Overall Satisfaction with MarkLogic
MarkLogic is used the most by the academic publishing unit. There it is used for an end to end data flow for the metadata and full text of our academic publishing products (PsycINFO database, Peer reviewed journals, Books, Psychological Tests and Measures). In addition, we use some of the newer capabilities, such as semantics and javascript support, for our "labs" where we develop cutting edge technology for our content.
- Indexing is a major strength of MarkLogic. The out-of-the-box configuration is set up to handle a combination of text and fielded data. The indexing is also highly configurable. Those configuration options are at the heart of a lot of our high-volume, high-performance applications.
- The industrial strength transactions and security are also a strength, particularly when we are dealing with user-created intellectual property.
- The engineering support is a strength. They are big enough to have a really strong support and engineering staff, but small enough so that a medium-sized customer has access to it. They are very responsive to questions and problem reports.
- The ability to move easily among XML and JSON is a strength.
- There is a steep learning curve to learn to use this tool, particularly the xquery and extensive associated API. The more recent releases and features have been responsive to this concern, but some of the core features still take some learning.
- The javascript implementation is new and there are still some spots where it needs to be made fully compliant with standards and conventions, such as file extensions
- Faster time to market for our content. This was important to us as we are a publisher.
- Increased effectiveness at innovation. We have been able to make advances in analysis and visualizations that have served our internal clients for business intelligence and our external users for more modern info pages.
We had Fast in place when Microsoft had bought it up and was going to change / deprecate it. One of the biggest advantages of MarkLogic for search actually had to do with the rest of the content pipeline - it allowed us to have it all in one technology. On the NoSQL side, we looked at MongoDB a couple years back. At that time, MarkLogic came in stronger on indexing, transaction reliability, and DR options. For us, that was worth using a commercial product.