So the JSON file store was not the method to use, instead I looked at storing the data in a Database, in this case a MySQL database. Video (7min).
My first exploration was to just have a large amount of data in the table and be able to plot it on a map with some information regarding the data.Map of Housing Assets by Suburb . Initially I looked at just identifying the properties by suburb, colouring them to suburb and giving general information about the properties in a pop up box in the maps. This had just over 300 rows of data.
Then I increased the information on each property by processing the data and of setting the Latitude coordinate for specific elements so that you could differentiate the roof/wall/door/window icons
Map of Housing split into external elements. You need to zoom in on this map to a group or just one property to see the different elements of the same building. This was done without data cleaning on a large dataset and Knime was used to shift coordinated in the latitude for the different elements. I did not try and style the icons as this is just a demonstration of using a larger dataset and displaying the data on a map. Each external element had a 3 letter code, Rof, Exw, Win, Dor etc and on clicking icon a pop-up box gives you data on that element.
As this is a dataset with over 3000 rows, and the data is written to XML before being displayed, it is slow to load the whole dataset onto the map.
An alternative is to select an area first then find the data, this way, only a sub-set of the data will be displayed and it will load much faster. There are a couple of ways to do this, one that I researched was with a polyline, where you drew a boundary around an area and the query only fetched the information of properties withing that bounded area.
Another method is the example of Shops in Cities. In this interactive example you type in a City , eg Boston, Los Angeles etc, and select a radius, say 25 miles, the search will only look in this area and display only shops within that radius, rather than in the whole of the dataset, thus speeding up the search result and display on the map.
This is the 2ndpost and video about this subject, the first post link at top of post.
The following one is about mapping that shows how to use the map as a place to link data from multiple sources and have a common reference point to find that data. How to use Maps to find Data on Assets.