Over the years working at ESRI on GIS software and later consulting with large companies on their GIS strategy, Goldin realized there was a need to provide immediate, actionable insight on an organizations’ data. Doing so would provide enormous value and would enable organizations to get the most out of their hefty investments in geospatial data. His idea was to teach modern search technology to understand location and to read the multitude of complex GIS formats. In 2008, he and his partner, Ryan McKinley, one of the forerunners of open source search, went on to found Voyager Search, a company targeting this unexplored corner where search and GIS meet. The objective was to develop an innovative solution that would enhance the manner in which organizations could find and manage their geospatial data.
Goldin took a cue from IDC’s study “The High Cost of Not Finding Information.” The research firm estimated that 35 percent to 50 percent of information is locked away in some offline repository that is unavailable to the rest of the organization. Additionally, IDC wrote that there was duplication of work and the bad decisions being made based on incomplete information. Goldin knew this was true in GIS. There was a clear need to make information discovery easier, to eliminate the GIS industry’s requirement for creating complex metadata in order to find something and ultimately to support information sharing between employees. Both of the problems described in the IDC article are compounded in geospatial organizations because those knowledge workers need specialized data, with a geographic reference to do their jobs, and if their corporate IT departments had developed enterprise search engine at all, it didn’t understand spatial data.
Voyage through Spatial Data
Voyager addresses these problems by providing an easy-to-use, out-of-the-box search solution that can support a wide variety of users around the globe. It is open and scalable, but with advanced security. Voyager can index a wide variety of content. Instead of requiring centralized data storage, Voyager offers a centralized search experience while leaving the data untouched in its original location. It contains the connectors to bring content into the best-of-breed, open-source search engine—Apache Lucene/Solr—giving organizations an easy-to-use, web-based interface for discovering their content. It even offers the tools for delivering documents and web services to the users.
Ryan McKinley, CTO
Instead of requiring a centralized storage of all the data, Voyager offers a centralized search experience while leaving the data untouched in its original location
Voyager can be incorporated into existing systems without adverse impact on current workflows. In other words, it does not require a mass migration of content to a centralized repository to work. Instead, it provides a single search experience across all of the various content stores. It all starts by creating a catalog or index of the contents of any repository— file servers, databases, or to service-based sources. “Our vast connector framework allows us to read the content with little impact on the storage device,” explains Goldin. “Voyager is like a librarian who builds a card catalog from a warehouse of books. It doesn’t require you to move all of your books into one big warehouse just to know what’s there,” remarks Goldin.
Once the catalog is created, Voyager’s entity extractors come into play and pull out all the text, metadata and properties. If the client has images, the entity extractor will extract the header information and glean its location as well. Voyager then proceeds to the pipeline, which transforms the data as it is indexed. It creates thumbnails for content, it can also take documents that do not have any type of location information, like a Microsoft Word report with place names, and geotag those documents so they can be found by location.
Finally, Voyager’s web user interface then allows users to access the search engine. Users can find what they need using spatial search, keyword search, as well as filtering and sorting tools common in most web search experiences. Voyager helps to find patterns in the data to make discovery easier. Once users find what they need, Voyager helps users share information within organization by tagging it. Then users can have that content delivered to them. “We can run ETL (Extract, Transform, Load) technology on top of the data to deliver the content to customers,” adds Goldin.
In 2010, Fon Duke, program manager for MDEP at Mojave Desert, was required to publish all of the metadata they managed. The implementation of the incumbent software would have required them to spend more than four months just to set up the system and even after that, he worried about whether it would work once it went live. Voyager gave MDEP a tool that let them access data without requiring metadata, yet leveraging it where available. It only took weeks to implement. Duke was particularly relieved to find an interface that didn’t require GIS expertise. After two weeks, the MDEP team had a website that could serve their map services and support data downloads to constituents and users around the world.
GIS and Enterprise Search
During the formative years of Voyager Search, the emphasis was on geospatial data. However, as the product reached a wider audience, the need to extend the number of content types that could be indexed by Voyager was pressing. Voyager evolved from a search engine for geospatial data to one that could read Office documents, PDF files, and more. “We started expanding the kit that we had for indexing to support some 2,000 different types of content—much of it non-spatial,” recalls Goldin. “Our business has evolved into enterprise search, so we are working with the entire universe of spatial and non-spatial documents.” Voyager Search is now the only vendor in the search industry that is a GIS-enabled, enterprise search engine.
This unique niche has enabled Voyager Search to scale from customers that have hundreds or thousands of records inside of the search engine to those that have billions of items to index. Utilizing the search engine, companies can find relevant content quickly and make informed business decisions in the areas where they need to direct their resources.
Regional Municipality of York, also known as “York Region” was required to replace its metadata portal. York Region also tasked its GIS group to enhance data discovery procedure. They were moving towards an open data policy and needed a website that would support easy access and downloadable data. Voyager Search came to the rescue by providing enhancements to its Commercial Off-The-Shelf (COTS) software to support exhibiting data to external customers. Based on these improvements, York Region was able to create a Data Discovery site with minimal configuration of the out-of-the-box software. York Region simply pointed Voyager at its datasources, and Voyager’s built-in scheduling allowed for an automated indexing of data to ensure any changes to the data or metadata were reflected in the web accessible search.
Voyager also caters to the specific needs of various defense and intelligence organizations and data intensive sectors oil and gas and government. “Companies functioning in these domains need to understand the kind of data they have as well as information for exploration or immediate response mechanism for adverse situations,” says Goldin. Usually, the organizations have various departments, which results in duplication of data. Goldin explains that a state government can buy data to be used by its department of agriculture, but that data might already be available at state’s university database. “Voyager ensures that there is no duplication of data within the organization as a whole, thus helping them reduce cost by avoiding buying existing data twice—or more,” exclaims Goldin.
An organization can opt for either of the two Voyager licenses. “The basic license, Voyager Server, is to help organizations get started indexing quickly to build smaller solutions, while the advanced license, Voyager Server Pro, offers the same indexing capabilities, with additional add-ons like geoprocessing, and integration with enterprise security systems,” says Goldin. “Voyager Server Pro is geared to the needs of enterprise customers.” However, the functionality can be sliced and diced to suit the needs of any sized organization. By purchasing the basic Voyager Server, an organization can adding extensions that are included in a Pro licenses, thus have more economical options. Voyager sells extensions to support Enterprise Security, integration with LDAP and Windows Authentication; Federated Search, to support linking indices together; Geotagging, to add a spatial dimension to non-spatial content; Geoprocessing, to transform and deliver content; and the Document Management Connector, to index other search engines including Microsoft’s Sharepoint and HP’s Autonomy. “A company can purchase just what they want a la carte, or they can buy a Pro license that has everything,” states Goldin.
Voyager Search is increasingly working towards providing a universal search solution that will leverage the existing investments of the enterprises. Voyager Search will continue to enhance its solution to generate meaning from unstructured data with or without geospatial information. Giving users the capability to search, analyze and visualize the content and to make sense to the data, Voyager will deliver on the promise of connecting different dots together. “We want to be known as the company that’s providing out-of-the-box search solution that helps people find everything they need to move forward,” concludes Goldin.