Coding of religion

[philipp-kempgen]

The current choice of religions is not that great, because it mixes up levels of detail.

https://github.com/animina-dating/animina/blob/8ae0328255735fd4660e0ac7e9ddd854a84170a7/priv/repo/migrations/20240228210500_seed_flag.exs#L1347

It's also a topic which you can easily get wrong. I don't think this is the number one priority, but I would adopt an established scheme instead of inventing a custom scheme:

Bormann, Nils-Christian: EPR Ethnic Dimensions Data: Codebook. Version 2021. https://icr.ethz.ch/data/epr/ed/ (Look at the Codebook, Table 1: EPR-ED Classification of Religious Segments, on page 5.)

Think of this as a tree.

• Someone may describe their religion as "EARC1A1" (Southern Taoism). Give them a choice of what level of detail they want to match with their partner, e.g.:

  • "EARC1A1" (Southern Taoism)
  • "EARC1A" (Taoism)
  • "EARC1" (Shenism)
  • "EARC" (Chinese Religions)
  • "EAR" (EAST ASIAN RELIGIONS)
  • "" (any)

• Someone else may not want to divulge so much detail and simply describe their religion as "EARC" (Chinese Religions) or "" (unspecified).

In theory someone describing their religion as e.g. "EARC1A1" (Southern Taoism) or "" (unspecified) might specifically look for a partner outside of their own "branch", such as "ERB2A" (Nichiren Buddhism), but I don't think this is an actual use case, so I wouldn't account for searches outside of the branch.

Maybe don't include the most detailed level in a <select> input so the usability doesn't suffer.

@wintermeyer You may potentially want to do that after the MVP milestone, I think. Just leave the issue for future reference then.

It's just an example of data that can't really be coded as flags.

[wintermeyer]

The religion is a test ballon. I just want to see if people use it. It is good enough for now. We might improve or delete these flags in the future.