Last month, I wrote about Elastic App Search which is Elastic’s curated experience of Elasticsearch. Since I wrote that, Elastic has officially released App Search from beta to general availability. More importantly, Elastic has clarified its licensing in the blog post announcing the release.
Best of all, Elastic App Search is free to use alongside the default distribution of the Elastic Stack.”
And with that, Elastic effectively wiped the market for about 2 dozen companies who offer ‘search’ solutions. Now they are doing it again with Enterprise Search. Enterprise search’s goal is to create a single source of truth within your enterprises content.
Elastic Enterprise Search is the former beta product from Swiftype which is now re-released under the Elastic flag. I remember seeing a demo a couple of years ago and at the time was impressed by the simplicity of the interface, its breadth of features. I joked to some of the principles at Swiftype at the time that Google would have been better off by acquiring Swiftype versus putting the R&D into Springboard.
Like App Search, this product obliterates many of the complexities in deploying Elasticsearch for organizations. It raises (or lowers depending on how you look at it) the bar for the minimum expectation for what Enterprise Search can be out of the box.
Steps to get started
(remember – not for production use yet!)
1 – Get Elasticsearch – The simplest way to get started is to run everything locally either by installing Elasticsearch via Docker, brew, or Chocolatey. Be sure to run the Elasticsearch version supported under this beta release which is 6.7.1 and not 7.2.x.
2 – Download and Unpack Enterprise Search – With Elasticsearch running on locally, goto Elastic’s website and grab the latest release from https://www.elastic.co/downloads/enterprise-search. Uncompress this.
3 – Add one line to the yml file. Edit the config/enterprise_search.yml by adding allow_es_settings_modification: true to it.
Next you will want to index some content. If you have a Google Account, you can easily create a connector to Google using these instructions.
After you put in the Oauth Keyset, you have it as available source as follows:
Now you can setup a source and begin to index content.
In Google Drive’s case, you do need to wait sometime before the project and api keys are federated to Google’s edge servers. Once everything is in order, it will start indexing. I got in about 3k documents in a few minutes.
Now is where you will start to see the magic:
There’s going to be some interesting releases over the next few months. I hope that Elastic releases the ability to write connectors inside of the product and not simply posting to the ingestion API. But next time, we’ll dive a little more into this product.