Intellectual Property Primer: Copyright

Academia has always been a world focused on ideas and knowledge. Increasingly, the world of business (at least in the major developed economies) is becoming focused on ideas and knowledge over physical products and their manufacture. While both business people and professors concern themselves with the resources and products of the mind, they relate to them in different ways. The professional life of an academic typically revolves around credit for discoveries and scholarship, hence the enormous emphasis on who gets to publish what and the order in which names appear in author lists. Private industry is mainly concerned with ownership, and the ownership of ideas and knowledge is the domain of intellectual property (IP).

Most physicists will spend the bulk of their careers in “industry,” as I have discussed. Unfortunately, most of us get little to no education in intellectual property. I had none during eleven years of top tier physics education and, to judge by the gripes of intellectual property attorneys, many of my fellow scientists and engineers are in the same boat. Eventually, I picked up a working knowledge of IP through experience and self-study. This is the first in a series of notes on the different types of IP: utility patents, design patents, trade secrets, trademarks, and copyrights. This one serves as brief primer on copyright. To be clear, my aim is to provide helpful information, not legal advice, so go get a competent IP attorney before you bet the farm on your new startup company.


Copyright is the first type of intellectual property that students typically run into, either when they are warned against plagiarizing or when they publish their first journal article. Copyright allows you to own works of intellectual and creative expression (e.g., articles, books, movies, music, paintings, computer code, etc.). Expression is the key term, as copyright conveys ownership of a particular fixed and tangible form of expression, like a text, program, or work of art, but not the ideas it contains. This is illustrated in every genre of popular culture. The character and stories of James Bond are owned by the copyright holder, but they don’t own the idea of stories about super secret agents, so there are many others centered on similar characters like Jason Bourne and the Kingsman stories. Similarly, the Mathworks company owns the copyright to the MATLAB code, which they then sell licenses to use. However, Mathworks does not own the idea of a data analysis and visualization software package, and people have written Python packages with similar functionality. The Python packages, and Python itself, are “open source,” meaning the authors explicitly license everyone to use, change, and distribute their code without charge.

Who owns the copyright on a particular work? The author(s) of a work are the original owner(s) of the copyright. The point of copyright, however, is turning a work of expression into a piece of property that can be sold, rented, lent, licensed, etc. Authors often transfer their ownership of their copyrights automatically. That’s because many of us work as employees of a university, government laboratory, or company. As an employee, the works you create on the job are typical “works for hire,” which are automatically owned by your employer. The legal mechanism for this is often an “Invention and Copyright Assignment Agreement” that you are asked to sign as a condition of employment. These agreements are not unlimited. They rarely apply to works an author creates on their own time with their own resources (e.g., the novel you write on nights and weekends). Similarly, publishing an article in an academic journal usually requires signing a “Transfer of Copyright Agreement” that gives ownership of the article’s copyright to the journal with an assortment of rights retained by the author(s).

An author’s copyrights exist as soon as they write something or otherwise put their work into a fixed, tangible form. However, if you anticipate the need to defend your rights through litigation over theft or infringement, then it is wise to give notice of copyright using a statement like “© 2017 Your Name” and register the work. In the United States, you register with the US Copyright Office, and the registration fees start at $35 per work. The costs associated with IP ownership and licensing, which can become substantial, will be a theme in these articles. It is always necessary to gauge the costs of asserting and maintaining ownership over a copyrighted work or other piece of IP against it economic value. For example, it simply isn’t worth it for me to spend the time and money to register copyright for every article on this website, but I give notice of my copyrights at the bottom of the page. Finally, copyrights last a long time, typically the life of the author plus 70 years.

Why Does It Matter?

As I stated, copyright gets little attention in the academic world. No one is making big money off academic papers, and suing a bunch of penniless graduate students over pirating software is about as rewarding as trying to squeeze blood out of a turnip. Years spent in academia gave me, and most of my contemporaries, bad habits and low awareness of copyright.

The situation changes substantially in private industry. Big companies are typically very concerned with ensuring everyone is running appropriately licensed software and accessing other copyrighted materials in legal ways. The reason is simple. Once an organization (or an individual) has significant assets, they are a target for profitable litigation. Preventing law suits or, a least, limiting the potential damage they can cause becomes a matter of limiting activities that break the rules and expose the company to liability. If generating and selling copyrighted works is also part of the company’s business, then their focus on following the copyright rules and securing their rights is probably even more intense.

Tips for Scientists at Startup Companies

I ran into a thorny copyright problem I had never considered when I first left the academic / government laboratory world and began working at small startup company. I had no access to academic journals. It is easy to forget that big labs and universities frequently have site licenses to most major academic journals, so when you browse the journal website from a computer on the university’s network, you can automatically access all the content. Go to the same sites outside the university’s network and you run smack into a paywall, and it is not a small one. When I asked my supervisors how I should access non-open-access journals for background research on company projects, they had no answer, so it became my project to find one. Here are the options I found:

  • Virtual Private Network – People at technology startups often still have affiliations with universities that include access to their networks through VPN. That means they only have to log into the university VPN on their company computer to access an all-you-can-read academic journal buffet for free. Unfortunately, this approach almost certainly violates the university site license terms of service, especially if the individual in question retrieves papers for other people. So, this approach to accessing journal articles for company research is like stealing cable or downloading pirated copies of movies. You can do it, but it violates copyrights. You may never get caught, or you could get hit by a lawsuit that bankrupts your company.
  • Subscriptions – A company can get institutional site-licensed subscriptions to journals, just like a university, to give everyone on the company network access to the full content of the journal on-line. The only problem is that the subscription rates favor economies of scale and do not fit the size or budget of a small startup. When I priced subscriptions, I found that a site-licensed subscription to just Physical Review Letters would cost ~$4,000 per year. Bundled access to all the APS journals was over $20,000 per year.
  • Head to the Library – Why not just head to the library? Well, you can if there is one nearby with the resources you need that you can access. Consider, however, the cost of lost time. Take, for example, a PhD researcher who makes a salary equivalent to $50 per hour. To employ this researcher, the company also must pay for payroll taxes, benefits, insurance, office space, IT equipment, safety gear, etc. Together, these overhead costs can easily add up to 100% of salary, which we will take as a rule of thumb. Therefore, the PhD researcher’s time really costs the company about $100 per hour. Suddenly, an hour round trip to the library for “free” access to research materials doesn’t look free anymore.
  • Documents Delivery – Document delivery services are librarians for hire. These services will find you any article (or report, book chapter, patent, etc.) you want from essentially any source, send you a digital copy you can keep, and pay the correct copyright fees. Using one of these services allows a small company to have a single source provider of articles and research materials with one bill detailing all spending and proving that copyright was honored (which is handy in case of litigation). The document delivery service also simplifies things from the employee’s perspective by giving them one simple procedure to use for all their document needs. We use Reprints Desk for document delivery at work and are satisfied with the service. Now, document delivery is not cheap, but, depending on your needs, it can be cheaper than subscriptions or wasted time. Most of the big journals want around $30 per article for copyright fees, and document delivery providers add a service fee of around $10.
  • Document “Steaming” – Yes, someone has applied the streaming model to academic journals. Just like Netflix will let you watch (but not download) as many movies as you want for a fixed monthly fee, DeepDyve will let a you view (but not necessarily download or print) as many articles as you want for $40 per month. There are also yearly and group plans. The only catch is, just like Netflix, they have secured rights to a lot of material, but not everything. So, you may still have to buy articles they can’t “stream” for you.

Note that the landscape of copyright and access in the realm of academic journals is changing fast, just like it has for music, movies, and television. There are strong forces pushing for more open-access journals, lower fees, and new paradigms for academic publishing. I think we are still in the buy a $20 CD for one song days of academic journal access. A great deal more change seems inevitable.


1 Comment

Comments are closed.