Daniel Porter
Nov 6, 2012
Featured

Patent Knowledge: Crowdsourcing

What is crowdsourcing?

Crowdsourcing is not a unique invention of the internet era, but modern connectedness has helped it come into its own. It is the process of distributing a problem to a non-specific public body of participants who can contribute to a solution, typically by a single entity which collects and combines all the contributing parts to the final solution. Crowdsourcing comes in several varieties depending on the problem that entity is trying to solve.

The Longitude Prize, Britain’s 18th century public contest to invent a more effective way of determining ship’s longitude, is an early example of the crowdsourcing concept. This allowed the government to recruit many of its subjects to help invent a solution to a problem that foiled naval scientists at the time. Today, the internet makes it easy to put problems in front of a large number of individuals, and companies are increasingly using crowdsourcing to solve difficult problems beyond simple idea creation. “The crowd” is now helping with problems of group evaluation, creative work creation, funding, and search through unmanageably large sets of data. Recently the intellectual property community has begun to recognize the value of crowdsourcing for prior art search.

Why crowdsource?

Not every problem lends itself to the widespread public distribution necessary for crowdsourcing, but this method of problem solving provides a number of benefits to problems that do. Crowdsourcing can benefit both the issuing organization and the public participants as well as significantly increase the quality of the results.

For an organization with a problem to solve, crowdsourcing is beneficial by reducing cost and increasing efficiency. Put simply, many minds, perspectives, and opinions are often better than one. To get the sort of diverse perspectives and experience accessible with a crowdsourced search, a company would need to make substantial investments into assembling a diverse team of researchers. Many of these researchers, despite best intentions, may not all contribute significantly to the final solution. Under most crowdsourcing models, only participants who contribute to this solution are rewarded for their efforts, significantly reducing the cost of the search per contribution to the final solution. In the case of the British government and the longitude problem, the government really employed the help of a large number of inventors, but needed only pay those who provided a solution. If the problem has no solution, or a search returns no results, the organizing body only has to foot the cost of publishing the search.

Perhaps even more advantageous for the issuing organization is that crowdsourcing lends itself to a much more thorough consideration of the problem at hand. Crowdsourcing a problem draws on a much wider variety of sources than more traditional approaches to problem solving. Often the collective knowledge and experience of a wide group participating in a search is much broader and more effective than that of a small team working intensely on solving a problem. In the case of searches (such as prior art searches), the many participants access a significantly larger portion of the available data than just a few researchers ever could. In other cases the varied experience and knowledge helps to search databases that would otherwise be inaccessible. Software code databases, for example, are accessible through the specific knowledge of seasoned experts that access these databases on a regular basis. These types of resources would otherwise be inaccessible without significant educational investment by non-experts.

Why participate?

The most common incentive for individuals to take part in a crowdsourcing is a contest-like reward system. From a wide variety of contributors to a problem’s solution, the organizing body will select the most helpful or “best” solution or solutions and give them a substantial reward of some kind. As was the case with the British government and the longitude problem, today many crowdsourcing models award a substantial prize to the one participant who “solves” the problem. For most participants, the cost of participating is outweighed by the potential reward for finding a solution[]. In many cases a crowdsourced search will reward users for knowledge and expertise they already have. This allows experts such as graduate students, researchers, and other subject experts to get paid for their expertise beyond their regular research activities.

Applications in innovation

As discussed in a previous article many of the major problems with current intellectual property systems stem from too much information. A number of companies are using crowdsourcing models to solve some of these problems. Here’s a quick rundown of some innovation crowdsourcing models:

  • Idea Connection: confidential distribution of R&D problem solving needs to managed teams of selected experts.
  • AskPatents: a StackExchange allowing anyone to participate in the patent examination process by contributing prior art related to patent applications.
  • Patentcrowdsource: research group focusing exclusively on patent infringement research, asking its users to identify patent infringement cases.
  • Peer-to-Patent: prior art search distribution for users to help with patent examination process. Rewards based on whether prior art is used by an examiner as a rejection basis.
  • Blue Patent: freedom to operate and intellectual property evaluation services using crowdsourcing model for searches.
  • Article One Partners: publicly distributed prior art searches distributed to a variety of patent researchers.
  • Patexia: prior art search distributed to select community of experts knowledgeable in specific field of searced patents.