Kurt Sutter – Defensive Publications Demystified

Kurt Sutter – Defensive Publications Demystified

Guest post by Kurt Sutter.

You had a great idea, but you don’t want to patent it, and you don’t want others to patent it, either.

Then publish it. This makes it “prior art” for any later patent applications. The publication can be used to invalidate later patents.

This type of publication is called a defensive publication or a defensive disclosure.

In the following, I elaborate on the concept: What are defensive publications? How and where to publish them? And when not to use them?

What Are Defensive Publications?

Defensive publications are public descriptions of ideas with the purpose of making these ideas… public. Once an idea is public, it cannot be patented anymore (at least in most countries; see below). Hence, if a third party tries to patent the idea after your publication, the patent will not be granted, or (if it does get granted) it will be invalid.

What Are the Requirements?


If you create a defensive publication, you can use it to attack a later patent. For such an attack to be successful, you must be able to prove, in litigation, where and when the publication took place and what was published. Also, you have to prove that the publication was, to some degree, accessible to the public. And you have to show that it describes the invention claimed by the later patent.

As usual, alas, different jurisdictions may have different requirements. In the following, I will concentrate on the European practice (i.e., the practice of the Boards of Appeal of the @European Patent Office), but see the notes under “Other Jurisdictions” below.

European Practice

Here are the, in my view, most important rules for a valid and useful defensive publication under European practice:

  • Describe all the relevant details of the idea, i.e., all the ideas that a later patent applicant might want to claim. Just outlining a general, abstract concept might leave room for later patents to claim important details.
  • Describe the idea in such detail that it can be implemented. For some ideas, once you outline them, a person of ordinary skill in your field will know how to build or practice them. But for some ideas, you need more information before you can implement them. Provide such information.
  • Make the publication searchable by public search engines under the relevant keywords. Case law says that it must be possible to find the document “with the help of a public web search engine by using one or more keywords all related to the essence of the content of that document.” (See, e.g., decision T 1553/06 of the Boards of Appeal of the European Patent Office; reference below.) Merely making the content accessible under a URL not known by search engines does not suffice. Nor should you publish non-searchable content, such as bitmap-based text or images without searchable, pertinent text. And search engines must not be blocked from accessing the publication.
  • Keep the content accessible for long enough for members of the public (i.e., people under no obligation of secrecy) to have direct and unambiguous access to the document.
  • The publication should clearly indicate when and where it was published. This alone does not prove anything (see next bullet point), but in practice, it makes the publication much more credible. For electronic publications, cryptographic timestamping by a trusted third party also adds to credibility.
  • Make sure you can prove all of the above. Depending on your publication method, this may be the trickiest requirement. You may have to provide such proof ten or twenty years from now. In a world that changes daily, with publishers, websites, and search engines coming, evolving, and going away all the time, providing such proof may be as hard as carrying water in your bare hands across a desert.

Other Jurisdictions

While I wrote this, I talked to colleagues in various jurisdictions (see acknowledgments below). I was happy to see that other countries apply similar criteria, even though a few details differ. If your defensive publication fulfills the above criteria, you may have a good weapon against later-filed patents in most countries. Just make sure you err on the safe side.

How to Publish?

If you’ve read the above, you will understand that just putting something onto your company’s website or a blog might not be the best way to do defensive publishing. To be on the safe side, you should make sure that the content is also indexed by the major search engines. More importantly, you may have to prove, in decades to come, what was published (and searchable) when and show that it has been up for a reasonable period.

There are various publication channels that might make your life easier. Some specialize in defensive publishing, and others just happen to publish documents you might use as defensive publications.

Let’s have a look at some of them.

Specialized publishers

Some publishers specialize in defensive publications. In the following, you will find a table of the most popular ones, in alphabetical order. Given the above criteria for robust defensive publication, I checked if documents published there can be found through a public search engine. I have also checked if they provide robust timestamping, such as cryptographic timestamping by a trustworthy third party, to prove the publication date.

List of some Defensive Publication providers

The links to these services can be found at the end of this article. Note that IP.COM and Research Disclosure both offer rebates if you publish more often.

I have sent all three of them a request for further details. So far, the only one who got back to me was IP.COM. If/when I get replies from the other ones, I will update the details in the above list.


Scientific or technical journals are a fairly robust and acknowledged channel for defensive publications. They are typically solidly indexed in public databases and libraries, and the patent authorities usually acknowledge their effect as prior art.

However, you need to find a journal that fits the field of your idea and is willing to publish something as detailed as a defensive publication. Further, your publication has to meet the journal’s quality criteria and may have to undergo a peer-review process.

Some journals publish quickly, while the review process can take weeks or months for others.

arXiv.org has, over the last few years, gained a lot of popularity. If you are endorsed to publish there, if your writing meets the required standards, and if it fits one of their subject fields, arXiv provides a quick way to publish something.

Be aware, though, that Journals can disappear as well (see the excellent article on @Ars Technica by @John Timmer mentioned in the section “Further Reading and Links” below).

Patent offices

When you file a patent application, it is usually published 18 months after its priority date (i.e., 18 months after its first filing date). This publication also forms a defensive publication—a very solid one: it enters the major patent search databases, has a clear publication date, and its contents are readily searchable. It is unlikely that a patent office or court will ever question its effect as prior art.

In addition, many patent offices provide a procedure for publishing patent applications early (i.e., before the expiry of the 18 months), typically within a few weeks from filing. This makes patent applications an interesting option for publishing something (comparatively) quickly.

Pricing varies from office to office. If you use a patent office where the search is not mandatory or not payable when filing, filing fees may be a few 100 CHF/EUR or even less. However, you will have to use one of the languages the patent office publishes in (e.g., German, French, or Italian and the Swiss patent office). Also, depending on the patent office, your patent application may have to fulfill some formal criteria before it can be published, which makes it advisable to have at least some patent attorney support.

Using videos

Using videos as defensive publications is tricky at best; see, e.g., decision T 3000/19 of the Boards of Appeal of the European Patent Office. Simple screenshots may not prove what has been disclosed. Also, it may be hard to prove that relevant content found only in spoken text and images was searchable at the time of publication.

Some Stats

The previous section shows that there are various channels for publication.

You may want to choose one that is regularly used by patent offices when examining patents. This reduces the risk that patents get granted that should not be granted in view of your defensive publication, and it adds credibility to your publication channel.

I have, therefore, run some queries. Which of the above channels has been cited in Chinese, European, Japanese, U.S., and international patent applications between 01.01.2023 and 31.12.2023? Here are the results:

(Note the log scale!) IP.COM was cited in 0.012% of the patent documents, Research Disclosure in 0.0028%, TDCommons in 0.00086%, and arXiv in 0.16%.

For completeness, I have included Youtube in the graph (it scored 0.08%), but for the reasons above, I do not recommend this for defensive publications.

So, Where to Publish?

Given the above choice of private providers, journals, and patent offices where you might place your defensive publication, which one should you choose?

If you need a quick publication (aka a publication on the same day or within a few days), I’d go for one of the dedicated publishers. Among these, TDCommons is the cheapest and very well-searchable, but it is younger and not as well-known among the authorities as ip.com and Research Disclosure. For all these outlets, though, you don’t know how long they will be around, and once they are gone, proving a publication date and searchability may become increasingly cumbersome as years go by.

arXiv and other well-established journals are likely more solid, but you cannot publish everything there, and they may be slower.

The patent offices are rock-solid publication channels. However, they are slower, and the administrative and formal hurdles are slightly higher. When you do defensive publishing only rarely, you’d probably best go through a patent attorney if you want to use a patent office for publishing. If you do defensive publishing regularly, have your patent attorney teach you how to get something published through a patent office. (If you want your patent applications to proceed to grant or even to have litigation quality, I strongly recommend having them written by a patent attorney.)

So, there’s no “best” solution. To make matters even more complex, you can combine two or more publication channels, e.g., by throwing something at TDCommons and one of the others.

Seek your patent attorney’s advice 😉

The Disadvantages

There are some obvious drawbacks of publishing your idea in a defensive publication:

  • Once published, you cannot patent the idea anymore, i.e., even your own patent application, if filed after the publication date of your defensive publication, will be invalid. (There are exceptions to this rule in a very small number of countries. In the U.S., for example, the inventor enjoys a 1-year grace period for filing a patent application even after publishing his invention.)
  • Others can use the idea. If you publish your idea without first filing a patent application, you substantially place it in the public domain.
  • If you publish under your name, others may learn what you are up to.

The Alternatives

Patent Applications

Patent applications are an interesting alternative to defensive publications. Since a defensive publication must anyway describe the idea in detail, drafting a patent application may not require much more effort. And if you choose an office with low fees (such as the Swiss Federal Institute of Intellectual Property), the advantages of getting at least some local patent protection may outweigh the additional costs.

Moreover, as mentioned above, you can request early publication, which makes it harder for third parties to file further patent applications in this field. (Note, however, that there are, in some situations, drawbacks to accelerated publications. Make sure to discuss them with your patent attorney.)

Keeping It Secret

Keeping the idea secret may, in some cases, be another alternative.

In that case, though, it can still be helpful to document that you were in possession of a particular idea at a specific time. Hence, it may be a good idea to obtain an electronically timestamped and signed copy of your documentation; see, e.g., the post by @Sebatien Ragot in the section “Further Reading and Links” below.

If you want to keep things secret, though, you should set up trade secret policies and take steps to prevent (or at least make it harder for) former employees to exploit their knowledge. But that’s for another article.

The Odds and Ends

Here are some miscellaneous points to keep in mind.


If you are in Germany, the Arbeitnehmererfindergesetz (the Employee Inventions Act) complicates matters. A defensive publication requires the inventor’s consent (which may entail paying them compensation).

Public use

Public use of an invention, e.g., by selling a device that contains the invention, may also generate prior art. For example, suppose you have invented a new cap for a bottle and start selling these bottles to the public. In that case, selling will make the invention public in most countries (with exceptions: for example, prior use outside Germany is not considered to form prior art against German utility models).

The greatest challenge when relying on public use as prior art is, often, providing proof (of what was sold when and where to whom) when trying to invalidate a patent. Further challenges arise if the invention is difficult or even impossible to recognize from the sold product alone (this topic is a question presently pending before the Enlarged Board of Appeal of the European Patent Office, case G 1/23).

Written publications provide a more solid basis against later third-party patents.

Publication License

Do you need to place your defensive publication under a specific license, such as one of the Creative Commons licenses?

Nope. There is no need to put the defensive publication under a particular license.

However, most defensive publication outlets will request quite far-reaching permissions to publish, copy, and otherwise use the documents.


Defensive publications can be an effective tool for preventing third parties from obtaining patents in a given field.

However, when using them:

  • Know their pros and cons;
  • Draft them carefully to include all relevant details (or have your patent attorney draft them); and
  • Publish them on a platform where they are well searchable and that provides solid proof of publication (or that is generally accepted by the authorities).


While writing this article, I had the pleasure of discussing the topic with colleagues in other countries to better understand their local requirements for making a defensive publication effective.

My thanks go, in alphabetical order of the countries, to

  • Ji Liu at CCPIT Patent and Trademark Law Office, China.
  • Atsushi Aoki and Masaru Nakajima at Seiwa Patent & Law, Japan.
  • Robert Mueller at Greenblum & Bernstein, P.L.C., USA

Further Reading and Links

Links to the defensive publication services:

https://www.epo.org/en/boards-of-appeal/decisions/t061553eu1 T 1553/06, a decision by the Boards of Appeal of the European patent office on the requirements for internet publications to be prior art.

WIPO PROOF under the loupe by Sébastien Ragot: About time-stamping documents.

https://www.wired.com/story/google-tdcommons-dpub-patents-prior-art/ A recent publication by Paresh Dave on WIRED focusing on TDCommons.

https://arstechnica.com/science/2024/03/study-finds-that-we-could-lose-science-if-publishers-go-bankrupt/Article on Ars Technica written by John Timmer: “Study finds that we could lose science if publishers go bankrupt.” Some interesting thoughts about the persistency of internet publications. It in turn refers to a more detailed study at https://www.iastatedigitalpress.com/jlsc/article/id/16288/ by Martin Paul Even of the University of London

The above article was first published here

    Did you like what you just read? Then subscribe to my free Tip of the Week!