Lookout, YARPP.
Are you a blogger that uses Yet Another Related Posts Plugin (YARPP)? There are a lot of similar plugins (hence the funny name).
But Google apparently thinks it’s novel, as the company has filed a patent application (pdf) for precisely this.
United States patent application 20120109932, titled “Related Links”, covers the process of identifying keywords within an article and then displaying related articles (or searches) based on those keywords.
In other words, it looks a lot link the “Further Reading” links at the bottom of this article.
Here’s how the inventors describe the problem they’re trying to solve:
Regardless of whether the web is static or dynamic there are cases where website owners want to display on their web pages links to other web pages within the same web domain or one or more other domains whose contents are related to the current web page, for example, to increase the page views on these web sites or pages. A conventional technique for presenting web page links is for a user (e.g., a web site owner or developer) to manually identify related web pages and then embed links to these web pages on one or more web pages of the user’s web site. The user can identify related web pages by conducting searches within the same web domain or one or more other domains based on the contents in the web pages, for example, by manually selecting some keywords from the contents on his web pages as most representative of the contents. These keywords are used as a search query to a search engine. The user can then manually select from the obtained search results. However, If the web site has a large number of web pages, and the textual contents in each web page is different from one another, it can be inconvenient for a web site owner to manually search for such related web pages and embed the links to them in the web pages during creation and setting up of the web site.
On the other hand, for some conventional web pages, especially for some dynamically created web pages, the contents on the web page may change when visited at different times. This can be due to changes to the text data during the time interval between different visits. In this case, it is difficult for the web site owner to predict what kind of content will be presented on the web page to the user in the future, hence it will be difficult for the user to find the web pages that can be considered related to the web page and embed them in the page.
Morgan says
Just a sidenote, this plugin has some serious security issues, I had two separate sites hacked due to an exploit in this plugin, even the most recent version.
People should proceed with caution when using this.
That being said, Google is definitely onto something here!
Francois says
Like probably many I thought about the “related links” concept before Google written this (abusive?) patent:
First as a proof I own the domain for many years:
RelatedLinks.com
Second After few months of reserach and development I succeed to create a javascsript that was able to classify a web page content in real time and thanks to that can show related content.
I admit that unfortunately I did not had a strong ad feed, plus it was again AdSense TOS to show other contextual sponsored links in web pages showing AdSense so my proposal was not very attractive and I did not pursue.
But there are a lot of variations of the concept and I can decide one day to try another one.
3 or 4 years ago I proposed to BuyDomains.com to create a specific javascript code they could give to affiliates and that will display domains for sale related to the page content the visitor is browsing. They did not reacted, but I just realized there is a growing huge player that could be smarter and use such technology to make more sale…
jp says
Who holds the patent on the concept of an “Internet search engine”.
We should be registering patents, apparently you don’t even need to think of a new idea to patent something.
Too bad there is no UDRP type thing for patents.