All our sample data sets are currently using a time-based index pattern. We often come across issues or features not behaving very well with non time-based index patterns. This might be, due to the fact, that most developers use the sample data sets during development. I think we should add a sample data set, that isn't using a time field, so that there is an easy way to test against those data.
I've no special suggestions about the content of the dataset, so every suggestion welcome.
A library, catalogue, or inventory of product (anonymized employee data could be interesting)
Could use various data sets from numbeo (yes, you could graph all of this stuff over time, but a snapshot of data would be just as good of a data set) (saving you a click, it's stuff like cost of living data, transportation data, property prices, crime, etc for all over the world)
Relationship between soil composition and properties of plants (can extend this idea to anything that is "state" to "result of state" data sets)
All our sample data sets are currently using a time-based index pattern. We often come across issues or features not behaving very well with non time-based index patterns. This might be, due to the fact, that most developers use the sample data sets during development. I think we should add a sample data set, that isn't using a time field, so that there is an easy way to test against those data.
I've no special suggestions about the content of the dataset, so every suggestion welcome.