Apr 112011

I sometimes take for granted what life would be like without Google. The ability to instantaneously get almost any information for free is simply amazing. A crucial component of practicing patent law in any area involves gathering the right information. What are the related patents? What language is used in similar patents? Does the invention stand apart from the published applications? Recently, I have been on a quest to find the best services that promote Google’s mantra: publicly available, fast, useful and comprehensive. I found that there’s no dominant, one-size-fits-all service; a lot depends on what you want to search for. Here, I compare 12 publicly available services that tailor to the English language. After my analysis of the strengths and weaknesses, I compare the results of actual test queries to come up with a “best of” list of three services. Ironically, my “best of” list does not include Google. Skip to the Best Of here.

Compared sites

Searching for patent information is tricky for a number of reasons. First, different search engines provide different search capabilities. Some allow you to search for terms that are in close proximity. Some allow for complex combined, boolean search logic (and, or, not). Some allow for wildcards so that you can generate a very expansive result list. In addition to all these search factors, there are differing options of searching through different fields. Some allow you to search through only certain fields at a time (title, abstract, specification, etc), while others allow you to combine the fields. This is why it’s difficult to come up with a list of “best of” list because they are all so different. Click on the links below to be redirected to the corresponding analysis below.


Boliven has an extensive database of 60 million patents and patent-applications. It includes these databases: US, EPO, PCT, JPO, KIPO, INPADOC.

I like the interface, as shown above. By default you can search through the entire fields, but if you want to drill down more specifically, you can use the Advanced Search. The big drawback to Boliven is that by trying to make money through subscription service, Boliven limits you to 30 results if you are not a paid subscriber. If you have a very specific query, this should be no problem though. If you would like a free 30-day trial, you can go to this site: www.boliven.com/freetrials. You will be prompted to enter some information. Once you submit that, you need to enter a promo code. The code you can enter for 30 days of Boliven bliss is TrentOstler3. Try it out.

Boliven has limited, but good analytical tools for registered users that do not pay the subscription fee. Also, the search syntax allows for proximity, boolean, wildcard, grouping, and even fuzzy searching. Here is a diagram of the advanced searching syntax:

Canadian Patent Database

This service provides over 2 million patent documents from Canada since 1924. Full document images including cover and abstract are available from 1975 to the present.

The syntax is simple and straightforward.



Esp@cenet also has an extensive database of 70 million patent documents. The service offers full text search for EPO and PCT patents and applications. In addition, it includes the English abstracts of INPADOC.  You can select and query individual countries in the country’s native language too.

I found the interface straightforward enough. The syntax did not allow for wildcard or complex searching, however.


FreePatentsOnline is great. The database includes US (applications and patents), EPO, Abstracts of Japan, and PCT. Notably, you have the option the ability to download original patent documents. In addition, FreePatentsOnline recently added non-patent literature so that you can now search the contents of several thousand journals concurrently with your patent search. Very nice.

The syntax for search queries is intuitive. They use the forward slash syntax that the PTO uses, which takes some getting used to. But the options for advanced searching, as provided below, give you the ability to do some intense searching.

Google Patents

The database is somewhat limited to only US patents and applications, a little over 7 million of them. Also, a search query does not show how many results were generated. You just get the 1-10 that appear on the initial page, then it’s a matter of clicking “0” after “o” to see how many more results there are. When dealing with a broad search term, this can be frustrating.

But what Google Patents lacks in coverage and results, it makes up for in speed and familiarity. Almost everyone knows how to use Google. It’s fast and gives you relevant results first. I use it to quickly look up filing dates. I also enjoy the “search within this document” feature.  Sometimes you need the text of an original claim in a patent or patent application. But PAIR only provides the PDF version. Google allows you to convert any words on the patent to text.

The advanced search options allow for more functional searching, but I couldn’t find proximity, wildcard, or advanced boolean searching capabilities. I understand why regular Google doesn’t use such searching styles, as there may be more advanced ways of providing relevant results, but I think this is a weakness for Google Patents to not include it.


KIPRIS includes 4.5 million patent documents. In addition to patents, KIPRIS houses IPR applications, legal status information and trial information. Plus since 1996, KIPRIS has been run on behalf of KIPO.

The interface is simple and intuitive. The advanced searching also provides a lot of options. Here is the syntax:

Patent Lens

Patent Lens is another great service. Boasting 10,870,431 patent documents, Patent Lens covers US, EPO, PCT and Australia jurisdictions. I like the ability to analyze particular results to see what the patent family is. For instance, if I’m looking at a US application, I can see where else in the world it was filed. Also of interest to me is the Patent Lens Sequence Project, which allows you to search through 80 million DNA and protein sequences disclosed in patents.

The advanced searching is also robust, with boolean logic and proximity searching capabilities.

Patent Surf

An interesting service, Patent Surf primarily focuses its search engine on natural language, rather than use boolean syntax. You can enter a word, phrase, paragraph, or even page and the search engine will generate the most relevant results using relationships that are stored in their database.

The syntax is simple, use natural language! If you want to emphasize a particular word, you can do so with an asterisk–even multiple times.


Great service! This neat and organized interface reminds me of Google. Not only is the interface sleek, it provides additional features that allow for analytics. You can organize your results by assignee, inventor, filing years, patent family. You can also easily export your results, downloading multiple patents. You can also subscribe to receive RSS feeds for particular queries. The database is extensive, including US, EPO, and PCT patents. It is even more extensive, adding Norway, China, Korea, and Japan, if you wish to subscribe to the monthly subscription.

The searching syntax is user-friendly when using the field search, as you can input queries for particular patent sections. The “command line” is a little different, but looks powerful.


Sumo Brain

SumoBrain is a useful tool for searching major patent collections. The database includes US (applications and patents), EPO documents, PCT, and JP abstracts. It compares well with other free patent search services in terms of managing queries and saving records.


The search syntax is straightforward, using the slash syntax for particular section searching. The weighting operators are worth checking out as a different way to hone searches. The search syntax they provide is the same as FreePatentsOnline’s document.



The PTO provides two services, PatFT which searches the issued patents and AppFT, which searches the published applications. PatFt provides full text patents since 1976. AppFT provides the full-text applications published since March 2001.

The USPTO’s website is still kind of 1990s in its design. It’s not fancy, doesn’t have analytical options, just provides the results in bare form.

Both services operate using a basic search engine. You have to get the hang of using the syntax.

WIPO PatentScope

Great service. PatentScope is made of two databases, each with its own search interface–the International Applications search, and the National Collections search. The National Collections includes additional breadth in that it includes the international document (PCT) plus individual countries. The full-text searching capabilities include PCT, African Regional Intellectual Property Organization (ARIPO), Brazil (BR), Cuba (CU), Europe (EP), Israel (IL), Mexico (MX), Morocco (MC), Singapore (SG), South Africa (ZA), Spain (ES) and Vietnam (VN). Less than full text for many other countries Argentina (AR), Chile, Columbia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panama, Peru, Republic of Korea, Singapore, Uruguay, Vietnam. This adds up to a total of 7,752,536 patent documents, as noted on their website.

The syntax is straightforward and allows for powerful querying. Below are some syntax examples.

My “Best Of” list

I performed three queries and the results are shown below. To the right of the public patent search services are some proprietary comparisons.

Three services really caught my eye in terms of the interface and the results, effectively making my best of list. They are:

  • FreePatentsOnline
  • PatentLens
  • PatSnap


As noted in the above graphs, free search services are very good, but I believe can get even better. I’m going to make a prediction–this area of patent law will experience significant growth and change in information mining. Gone will be the days of paid-for database services. The Google model will become the norm. In sum, the more advanced our society becomes, the more information we will get. Our task is to optimize the tools for gathering such data because there is a demand out there that frankly demands it.

 Posted by at 1:08 pm
Mar 022011

I recently came across a pretty interesting blog http://tacticalip.com/. As I was reading about Patent Reform, I came across something hilarious: a terms and conditions section for viewing the blog. I thought it was a joke, but then I realized that these people were serious.

Hmmm. A website without any architectural restrictions on accessing content trying to restrict its viewers under such a legal obligation? It sounded fishy to me too. I figured I should probably look into the terms and conditions that I was “bound to,” considering I was using the site, so I clicked on the link. Here is the first paragraph.

So let me get this straight, TacticalIP. As a user, I am granted a non-exclusive, revocable license to access the Website under its terms and conditions of use. If I do something that goes against this agreement, my access becomes terminated. My first thought was, why threaten to terminate access to the website? There is no practical way to implement such a threat. Did they think that they could terminate your use by blocking your IP address? No, because here is a clause that they have later on in the terms.

Not only can IP address information not be used to identify someone, this blog does not register viewers. If I go against the terms, how are they going to know who I am? Unless these people have some advanced way of tracking every possible protocol that I use to receive the web pages they have freely offered on the web, then I don’t see a point in claiming such bold termination. I think it’s interesting to think about how they could architecturally restrict one’s use, but it’s also interesting to think about whether a restriction can be legally enforced. I will defer that topic of conversation for another day.

Back to the terms and conditions, the next paragraph really had me going.

So my use is restricted to only reading the text of this website. Consider this paragraph copyrights on steroids. Basically, you can only use their content for reading. But such a regime should not be encouraged, especially on openly-accessible websites. If I were to provide consideration to a website (user registration or a monthly subscription fee) to access this information, I could understand the legal basis for restricting use. But we’re talking about a blog that is as open as can be. Isn’t the open Internet a medium that is designed for modifying, publishing, transmitting, transferring, and otherwise reproducing content within the framework of the copyright law? I think that it is. I want to know under what law is TacticalIP seeking to create such a seemingly legal obligation. Copyright law? Contract law? These are questions that need to be answered, but I’ll tell you one thing in regards to contract law; there’s no way that I would agree to these terms.

In sum, websites should only use such clauses as TacticalIP’s terms, when they are playing an April Fool’s joke or when they have registered users. I may have violated Tactical’s terms with this post by copying from their website, but I want to see how they are going to terminate my use.


 Posted by at 10:20 am
Jan 262011

A few years ago, I took a class at the Harvard Extension school entitled Internet and Society: Technologies and Politics of Control. We covered a lot of emerging Internet concepts including the power of the masses and net neutrality. A lot of the concepts we covered were very much in their infancy  including Wikileaks, Amazon’s mechanical Turk, Wikipedia and crowdsourcing. I recently came full-circle with crowdsurfing on my blog. According to Wikipedia, crowdsourcing is the act of outsourcing tasks, traditionally performed by an employee or contractor, to an undefined, large group of people or community (a crowd), through an open call. While doing price checks for patent translation services, I came across a free translation WordPress plugin by GTS Translation. The plugin provides a widget (utilizing crowdsourcing) that allows you (or your readers) to convert your blog (or whatever other text you have) into Spanish, French, or German. It also caches the results so that search engines can mine your content in other languages.

I have used computer-generated translation services before. Google’s translate service, for instance, used to be awful, but has made a lot of improvements lately. I was expecting this plugin to be awful as well because most of my posts are very specific and it would take someone that knows what I’m talking about in English to make a good translation into another language. But I don’t know French, Spanish, or German to really evaluate each translation. That is why I was pleasantly surprised when my good friend Spencer looked at it for me and informed me that a reader could get the gist of what I was talking about. To me, that is the most important. I am sure the technology will only get better as more people participate in crowdsourcing. My next hope is that they offer additional languages, like Russian.  After all, the world is obviously becoming increasingly interconnected. You may as well get your message to this increasingly interconnected audience.

The plugin can be found here.

 Posted by at 6:33 am
Apr 232010

At the SIPLA symposium today, Ilan Barzilay commented on his new website BostonIPBlog.com. The website focuses on tailoring their cases to judges, so that they aren’t wasting either their time or the judge’s time. What an interesting idea! Especially considering there are undoubtedly SO many tricks of the trade that a new attorney would have no way of figuring out. For someone who plans on practicing law in Boston, this website is not a bad idea. At the moment, it’s got room to grow, limited to one person generating content. But ideas like this are sure to grow in number throughout the country.

 Posted by at 12:52 pm