Thinking out loud: using follow/unfollow you transmit a signal with no information. using the timing between the signal you are encoding symbols. I think this is called Pulse Position Modulation[0]

First step is to characterize the delta in the timestamps. That will give you an idea of what time-between-follows (TBF) should be to ensure reliable detection.

You could probably make a client for this. Build comms on top of twitter protocol.

Haven't used twitter in over 10 years, but this kinda makes me want to again.

Side note: reminds me of why gmail scans drafts. Criminals used a signal account and create drafts to communicate back and forth. This prevented detection because intel agencies scan email traffic, not drafts.

EDIT: why stop with twitter. what makes this powerful is that it leverages a feature every platform has. Youtube, Twitch, you name it. Let's go my dude.

[0] https://en.wikipedia.org/wiki/Pulse-position_modulation

No, because the notification timing is not reliable. It can vary a lot depending on the load on that microservice at any given time. Also, how would you handle dits, dahs, and gaps?

You only have 1 variable in your control, the time in-between follows. (I say this because there is no notification for "unfollowing" someone)

You need to represent 2 variables : the dot/dash to represent symbols, and the dit, the time in-between symbols.

If you want to do this you need some sort of 2nd action, a comment or a twitter like, that way we could represent the dot/dash with either action, and then space them out to time the silence.

Were you blocked?

More seriously, the events end up in a list, so a recipient would more likely "decode" them by comparing timestamps instead of "live" watching the actual sequence.

But I suppose if you timed your follows/unfollows really carefully, you could get certain patterns in the timestamps to appear which could be used to distinguish "long" and "short" pulses.

Let's say you could follow and unfollow once a second and they knew it. Then you could replicate the dots and dashes by the length of time you toggled the status plus the gaps where you weren't toggling at all.

Eg in the first ten seconds 10 toggles, in the next 10 seconds a 5 second gap then 5 toggles. That would be dash, quiet, dot.

I had the same thought.

.. ..-. / -.-- --- ..- / -.-. .- -. / .-. . .- -.. / - .... .. ... --..-- / - --- ..- -.-. .... / --. .-. .- ... ...

With just follow/unfollowing, you can't truly encode Morse code. You would have to override a metadata field or agree on an external clock source to overcome the idempotent follow/unfollow action and encode language into the state machine.

Do you mean following and unfollowing in some sort of rapid succession?