Six months after revealing that it planned to fork Elasticsearch, Amazon has delivered a production-ready version called OpenSearch. The Java-based open-source search and analytics engine is used for large volumes of data, in conjunction with a visualization dashboard called Kibana.
Elasticsearch is mostly used by enterprises for tasks that require data or document retrieval, log data analysis, security analysis, and more. Amazon opted to create the fork after the company behind both Elasticsearch and Kibana complained that enterprises tend to take without giving back, and made changes to protest that.
To fork off
It all started when the company said it was transitioning from a permissive Apache License to a dual source-available Server-Side Public License (SSPL), as well as a proprietary Elastic License. The company also said that the motive for doing this was to prevent cloud service providers (like Amazon Web Services) from offering Elasticsearch as a service without collaborating on the project.
For an open-source project, forking is often seen as the way forward when two entities using it have different or conflicting priorities. The process involves taking the original source code and creating a separate project maintained by different people, in conjunction with the original project.
Fork to mainstream
Amazon’s fork now wants to be the official open-source project and is offered under an Apache 2.0 license. Meanwhile, the original Elasticsearch is now proprietary software, classed as ‘free and open.’ The first release of OpenSearch was back in April, but it was not production-ready then. The release was still in ‘Alpha’ and only recommended for testing purposes. Version 1.0 is available on GitHub now, with Amazon anxious to remind us all that it is an open-source offering. Maybe the uptick in OpenSearch users may take the fork mainstream.