openHPI / m.e.i.n.e.l

Modern, Extensible and Interactive Number Exploration Library (or short: m.e.i.n.e.l)

Home Page:https://openhpi.github.io/m.e.i.n.e.l/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Technology Stack Decisions: Visualization

flowirtz opened this issue Β· comments

commented

Currently, we are deciding between native D3...

D3 (vanilla)

Good

  • plenty (endless) chart types
  • vanilla => expandable

Bad

  • vanilla => complexity πŸ‘Ž => lots of code for standard charts

... and these four frameworks built on top of D3:

NVD3

Good

  • ~30 chart types

Bad

  • <=IE8 not supported?!

Plotly

Good

  • ~20 chart types
  • quickstart guide

Bad

  • documentation not great (good enough though?)

C3

Good

  • ~16 chart types

Bad

  • last update: spring 2016

Vega lite

Good

  • ~20 chart types

Bad

  • all JSON
  • no interaction?
  • no data updates?
  • components?

Comment further arguments / frameworks below to add to the discussion. 😊

I remember we discusses an hybrid approach, so something like plotly + vanilla

commented

Yes, we also decided on that.
Plotly offers a lot of comfort, but as it makes use of predefined chart types it does not offer the full range of versatility. Therefore, we will also include some parts of D3. 😊