beeminding a conversation stack

I’m hoping some of you folk will have good ideas about this stuff.

Part 1: Data structures for conversations?

Since hearing Robin Weis speak at 2017 QS conference in Amsterdam last summer, I’ve been thinking about ways to be more intentional about tracking open conversations and keeping track of conversations that I want to have with my partner.

A lot of these discussions happen over email, since that’s a nice medium for longer form, more discursive, or in depth topics. So there’s already a record of a lot of our conversation, but I still think it would be interesting to have some more structure to it. I wonder if it would work to, off the top of my head, use github issues in a private repo for this… we could have discussion back and forth, use labels, schedule things, search for topics, explicitly close topics when we agree on a resolution (or agree that it’s stalled)…

Has anyone done anything like this? How crazy (awesome) does I sound? Who keeps an actual conversation stack of their conversations with their partner? Or other tales of unusual tracking of conversations, or unusual tools / software for conversing?

Part 2: Beeminding the conversation stack (of course)

Any ideas on this? I thought at first that I’d label all the email threads with a special label and start an inbox goal for those conversations, but I don’t think that’s really quite what I want. I want to allow room for the stack to grow, so it’s not strictly a backlog problem. But I also do want to incentivize myself to keep touching base and not let old topics starve (or explicitly resolve / close them, I suppose if the topic is exhausted). So beeminding number of threads with a particular label doesn’t seem like quite the right metric, because I don’t just want to remove threads from the backlog, I want to incentivize engaging, and give credit for being responsive and stuff.


What a cool article! I want to read that again, actually, let my subconscious noodle on it a bit.

I don’t have a ton to add logistically – my long distance relationships usually had one massive email thread and just answered Everything All The Time – and we never got down to tracking things like that, much less incentivizing them. But…but this topic is so cool that I wanted to reply to it.

What’s the qualitative goal? Obviously it could be and probably is a lot of things. But e.g. is the goal to make sure the partners are covering a lot of topics with each other? Or, could be that the goal is to have a greater percent of the conversation go beyond “how was your day” to include deeper/discursive topics. Perhaps it’s response time – if partners respond quickly to each other, then that shows respect/affection/prioritization.

I think there can be metrics that track all those things. You probably know all these. You can track “Add # Conversation Topics to List per (time period).” Or “Reference # Deeper Conversation Topics per (time period).” Response time is easy – FRT is an old SaaS support metric – “Average First Reply Time to new email.” Or something.

Thing I don’t know how to quantify or incentivize is quality of response / quality of thought. I don’t know if this is something you even care about. Top of my head I would think that Length of Response (word count) would match well with Time Spent Responding, which might also match with Quality of Response, but that’s assuming that more words and more time = better thinking, and also assuming that the opposite is true. I don’t think it is. I have written heartfelt things quickly, and insightful things concisely. (Though, not often.)

Time Spent Responding could be a metric in itself, too, btw. “Spend X Minutes per Week on Discursive Emails.”

Not sure if any of this is helpful. But thanks for posting this idea out there.


If you track via GitHub, you could create milestones and use those for progress tracking. E.g., newly proposed conversation topics could go into the general backlog, then (say) once a month, you and your partner could go through the backlog and add topics to either a annual, quarterly, or monthly milestone (or several concurrent milestones of different lengths). You could then Beemind the burndown chart for that milestone.

A weirder but perhaps more interesting thing to Beemind would be the time it takes to close a randomly-selected topic. For example, write a script to randomly select a topic and email it to you and your partner. When the script detects that topic has been closed, it adds a +1 datapoint to a Beeminder goal to close n topics a week/month, randomly selects another topic, and emails it to you and your partner. The idea is that you’d only have one Focus Topic at any given time, although of course there’d be no barrier to discussing any other topic concurrently.

An advantage here would be that the random topic selection will occasionally select difficult-but-important topics that neither of you may want to bring up as a result of akrasia[1]. It would also sometimes select easy or fun topics in proportion to the number of those topics pending, which could help encourage both of you to record those conversation topics on the list. That could keep both of you in the habit of adding items to the topic tracker as soon as they came to mind, maximizing your opportunities to both collect data and have interesting and useful conversations.

[1] I’m probably just projecting my own cowardice at bringing up difficult topics, though.