Patents are documents with the force of US statute that reserve intellectual property to a specific entity for a given period of time, generally 20 years. Patents play a significant role in the area of business, management, engineering, design and fine arts, and marketing. Patents are awarded through a formal process known as patent examination and award.
For library personnel, it is important to note that we must never assert that we have created a comprehensive patent search. The only source of an authoritative, comprehensive patent search is a patent attorney (PhD + Juris Doctorate AKA JD). The most authoritative search is conducted via a patent examiner, patent attorneys employed by the US Patent and Trademark Office, whose job is to assess patent applications and indicate if those applications will be awarded.
We can explain the steps of patent discovery to users. We can point out key features that will allow end users to create a more successful patent search. We do not offer patent searching as expert searchers.
There are three strategies for patent instruction. Free text searching is the mode of searching that users are most familiar with but it can also be very problematic, given the language that is used to describe the technical innovations by the applicants and patent examiners. In order for individuals to be successful in free text searching, base-line instruction on the difference between technical jargon and vernacular English is key.
For users to successfully implement classification search, they have to understand hierarchical structures of data bases. In many cases, novice users have difficulty understanding this concept, particularly in the case of Graphical User Interface/ tablet focused digital natives. In most cell phones and tablets, as well as Mac computers, the hierarchical structure of databases are hidden from the end user, so many not only have not used traditional paper systems such as indexes, but also have not become experienced with a folder>file>document hierarchical structure on their computing devices either. This is a baseline skill that must be understood for eventual searching success.
For users to successfully implement patent citation searching, users have to understand the semantic (relational) context between multiple items. Those semantics may be related across many fields of metadata. In order to make meaning and predict which citations are likely to be useful, individual searchers need to understand the co-location of citations within the classification hierarchy, the relationship between assignee, inventor and applicant, and the types of information that users are likely to find in a scholarly journal article, technical paper, and patent application/awarded patent. Baseline instruction that is needed includes defining metadata and instruction on how metadata works, how classification co-locates patents by innovation, not by the author, assignee or date (although you can return search results in this manner if it is appropriate for your search), and in types of technical literature.
There are many roles for liaison librarians to facilitate discovery of patents, particularly with providing technical information literacy that allows searchers to make meanings of their findings.
Strategy 1: Full text searching
Full text searching for patents involves developing a list of keywords that describe the innovation, and then using those keywords to create search strings in patent discovery systems such as Lens.org, USPTO Patent Public Search, or Google Patents. Commercial databases also include patents. Those databases are linked in the "Discovering Patents" page of this LibGuide. Patents use complex technical language to describe design innovations. This can be a barrier to effective patent searching for novice searchers. Performing a basic Google search is a way of mining these complex technical terms, which then can be effectively searched within the patent discovery systems. These full text searches should be conducted against multiple databases to ensure a comprehensive search.
All patents are assigned multiple classification numbers as appropriate. The classification numbers are part of the Cooperative Patent Classification, an extensive hierarchical structure specifically designed for ensuring that the patent system is searchable. Classification numbers are assigned to patents during the application process and can be found in Boxes 51 on a granted patent. The classification system is grouped by area of technology. You can use the classification system to narrow or focus your search on the specific design innovation, without knowing the terminology used by prior inventors.
The CPC is divided into 8 main areas, A-H:
– A: Human Necessities
– B: Performing Operations; Transporting
– C: Chemistry and Metallurgy
– D: Textiles and Paper
– E: Fixed Construction
– F: Mechanical Engineering; Lighting; Heating; Weapons; Blasting;
Engines or Pumps
– G: Physics
– H: Electricity(USPTO, nd)
The classification system is further divided into sections, classes, subclasses and groups within each of the areas. Explore the patent classification system through the online interface hosted at USPTO. This interface allows you to keyword search across the classification descriptions in order to determine likely classifications relevant to your design.
Patent citation searching involves finding a patent that is related to your innovation, and looking at Box 56 on the patent cover sheet to identify relevant other patents that may be related to the innovation being described. Those patents can be mined for relevant other sources. Patent sources are not limited to patent citations. Box 56, References Cited, may also point to external sources, including academic journal articles, company websites, manual and catalogs, and other relevant materials. Generally at least one patent will be cited.
For this week, please use the Lens.org to search for a patent using the three types of search strategies. Note the challenges for each type of search. There is no set way to approach the strategies in terms of order. Maybe beginning with free text is appropriate for your search. If not, perhaps starting with a known patent's citations is a logical first step. If you have a classification code, you could start there.
Do: Write down what you hope to see as the result for your patent search (otherwise known as the topic of the search).
Do: Use multiple strategies to research that topic.
Do: Reflect on the challenges of each type of strategy and note potential gaps in knowledge that an end user may need to fill in order to successfully complete a patent search.