Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

PoC: implement a webpage to visualize the global graph of public comparisons #1795

Open
amatissart opened this issue Oct 3, 2023 · 1 comment
Assignees

Comments

@amatissart
Copy link
Member

amatissart commented Oct 3, 2023

This is an idea suggesting during a recent retrospective session.

TODO:

  • propose an implementation to integrate such a graph on a webpage
  • how to make it visually attractive ?
  • potential interactions to access video details ?
@NatNgs
Copy link
Collaborator

NatNgs commented Oct 4, 2023

Open questions prior to implementation:

  • Do we compute the graph in backend (once every x time) or dynamically in frontend ? Maybe half on the back (preprocessing data) and half on the front (for display) ?
  • What does the dataset looks like (in backend if computation in backend; in frontend if compute in frontend) ?
  • What the graph should look like ? Should it be static (like a picture) or dynamic (like a video or an interactive object) ?

Current idea:

  • Display videos as nodes (maybe squared that become video icons when zooming)
    • The more users have seen (compared at least once) a video, the bigger the node
    • Under 3 users having seen the video => display as a dot (no video icon)
    • Maybe some shadow or border of color according to global tournesol score (black -> red -> yellow -> green)
  • Edges (links between nodes) if comparisons between two videos
    • More comparisons between two videos => Larger the edge
    • Try to make all egdes length the same with graph layout algorithm
      • That will obviously not be possible (graph isn't planar)
      • Comparisons made per multiple persons to have the priority to be of the fixed length (weight force between videos by comparisons count) => for large comparisons not to be stretched too far away
    • Color comparisons with red to green color gradient (green on the + side, red on the - side)
      • Color intensity depends on average comparison value between videos (0=gray, 1=grayish color, ... 10=bright color)

I imagine with this display to see groups where videos have been compared a lot to eachother (so maybe groups by video airing date or theme), and globaly red and green zones indicating where good and bad video are. I expect some moon like dots all around (all the less compared videos), displaying all the diversity available on Tournesol

Demo picture incoming, feel free to give others visualization styles idea or to suggest improvements to this idea above

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants