Epistemic status: It appears like it works for me. Not meant to be a hard/fast rule.

A frame I frequently examine conversations through is "are we building a product together, or not?"

Many conversations (online or in person) are "just sorta hanging out." Just-sorta-hanging-out can be quite important, both because it's fun, and as part of building up friendships, etc. But if one or more participants isn't having fun, it's more likely that the conversation should end, or change.

"Building a product" conversations have a goal, and the goal is "create something that someone is going to use someday." Here's a few types of products you might be building:

  • A literal product
    • i.e. programming a website or designing a widget
  • A relationship
    • you and a friend, or a romantic partner, are trying to improve the foundations of trust or communication
  • A party
    • you and your roommates are planning a surprise birthday bash for a friend
  • A new set of norms
    • you and your friends, or your coworkers, all agree that something about the current social equilibrium is off, and should be replaced by something new
  • A fun idea
    • you're doing random brainstorming on something kinda crazy, but you still want it to be the best something-kinda-crazy that it can be. Maybe it'll later turn into a literal product, or maybe it'll turn into a fun story to tell.
  • A felt sense
    • Maybe the object level discussion isn't about the product. Maybe you're exploring a philosophical idea. But your goal isn't to output a useful philosophical idea. Your goal might instead be:
      • develop a sense of what a particular flavor of confusion feels like
      • develop some shared intuitions that underlie the philosophical idea

In some of those cases, there are "just hanging out" conversations you might also have about it. You can talk about "a new set of norms" that's less goal directed and more just meandering and exploratory. A fun idea can certainly be nothing more than "a fun idea."

But, I find product-building conversations way more interesting, most of the time. In particular, I'm willing to invest a lot more effort into a conversation if it seems like it's about building something.

Product-building-conversations don't have to be rushed, or laser-focused. Sometimes the best way to design a product is to have a long, meandering conversation that gives you time to soak in each facet of the design constraints, or share intuitions about it.

Building a product imposes constraints on the conversation

There are lots of different conversational styles you can be building a product in, depending on your culture. Gruff auto-mechanics might brusquely swear at each other when they screw something up. Some companies might have weird politeness norms, and those norms might be different in the US vs Japan while designing their latest widget.

I discussed some examples recently of (what I saw as) product-building in a collaborative frame. My comments then were highlighting my own preferred norms. But in this post now I'm trying to make a broader point.

I have best guesses about what sort of norms make for productive product building.
But I have a much more important meta-level belief: If you're building a product together, there are some kinds of constraints on "what sort of conversation is actually worth having?". And one of the most important constraints is:

If you're not building the same product, you're going to have a bad time.

When a conversation seems to be getting confused, and people talking past each other, some questions worth asking might be:

  • Are we building a product, or just sorta hanging out?
  • If we're just sorta hanging out, are we having fun, or otherwise getting value? If not, stop, or change something about the conversation.
  • If we are building a product, do we have agreement about what product we're building? Do we have meta-level agreement that, if we don't agree on what product we're building, at least one of our goals should be to figure that out?

For me, at least, that last question dictates whether this is a conversation that I'm going to put a particular kind of serious effort into.

New to LessWrong?

New Comment
6 comments, sorted by Click to highlight new comments since: Today at 6:29 PM

There's a general mental slider I use called "intentionality" which is something like "how much am I treating this situation as a product vs. an exploration. It seems related to explore/exploit and babble/prune distinctions, as well as the concept of blending with or unblending from subagents.

One of the reasons you might want to go into a conversation with low intentionality is that it provides a wider area to explore and you can discover more potential exciting products to build. Another reason is that it doesn't make you lose sight of your other products/intentions as you're going (for instance, a chance to build the relationship while you're talking about a technical problem.)

One move I think is useful is to deliberately signal when you're changing your levels of intentionality. For instance, someone says something and I reply with "Now THAT's interesting, lets talk about that." Alternatively, when a particular product feels like its' fizzling out "I notice that I'm not as interested in this topic anymore. Do we want to keep talking about it?"

Something which might or might not be confusing here (I can't tell from the way you phrased things) is that I see "intentionality" as sort of orthogonal to "are we building a product or not?". This might make the metaphor more confusing than helpful.

(Since, even at a literal company building literal products, there's value in conversations that are exploratory. But in that case the "sub-product" that you're working on might be "an exploration of the space.")

Since, even at a literal company building literal products, there's value in conversations that are exploratory. But in that case the "sub-product" that you're working on might be "an exploration of the space.")

I think even in this case (as well as the conversation one) there's "levels of intentionality" that restrict your conscious and subconscious search space. For instance, at LW you probably have conversations that are "What features could be good?" and some conversations that are "What would this feature look like?" and some conversations that are "How might we solve the problem of common knowledge and peer review".

I see these sorts of things as levels of intentionality, and still fit into the frame of babble/prune, explore/exploit etc.

Similarly, you could have a conversation where you're building a product, a conversation where you're exploring different products to build, or a conversation where you're open to working on multiple products at once.

On the other hand, this is stretching the metaphor a bit.

Fascinating. The word "project" encapsulates a lot of these — from software to a party — establishing norms without a more tangible/legible "stake in the ground" (say, starting a Running Club) seems more ephemeral and we don't really have good words for it.

Thought-provoking post here. There's a gap in the vocabulary around this.

There's a more general version of this post that might have said "does your conversation have a goal, or not?". I thought about using that lens, but something about the product-ness makes things more concrete to me.

[edit: I missed that you said 'project', as opposed to 'product.' Project might work, but I think 'product' still carries a bit more of the connotations that I care about. There's a particular way that I work together with someone on a team if we're building something that we're planning to sell. Or rather, there are particular failure modes I don't fall into there. And part of what I'm (weakly) claiming is that those intuitions are useful to port over]

I really like this distinction between "just hanging out" conversations and "product building" conversations. And I too usually prefer the latter.