Earlier this year, iPhone developers Marco Arment and David Smith sat down to discuss how they efficiently manage their apps as sole proprietors. Arment makes a podcasting app and Smith makes an app for customizing your homescreen with widgets. Both are successful businesses.

One thing that made me a little sad was that both men said they found it important to avoid facilitating direct communication between human beings.

Here's Arment, for example:

I intentionally have not created any way in Overcast for users to write text or post images that would be visible to other users. Because then you have user to user communication, then you have things like harassment, illegal content being posted, takedowns, all sorts of stuff you have to deal with.

And so, you know, and you look at the problem set of a podcast app, it's like I can do 95% of what people want without incurring some of those giant burdens that would require a staff to moderate and things over time.

Smith similarly called it "a massive sort of obligation" to enable features like allowing users to, for example, send pictures into one another's widgets. "So I just don't go near those," he said.

On the one hand, I completely understand where both developers are coming from and can't fault their choice. I've actually been the editor sending a reporter to ask a tech company why they allowed bad behaviors to fester on their platform for so long (if only because others were leveling this criticism). I've become very familiar with how far some powerful people expect platforms to go to police their content, and why that expectation exists.

But I'm also old enough to remember when the internet was a more free and hopeful place, and when connecting people directly — strangers very much included — was considered one of its great superpowers. We've spent the last decade or so becoming intimately familiar with what can go wrong when you connect people online and how to mitigate the risks inherent in doing that.

There are also costs on the other side, though, in creating expectations for how platforms should police communication and in focusing a good chunk of interface design on stopping the worst actors. If we designed parks and cafes and sidewalks to thwart the most malevolent people who might use them, they would be a lot less enjoyable for the rest of us, and much of the web increasingly feels nerfed in this way.

I hold out some hope that the pendulum of conventional tech wisdom will swing the other way soon, and that we can collectively see the value in online spaces that bring people together and find some ways to make it easier for people like Marco and David to build and manage those spaces in a way that feels safe for everyone.

Because I think we've only begun to scratch the surface of how the internet can bring people together to accomplish great things.