social federation statistics project https://loadaverage.org/fedstats

romina 0cd3ed3a1a describe sample data 2 months ago
etc 2d81b35a3c update config url, revise writeup 2 months ago
share 29446cc40b add flow config, icons 2 months ago
LICENSE c825956c15 Initial commit 2 months ago
Makefile 2d81b35a3c update config url, revise writeup 2 months ago
README.md 0cd3ed3a1a describe sample data 2 months ago

README.md

#fedstats - social federation statistics project

if a man knows not to which port he sails, no wind is favorable.

icon

gnu/social (/n(j)uː/ /səʊʃl̩/) is a free software project to create a decentralized social network. the network is already quite large, but the full picture is far from well understood. extracting useful metrics from this system has been a fun and interesting challenge. for more information about the federation and gnu/social, see gnu.io. the goal of this project is to have an accurate and transparent system for measuring the public activity of the federation. #fedstats measures network activity by taking snapshots of the whole known network timeline of several popular gnu/social servers.

each notice has the following activity data extracted from it:

{
  "user": "xj9 (https://social.heldscal.la/user/27176)",
  "url": "https://social.heldscal.la/notice/1465766",
  "feed": "https://sealion.club/main/all",
  "id": "tag:social.heldscal.la,2017-04-02:noticeId=1465766:objectType=note"
}

we keep track of every url that has been seen in the last three days and use the list to filter out duplicates. another group of agents process this data into the following reports.

  • posts (24h)
  • posts (30d)
  • posts (30d peak)
  • active users (24h)
  • active users (30d)
  • active users (30d peak)
  • new users (24h)
  • new users (30d)
  • new users (30d peak)

these are formatted and published to gnu/social daily at 0900 UTC. the data used to generate these reports is available here:

  • https://huginn.bbnet.io/_raw/fedstats/daily.json
  • https://huginn.bbnet.io/_raw/fedstats/new_users.json

the next revision will utilize a bot instance to discover the full size of the network and improve the accuracy of the survey. i have only begun to scratch the surface of what sort of useful information i can extract from the data that has been collected so far. i hope it will be of some use to the community.

questions and comments are welcome.

see also

  • https://en.wikipedia.org/wiki/Daily_active_users
  • https://en.wikipedia.org/wiki/Monthly_active_users
  • https://en.wikipedia.org/wiki/Customer_engagement
  • https://github.com/cantino/huginn

contact

@xj9