RFC: Protocol for initiation/termination of dialogue

Abstract: Many people sometimes wish to talk to other via instant messaging applications about their lifes, news, common interests and other topics without having important information to tell and without wanting to make a request (we will call these conversations “chats”). However, it may be difficult to find out whether your interlocutor wants to chat or would prefer to end the conversation because we encounter problems related to courtesy. For these reasons, we present a new protocol for chats that could save time for both interlocutors and help them lead the conversation for exactly as much as they both desire.

This protocol concerns chats over internet only.

We consider that each interlocutor, at each moment during a chat, is in one of these states vis-a-vis the other:

  • L1: I absolutely need to talk to you, even if you don’t want to, because talking to you will likely make me happy, and I really need that right now.
  • L2: I would rather tallk to you than do anything else I could be doing at this moment (e.g. schoolwork, games…)
  • L3: I wouldn’t mind talking to you but I would just as gladly do other things (e.g. schoolwork, talking to somebody else…)
  • L4: I would prefer to do something else than talk to you right now, but if you want to talk, I will be glad to.

The protocol takes effect at two points in time: at initiation time and whenever you think of ending the dialogue to the satisfaction of both parties.

  • When you initiate a conversation, you announce the level at which you are and thus the “negotiation” begins. Based on the result of this negotiation, the conversation will either happen or not.
  • If the conversation is already in progress and one participant realizes he likes it less than before, he announces the change (deterioration) of his level, and this causes “negotiation” to occur again. This negotiation can end the dialogue.


  • If you announce the L1 level, it is discourteous if the other person announces their own level (unless it is also L1); it is equally discourteous to refuse to talk, regardless of one’s level or what’s going on around him or her IRL. The conversation must begin or continue.
  • If you announce the L2 level, then:
    • If the other person is at L2, the conversation begins/continues.
    • If the other person is at L3, the conversation begins/continues.
    • If the other person is at L4, the conversation begins/continues, but you should try to keep it relatively short. The other person (L4) may explicitly say that he is willing to lead a long conversation regardless – that is an expression of generosity; however, it is then discourteous if he or she inwardly thinks of it as their sacrifice, or – even worse – says as much aloud.
  • If you announce the L3 level, then:
    • If the other person is at L2, the conversation begins/continue.
    • If the other person is at L3, the protocol does not specify what should happen. Participants decide based on the situation whether they want to keep talking to each other or not.
    • If the other person is at L4, the conversation ends.
  • If you announce the L4 level (only in the middle of a conversation, otherwise you wouldn’t even begin to talk), then:
    • If the other person is at L2, the conversation begins/continue, but the other person will try to keep it relatively short (see L2/L4).
    • If the other person is at L3 or L4, the conversation ends.

Example 1:

A: “Hello. How are you? (L2)”
B: “Hi! Long time no see. I’m also at L2. How are your exams going?”
… conversation …
A: “Hey, it was nice to talk to you but maybe I should do some Algebra schoolwork today, too (L4).”
B: “Wait, wouldn’t you tell me a bit more about your diploma thesis please? How’s it going?” (implicit L2)
A: “I would be happy to. I don’t need to do the schoolwork today…” (A must not end now, but is permitted to answer only shortly)
…conversation about the thesis…
B: “Ok, bye.”
A: “Bye.”

Use of the protocol made it apparent to B that he or she should be slowly leading the conversation to its end.

Example 2:

A: “Hello, B. Do you remember how to prove the Steinitz exchange lemma?” (a request, not yet chatting)
B: “I don’t, unfortunately. But wait… here [link] is a textbook that explains it pretty well.”
A: “Ok, thanks. By the way, how are you? (L3)” (offer of chat)
B: “Pretty good, I’m watching TV right now (L3/L4). Bye for now!”
A: “kk, kthx bye.”

Use of the protocol prevented a chat that nobody was really interested in from occurring.

Notes: Most of the time, every one of us is in the L4 state vis-a-vis most other people in the world, or perhaps even in some kind of L5 state – that’s why we’re currently doing what we’re doing rather than being in a conversation at all times.

Many people are able to deduce their interlocutor’s level in many situations without using explicit level numbers (such as implicit L2 in example 1), however, this is not as easy it seems and there are very few masters who would be able to do it in every situation. Surveys that demonstrate this can be found in literature.

Even in such situations, however, this level system may work as a good theoretical basis.

Conclusion: Not all people will be interested in using this protocol and not all pairs of people either. Even so, it is our belief that people who find this protocol natural or useful may save time they would rather spend elsewhere and also strength their friendships because they will eliminate a number of misunderstandings.

There are certainly errors in the protocol right now and it isn’t useful too often. Despite that, it could serve as a good basis for further research in this area. This blog post is a request for comments – please use the comment feature on this webpage.

Leave a Reply