Open conversations are worthwhile

One of the hardest parts of participating in open source projects is, in my experience, having conversations in the open. It seems like such an obvious thing to do, but it’s easy to fall into the “I’ll just send a direct message” anti-pattern. Even when you know better, it happens. I posted this on LinkedIn a while back:

Here’s a Slack feature I would non-ironically love: DM tokens.

Want to send someone a DM? That’ll cost you. Run out of tokens? No more DMs until your pool refills next week!

Why have this feature? It encourages using channels for conversations. Many 1:1 or small group conversations leads to fragmented discussions. People aren’t informed of things they need to know about. Valuable feedback gets missed. Time is wasted having the same conversation multiple times.

I’ve seen this play out time and time again both in companies and open source communities. There are valid reasons to hesitate about having “public” conversations, and it’s a hard skill to build, but the long-term payoff is worthwhile.

While the immediate context was intra-company communication, it applies just as much to open source projects.

Why avoid open conversations?

There are a few reasons that immediately come to mind when thinking about why people fall into the direct message trap.

First and — for me, at least — foremost is the fear of embarrassment: “My question is so stupid that I really don’t want everyone to see what a dipshit I am.” That’s a real concern, both for your ego and also for building credibility in a project. It’s hard to get people to trust you if they think you’re not smart. Of course, the reality is that the vast majority of questions aren’t stupid. They’re an opportunity for growth, for catching undocumented assumptions, for highlighting gaps in your community onboarding. I’m always surprised at the number of really smart people I interact with that don’t know what I consider a basic concept. We all know different things.

Secondly, there’s a fear of being too noisy or bothering everyone. We all want to be seen as a team player, especially in communities where everyone is there by choice. But seeing conversations in public means everyone can follow along if they choose to. And they can always ignore it if they’re not interested.

Lastly, there’s the fear that you’ll get sealioned or have your words misrepresented by bad faith actors. That happens too often (the right amount is zero), and sadly the best approach is to ignore or ban those people. It takes a lot of effort to tune out the noise, but the benefits outweigh this effort.

Benefits of open conversations

The main benefit to open conversations is transparency, both in the present and (if the conversations are persistent) in the future. People who want to passively stay informed can easily do that because they have access to the conversation. People who tomorrow will ask the same question that you asked today can find the answer without having to ask again. You’re leaving a trail of information for those who want it.

It also promotes better decisions. Someone might have good input on a decision, but if they don’t know one’s being made, they can’t share it. The input you miss might waste hours of your time, introduce buggy behavior, or other unpleasant outcomes.

The future of open conversations

Just the other day I read a post by Michiel Buddingh called “The Enclosure feedback loop“. Buddingh argues that generative AI chatbots cut off the material that the next generation of developers learns from. Instead of finding an existing answer in StackOverflow or similar sites, the conversations remain within a single user’s history. No other human can learn from it, but the AI company gets to train their model.

When an open source project uses Discord, Slack, or other walled-garden communication tools, there’s a similar effect. It’s nearly impossible to measure how many questions don’t need to be asked because people can find the answers on their own. But cutting off that source of information doesn’t help your community.

I won’t begin to predict what communication — corporate or community — will look like in 5, 10, 20 years. But I will challenge everyone to ask themselves “does this have to be a direct message?”. The answer is usually “no.”

This post’s featured photo by Christina @ wocintechchat.com on Unsplash