Welcome to the RD thread!

This is a place for casual random chat and discussion.

A reminder for everyone to always follow the community rules and observe the Code of Conduct.

Image

Mobile apps:

Quick tips:

Footnotes:

  • Daily pixel art by Paul Sabado
  • Report inappropriate comments and violators
  • Message the moderation team for any issues
  • megane-kun@lemm.ee
    link
    fedilink
    English
    arrow-up
    12
    ·
    edit-2
    1 year ago

    I was bored/procrastinating on some work.

    It’s not really an in-depth thing, I just used Lemmy’s search capability and took note of the number of responses and the number of deleted responses. I added the number of responses with the number of deleted responses (a negative number for new responses, a positive number for new responses is ignored). I did it for both Lemmy.world and Lemm.ee and took the average. For the smoothed out version, I just took the exponential moving average with the following factors:

    • sampling frequency = 1 day
    • average age (ϰ)= 7 days
    • EMA factor (α) = 1/(ϰ+1) = 0.125

    That is, I am trying to smooth the values accounting for weekly variability in activity.

    I guess I could have ignored the spike in activity during the Reddit blackout, but it seems that the smoothing took care of that quite well.


    EDIT: Replaced chart image with a one that has a black background for legibility for people using light mode.