Chapter 5:
Raster tile layers are large and can take a long time to render, but once they are created, they are an important piece that works well as a base for other projects.
Vector tile layers are smaller and can be updated easier because it generates tiles based on data density.
Map image layers are mostly used for large datasets that need server-side rendering.
Map layers aren’t reusable because it’s rare for two map requests to cover the same data. All map requests read from the database and generate after the request, so they take the most time. Raster and vector tiles don’t read from the database because they are already cached or generated on the web server, so they are faster. Feature tile requests must be generated the first time they are requested, but will then be cached in the web server and faster to load.
Raster layer: the labels don’t rotate to stay upright, it is less clear, and it doesn’t display when zoomed too far. Vector layer: the labels stay upright when the map is rotated, it is clearer , and it displays at closer zoom levels.
Chapter 6:
Spatiotemporal data is observations of objects and events that change over time.
Moving: something that moves such as live feeds of vehicles.
Discrete: something that “just happens” such as earthquakes, geotagged feeds, criminal incidents.
Stationary: stands still but values change such as wind speed and direction from a weather station, highway traffic speed.
Change: change or growth such as perimeters of wildfires and floods.
Application based on 5 & 6:
I could make an application that tracks when the trees around campus drop their leaves, grow new leaves, grow flowers, grow fruit, drop fruit, etc. Pairing it with a gauge noting how many are flowering, growing fruit, etc. a list of what types of trees they are, and a map that shows where they are on campus.