Why is Linked Data Like an Aspen Grove?
October 17th, 2016
Two seemingly dissimilar systems that are strikingly similar in how they thrive.
Your library is like a tree. No, we’re not writing bad poetry over at SirsiDynix. Bear with me.
Your library is like a tree. Or at least your library’s internet presence is like a tree. You stand boldly alone in a jungle of digital information. If your library is part of a system or a consortium, you may unite and have a grove that buoys your presence. In general, however, your grove is lost in the broader forest. Biodiversity is great for ecosystems—just not for your library’s web presence. The stronger more dominant trees like Amazon and Wikipedia grow a canopy that smothers your library down on the forest floor of search engine results, where people don’t often see you.
Search engine preferences make it difficult for your library to be seen. Algorithms favor large, popular sites. It is incredibly difficult for a single library to climb up the ranks of search engine results when internet giants like Wikipedia dominate the top tiers. Even a large, urban library that serves a population of more than 500,000 will pale in comparison to Wikipedia’s 29 million registered users—and that’s just the English Wiki.
In addition to popularity, your library’s web presence is limited because MARC records aren’t readable by the Semantic Web. MARC was created before the popularization of the internet and before Google was so much as a twinkle in the eyes of Larry Page and Sergey Brin. (In fact, Page and Brin themselves were only eye-twinkles when MARC was developed.) Libraries eagerly cast aside their card catalogs and jumped on the MARC bandwagon, unable to anticipate the limitations of the new standard. Library collections can be accessed via the internet, but individuals have to know to go to your catalog to look at the items in your collection. These days, most searches for information start on a search engine, and Wikipedia with its millions of users almost always land at the top of the search results. No matter how relevant or authoritative your collection items might be, there’s no way for your records to land on a results page because Google, Bing, and Yahoo can’t read your records. So not only are you a small tree, you’re an invisible tree.
Don’t bust out the wood chipper just yet. Your tree has problems: small, outnumbered, invisible. But there are solutions. This is where the BLUEcloud Visibility, Linked Data, and the Library.Link Network come in. These solutions conquer each of your problems one-by-one. Let’s start by conquering the biggest problem: invisibility.
The first step in getting your resources on the open web is BLUEcloud Visibility. Your library’s content is invisible to search engines because MARC records are not harvestable or understood by the Web. BLUEcloud Visibility resolves this by transforming your MARC records into connected Linked Data resources. Linked Data is readable by the Web, so now your collection can be a visible part of the forest. Hooray! It’s good to be seen. But we’re not done yet.
If we stop at Linked Data your library still stands small and outnumbered in the forest. You may have a collection of a few hundred thousand, which makes for a good-sized tree. But this forest has some redwoods in it, like our earlier example Wikipedia. Wikipedia has over 5 million content pages in English alone and is available in over 200 languages. That’s not just a big tree; that’s a mammoth grove to compete against. You certainly can’t go it alone. What you need is a network. The next step after BLUEcloud Visibility converts your MARC records to Linked Data resources is to publish your records to the Web via the Library.Link Network.
Join the Pando
The Library.Link Network is the shared publishing platform by Zepheira. The platform connects libraries—and their collections—to other libraries and to the Semantic Web, creating a giant library network on the web. Dendrologically speaking, Library.Link connects libraries into a massive grove, comparable to Pando, an actual grove located in southern Utah. Pando is a giant colony of quaking aspen made up of thousands and thousands of individual trees. The aspens of Pando are connected by a vast underground root system, making Pando one of the largest living organisms on earth.
But don’t all libraries have similar problems with low-ranking visibility? How does a group a libraries with the same problem have more success than an individual library going it alone? Digging into the technical side of how search engines work, Zepheira’s Library.Link Network connects BLUEcloud Visibility libraries to each other and all other participating libraries. Rather than each library standing alone and seeking to grow visibility, Library.Link publishes library Linked Data resources as an interconnected network, amplifying common, unique, and locally relevant content for all. The connectivity serves to improve overall library visibility and Web users’ experiences. Activity with any participating library’s data or site benefits all libraries using the network. This is significant because massive networks of active content are exactly the kind of large-scale and connected information that search engine algorithms favor. As more libraries transform and publish their data to the Web, the search engine usage and result placement grows for all participating libraries.
The Library.Link Network uses data formats that search engine algorithms use to make library content visible. Search results are based on algorithms that favor authoritative data, which positions library content well for placement in search results. Authoritative data is only a part of the equation for achieving search engine relevance, sites must also be shown to satisfy user need by drawing users deeper into a site for longer viewing. Search engines favor sites with high traffic, significant linking from other sites, and frequent content updates; these features are considered indicators of authority, popularity, usefulness, and relevance. BLUEcloud Visibility and the Library.Link Network help libraries fulfill all these requirements. By connecting your library into the Library.Link Network, your tree joins the Pando. The Library.Link Network connects the data of over 1,000 library sites. Library.Link has transformed more than 27,000,000 MARC records into linkable, Web-visible resources. This gives your library a massive strength in numbers as a large, authoritative network on the Web.
Be Your Own Tree
The beauty of the network is that you still remain your own, distinct tree. Your library leverages the network to appeal to engine algorithms that make your library appear in search results. While the Libarary.Link network helps your visibility, it doesn’t absorb your identity. The network of these connections happens underground at the roots with local and distributed communities through the Web; on the surface you maintain a unique, local identity.
As part of the Pando remaining your own distinct library, your users and new users can discover your collection through search engines. Geo-location pairing ensures your library’s resources are connected to relevant searches within your area. Individuals who are part of the region you serve, but who rely on the internet for information, can now be pointed to your library. Our early statistics show that as BLUEcloud Visibility libraries begin to place in search engine results, new visitors to their catalog spike and then slow, as new visitors become return visitors. Libraries are beginning to be seen by people who might otherwise get lost in the woods.
Your tree may be small but doesn’t have to stand alone. Join the network of libraries coming together on the Semantic Web. The Pando is flourishing and through it, the relevance of BLUEcloud Visibility libraries grows.