virtual-scrolling
When trying to view extremely large datasets on browsers, traditional lists are unperformant, as each data point (even when "display:none"), exists as a DOM node. The more DOM nodes that exist on a browser, the more work the browser's rendering engine needs to do, which eventually leads to a sluggish user experience.
Pagination is a popular solution to this, but:
- It doesn't feel as fluid and seamless as virtual scrolling (in my opinion) because it forcibly breaks up the data at arbitrary points.
- Pagination doesn't work for datasets that scroll in more than 1 dimension (ie. a 7000 x 7000 matrix).
This is an attempt at trying to implement virtual scrolling.