Comparing Mindbreeze to Google Cloud Search, Coveo, Lucidworks, Yippy, Elasticsearch, and Solr
As a Google Premier Partner and GSA implementer, we naturally looked to Google for GSA replacement options. At the time of our evaluation, Google Cloud Search did not have any features available to address indexing on-premise content or serving that content through websites or web applications other than their own cloud search interface. In addition, the status of their security integration options and administration experience remained widely unknown. While it was always clear that Google’s new enterprise search index would be cloud-based, the options for pushing enterprise content from on-premise repositories into that index remain unclear. The initial product direction for Google Cloud Search (previously referred to as Springboard) focused on indexing Google’s G Suite data sources such as Gmail, Google Calendar, and Google Drive. Google has since changed their directional statements to reemphasize their intention to implement indexing mechanisms for on-premise content, but even at the time of this writing, that technology is yet to be released.
Our decision to pursue solutions other than Google, and ultimately partner with Mindbreeze, largely came down to the fact that we couldn’t confidently assure our customers that Google would have a replacement ready (and able to meet the aforementioned requirements) in time for the GSA’s end of life. While I continue to be impressed with Google’s cloud innovations and hope those eventually materialize into enterprise search options, Google Cloud Search remains in its infancy.
As a leader in the enterprise search and knowledge management space, Coveo has ranked well for the past several years among the analyst reports for this market. They have a mature product which made our short list of possible solutions. Two primary concerns surrounded Coveo when compared to Mindbreeze and other vendors. First, their product direction is heavily cloud-focused, available only on Amazon Web Services, with a decreasing investment in on-premise search. Our customer base has a strong need to index on-premise content along with a reasonable amount of customers who prefer the search engine itself be available on premise for governance reasons.
The other concern surrounding Coveo was price. By their own admittance, it is one of the most expensive solutions on the market. Mindbreeze was able to meet our requirements as well or better than Coveo, while providing a stronger commitment to on-premise indexing at a more attractive price point.
Mindbreeze vs. Lucidworks
Lucidworks offers enterprise support for the open source search platform Apache Solr. Their flagship product, Lucidworks Fusion, builds on Solr to add enterprise search features, including connectors and administration interfaces. Our primary reasons for preferring Mindbreeze over Lucidworks concern the ease and speed of both deployment and ongoing administration. While the Fusion platform goes a long way in creating a productized layer on top of Solr, the solution still requires comparatively more work to size, provision, configure, and maintain than Mindbreeze.
Another concern during evaluation was the less-flexible security model available with Lucidworks when compared to Mindbreeze. Mindbreeze supports ACL inheritance from container objects which means if a new user is granted access to a folder containing 50,000 items, only one item (the folder container) must be reindexed to apply the new permissions. Lucidworks applies permissions to each document, so all 50,000 documents would need to be reindexed. While Lucidworks was able to meet our indexing requirements, we felt Mindbreeze offered a shorter time to value, easier ongoing administration, and more flexible security options.
The Yippy Search Appliance attempts to offer close feature parity to the GSA and is available as a cloud solution or an on-premise appliance. Our biggest concern with Yippy, when compared to Mindbreeze, was its immaturity as an enterprise search product. Born out of the Yippy metasearch engine, the Yippy Search Appliance was introduced in 2016 specifically in response to the GSA’s end of life.
The solution is notably absent from consideration by both Forrester and Gartner in their respective 2017 market reports which base inclusion criteria on factors such as referenceable enterprise customer base and proven market presence. The solution also lacks interfaces for customers and partners to create custom connectors to proprietary data sources, an important requirement for many of our customers. As a search appliance, we felt Mindbreeze offered a lower risk solution with a longer history, large reference customer base, and mature feature set.
What about open source options?
Open source options were considered during our evaluation but quickly eliminated due to the vastly greater amount of development time and steeper customer learning curve associated with their implementation. For these reasons, we felt open source search solutions were not a good fit for our customers. Due to the high volume of questions we get regarding these options, I felt it worthwhile to include a few comments on the most popular open sources search tools.
Elasticsearch is a popular open source search and analytics project created by Elastic.co. Elastic itself doesn’t claim to be an enterprise search solution, but they do offer enterprise analytics solutions, and the Elasticsearch technology is often embedded into enterprise applications to provide search functionality. It’s easy to see the confusion this can create. Gartner did not include Elastic in their 2017 Magic Quadrant for Insight Engines. Elastic was included in the Forrester Wave on Cognitive Search and Knowledge Discovery as a nonparticipating vendor where Forrester stated, “Elastic says that it is not in the enterprise search market, but many enterprise customers ask Forrester about Elasticsearch, so we have included Elastic…” As a search tool, we found Elastic was better suited to log analytics than enterprise search as it lacks many enterprise search features including security, connectors, and pre-built search apps.
Apache Solr is a widely used open source search project. Many contributions to the project are made by Lucidworks (mentioned above) whose Fusion platform extends this core technology. Standalone Solr is a framework for creating a custom search engine implementation. While powerful and often used to build highly specialized search tools, it is missing out-of-the-box enterprise features including connectors, administration interfaces, and mechanisms to support secure search.
Apache Lucene is a popular open source search engine framework. It’s a low-level library which implements indexing and search functionality and must be integrated into another application for use. Lucene provides the base search engine behind both Solr and Elasticsearch.
Finding Success with Mindbreeze
After undergoing our evaluation last winter and joining the Mindbreeze partner network, we continue to find Mindbreeze offers an excellent combination of built-in features with tools for extending capabilities when necessary. In the past year we’ve released our Oracle WebCenter Content Connector for Mindbreeze, had ten employees complete the Mindbreeze Expert Certification and helped a long-time customer migrate from GSA to Mindbreeze. If you have any questions about our experience with Mindbreeze or would like to know more, please contact us.