Report:Google Patent Search/Overview
|Report||Patent Coverage Map||Ratings||Comments|
|This search system report was created by the Intellogist Team and is available for viewing only. If you'd like to share your knowledge on Intellogist, please visit the Best Practices, Glossary, or Community Reports pages. If you are a registered user and would like to be notified of any substantial changes to this report, you may place a "watch" on the Revisions page, which is the last page listed on the table of contents. To learn more about using the Intellogist "watchlist," see the Watchlist Help page.|
The following section contains subjective comments about the system that represent our editor's opinions, and should not be viewed as fact. Editor's opinions include positive and negative judgments about the product written in consideration of wider context, including related products and the industry at large. Further subjective information is presented in clearly labeled "Editor's Notes" throughout the report.
Google's 2006 foray into the world of patent search engines was greeted with fanfare. Browsing patent information on the web is often a slow affair, and Google was able to bring fast, efficient patent image searching to the web. Among the unique new features offered by the system was a way to superimpose keyword highlighting over the patent image itself, allowing users to skim the text of patents in their original form, find relevant passages quickly, and cite column and line numbers with ease. The availability of full patent images also made viewing the complete set of drawings from a patent quick and easy: Google Patent Search loads each drawing page as a thumbnail image on its summary page for each record, providing a new way to get a quick overview of a patent’s drawing pages. Additionally, users can now (as of mid-2008) view the hit list as patent drawings in a thumbnail view, to allow users to scan results by scanning patent drawing images. The tool was also updated in 2008 to include published applications (pre-grant publications); users now have the ability to view either only granted patents, only published applications, or both. Finally, the system itself is free, and even offers free patent PDF downloads and free RSS alerts.
After the thrill of the initial launch, some downsides to the system became obvious. The system only offers a search of US granted patents and published applications, excluding any non-US patent art. The pre-1976 US full text collection contains a significant number of scanning errors. Both US and international classification data in the system is outdated. Total hit counts are not displayed, and browsing is limited to around the first 600 records that match the query. Example searches for "car" and "computer" on April 5, 2010 yielded merely 552 and 576 hits respectively. Searches cannot be limited to the abstract or claims. And advanced query construction using truncation or proximity operators is not permitted.
Google Patent Search's retrieval accuracy has also been called into question. An informal 2008 test of the system's capabilities, performed by patent librarian and blogger Michael White, showed that Google Patent Search performed especially poorly when compared to three other online patent collections. Searches by assignee name, inventor name, title keyword and US patent classification all returned significantly fewer results than other search engines. These results show that Google Patent Search should never be used to perform a comprehensive review of the patent art. However, this investigation did not test Google Patent's proprietary retrieval algorithm to determine whether keyword searches were likely to bring the most relevant search hits to the forefront, as some users claim. So it is possible that the appropriate place for Google Patent Search in a prior art searcher's toolbox is as a quick-and-dirty search tool for keyword concepts.
Additionally, despite its innovative features, being a Google offering may actually be considered a negative for this system in the patent world. Because Google is known for tracking user patterns, the system cannot convincingly offer search string anonymity. In addition, development momentum from the Google team may have stalled, for the moment. Google Patent Search was developed as one of the company’s famous “20% time” projects, where employees are allowed to pursue projects that interest them. Almost a year after its release it is still in beta testing mode, and it is unclear when (or if) this will change. Finally, although technical support was initially described as responsive, it is unclear how many Google employees are currently working on providing user support and further developing the system.
Despite all these downsides, there are still very good reasons to use Google Patent Search. The system is unique for its speed and for its in-image highlighting, and it's especially valuable because it is one of the two existing US full text collections that extend back to the beginning of the US patent system (the other is currently hosted by several Thomson Reuters products). Because these two collections were produced independently by OCR technology, there is a chance this source may contain data that the other US full text collection does not, and vice-versa. And finally, Google's proprietary search engine and relevance-ranking capabilities ensure that the tool can stay useful as a quick-and-dirty search provider that seems to pull accurate US results to the forefront.
- ↑ White, Michael. "Comparison of Free Patent Databases." Blog Post, The Patent Librarian's NoteBook. http://patentlibrarian.blogspot.com/2008/09/comparison-of-free-patent-databases.html. Accessed November 4, 2008.