Suppose there is a social group that likes to communicate via group chat. We’ll call its members Akira, Bilal, Chiara and Dan and refer to them as A, B, C and D for brevity.

The subset saturation law of group chats states that eventually, group chats will splinter off the group chat corresponding to all possible subsets of the main group. So in this case given that A, B, C, D is the main chat, the following chats will eventually appear:

  • B, C, D
  • A, C, D
  • A, B, D
  • A, B, C
  • 1

Splinter chats form to arrange a surprise birthday for one of the main group members, arrange an outing when one or several members aren’t available, organise an activity in which only some members of the group have an interest, and so on. In the example given, the main group has only 4 members but this is theoretically extensible to chats and social groups with large numbers of members.

One might take the degree of subset saturation as an index of the strength of ties between members of the group. Larger groups would probably score lower on such an index compared to smaller groups, unless the large group had been around for a very long time. That makes sense - one would expect that smaller groups, or well established groups, would be more tight-knit. Alternatively one might take

  1. 1-1 chats should be included in this list but have been omitted. They aren’t what people normally call a group chat, and the list is illustrative enough as it is.