Introduction

The abundance of social networks poses questions regarding their similarities and differences to identify redundancies, but potentially also missing services in the social network landscape.

Similarities

Social networks have the basic feature in common of connecting people via electronic channels.

Communication can be based on one or more of the following basic types of media:

• text (with varying character count per post)
• image
• audio (live, recorded)
• video (live, recorded)

At present most social networks allow sending all types of media although there are differences in the usability of using these types of media, for example:

• text, e.g. E-mail, IRC
• image, e.g. Instagram
• audio, e.g. Spotify
• video, e.g. TikTok, YouTube, PeerTube

Differences

Social networks can be differentiated based on the following features.

Objective of communication:

• Publicity, e.g. X/Twitter, Facebook, LinkedIn
• Privacy, e.g. Signal, Telegram, Wired

Object of interest:

• influencer, e.g. Taylor Swift, Antonio Guterres
• topic, e.g. programming, knitting

Identifiability of the agents (assuming that any person can be identified with sufficient motivation and resources):

• non-identifiable, i.e. anonymous, single-use (“burner”) user/nick name, e.g. abc123
• partial identifiable, i.e. user name which is identifiable for an in-group, e.g. first name or known characteristic, e.g. TommyMusicBerlin
• fully identifiable, i.e. First Name and Last Name, Personal Foto

Directionality of communication (symmetric, asymmetric):

• Limits for sender, e.g. types of media, character limit, types of replies
• Limits for receiver, e.g. types of media, character limit, types of replies

Note: Communication is asymmetric if the sender has more permissions than the receivers, for example, if in a moderated channels receivers can only read posts, but cannot reply or only reply with a reaction (“like”). Communication is symmetric if senders and receivers have the same permissions to post messages. For example, YouTube or PeerTube allows the sender to publish videos, but only textual comments or reactions as a response.

Level of Encryption:

• unencrypted, e.g. plain text messages
• encrypted (e.g. using various encryption technologies, such as TLS, PGP, Bluefish, OTR, OMEMO, …)
  • client-to-server/server-to-client
  • client-to-client (end-to-end, E2E)

Level of Privacy:

• private
  • sender-selected receivers, e.g. invited by moderator
  • receiver-selected receivers, e.g. invited by group members
• public
  • accessible to accepted receivers (followers) only
  • accessible to all receivers (followers)
  • accessible to all users of social network, e.g. on X/Twitter
  • accessible to all users of the internet, e.g. search machines

Note: I have distinguished between Level of Encryption and Level of Privacy, because encrypted messages may not be private if the sender cannot select the receivers and verify their identities, e.g. if anyone can join a Telegram channel any encrypted messages may not be considered private.

Level of System Integrity:

• easy to compromise, e.g. unencrypted email via open WIFI
• difficult to compromise, e.g. end-to-end encrypted message via VPN tunnel

Type of Architecture:

• Centralized, e.g. most proprietary networks
• Closed-federated, i.e. federation on server-level, ie users can communicate with any other user hosted on servers of the same federation (e.g. IRC server networks such as libera.chat)
• Open-federated, i.e. federation on user-level, ie users can communicate with any other user hosted on any other server using the same protocol, e.g. E-mail (SMTP/IMAP), MASTODON (Activity Pub), Jabber (XMPP), …