. But the white-pages exception is very narrow. If the database is curated in some way, or sorted, tagged, etc., then there is a good chance it is copyrightable. See http://www.copyright.gov/reports/db4.pdf
> It may be that this was the original purpose of copyright long ago, but it has thoroughly been left behind by actual copyright law, under which one clearly can copyright phone books
Are you sure about that one? I have a vague memory that phone books are a standard example of something that can't be copyrighted. Specifically, a listing of names and associated phone numbers. This is (in my memory) because (1) a listing of names and their phone numbers is a collection of facts, and facts can't be copyrighted; and (2) there is no creative expression at all in the listing -- it is in alphabetical order. Anyone else using alphabetical order will necessarily produce exactly the same text.
You're exactly right (though strictly speaking, white pages are probably not copyrightable, but yellow pages may be). A phone book is maybe the worst example I could have picked. You can copyright collections of facts--the white-pages exception is very small--but courts will look at whether you have put "creative" thought into how you selected entries, organized them, etc.
Saying human created DNA sequences shouldn't be copyrighted is like saying novels shouldn't be copyrighted because they use the words.
In the US at least, case law states that we can't patent natural genes, which is great. However, scientists are entitled to protections for their novel work.
CRISPR is no different from any other biologic in terms of patents, and in the long run, 20 years of protection is a drop in the bucket. Hell, by the time they figure out this CRISPR fiasco, it'll be close to out of patent.
Not to derail this conversation further but as far as software patents go I've been wondering for a while why somebody doesn't try this defense:
In order to have a valid patent you need to share a sufficiently detailed design in the patent application. What constitutes a sufficiently detailed design? One that a current professional in the field could produce with no further input from the patent holder and secondly one that if given to two different implementers results in the practically the same widget. If either of these aren't true, the patent holder has essentially not held up the spirit of the law in so far as patents are meant to contribute back to the public.
I'm on mobile now, so I'll stop there but I think you get the idea. For software the practical implication is that you need to submit your source code or pseudo code as part of a valid application,because much like a mechanical blueprint that is the design for software. Not:
Input--->block diagram--->output.
With that level of spec I could patent strong AI. I know some software patents are more sophisticated than that but I think most fail the test above.
I honestly don't know if this has been tried before and shot down or what. Can anyone comment on that?
Also as far as that goes, if the patent holder were to share their code in this fashion, and it passed all the other requirements then I wouldn't really have a problem with it.
Anyways, I think the larger point is that patents can be bad, especially for synthetic biology if they cover very enabling technologies (they can be good as well in many cases). The same argument could be made for copyright - however, having both forms of IP protection would probably make things much worse. I'm still of the opinion that it's probably better leaving things be the way they are now.
[1] http://msb.embopress.org/content/1/1/2005.0018
[2] http://journals.plos.org/plosbiology/article?id=10.1371/jour...
[3] http://scholarship.law.duke.edu/cgi/viewcontent.cgi?article=...
[4] http://arep.med.harvard.edu/pdf/Church_Science_12.pdf