Comment by btown | Hacker News Reader

Comment by btown | Hacker News Reader

From the article:

> The central thesis is that what these villages can’t tolerate is a sustained large influx of strangers [...] A slow trickle of strangers is tolerable, a brief large influx is fine; the strangers’ average interaction is eventually stabilizes and biases toward the a stable group of members, and they quickly find shared values and become villagers too.

The tragedy of Reddit is that I believe this thesis is fundamentally wrong. Reddit has shown that villages can tolerate a sustained large influx of strangers if they are given the tools to scale the onboarding process - ensuring people understand those shared values, and ensuring that bad-faith actors are identified and prevented from destroying the fabric of the overall community.

These tools exist - they are the ecosystem of API-based moderator tools! People have been iterating on them for years to serve the needs of the moderators of Reddit's diverse communities. And these are the very tools that Reddit is destroying without providing sufficient replacements, without providing time to adapt, and with a remarkably sardonic and vitriolic corporate communication strategy.

I'm a member of multiple subreddits that are shutting down indefinitely because the thinly-stretched moderation staff is constantly under attack - by everything from karma bot farmers to politically motivated aggressors (including would-be infiltrators applying for moderation positions). The moderators' ability to use third-party apps was the only thing keeping the effective DDoS at bay.

And I truly don't think that those moderators would ever say that the number of strangers was the problem, because the tools did scale, and they were able to provide supportive and positive communities to thousands of those "strangers" who would never have had access to those communities before.

I always saw Reddit as a model of how communities could scale beyond physical constraints - and how some of the learnings of scalable community governance could perhaps be ported to real-world scenarios. Now, people will read posts like OP's and simply think that this was always impossible, and that's just sad.

The article stated that explicitly as its central thesis, yet I came away with a different insight.

> In this context, the defining trait of a village is that it’s group of people where the average interaction over time is with people you’ve seen before.

What I find especially insightful in this description is that it applies to lots of communities at all scales, both online and off, as diverse as a childhood friend group, a technical project team, a corporation, or an alliance of world superpowers. What defines the village is that the members you interact with are mostly the ones you have been interacting with. Change happens, but it is gradual.

What destroys the village is whatever upsets that defining trait. In the article's telling, a sustained influx of strangers is one way to kill it; there are others (for example, a large set of village members exiting, or the village splitting in two would also work.) And as you rightly point out, a way to effectively manage that change can allow the village to survive, be it tools to manage onboarding, or gradual acceptance (ala SO's points-based permissions), or a well-led corporate merger where changes are introduced gradually and with the buy-in of all participants.

I think the thesis is wrong. It’s not that the average interaction is with familiar individuals. I’ve been a mod, lurker and commenter on a lot of subreddits. And on each of those, there are maybe 10 users I recognize and whose comments I see occasionally.

What is the case, is that on short timescales, the average interaction is consistent with what I expect from the community. The anonymity of Reddit has show that the specific face or individual doesn’t matter. It has a ‘friend’ or ‘follow’ option that I’ve never used and really never felt like I needed to use it.

Reddit actually became less familiar to me when I saw users tag other users in a post to share them with other users. The same with user specific subreddits. It’s when Reddit moved from content centric to user centric.

Interesting. I had a superficially similar thought, which is that subreddits look like a path to keeping "villages" while letting the platform grow. That would tend to entail multiple "villages" for the fast-growing interest groups, which is weird. I don't know how well that could work/be worked around.

Many groups and identities (e.g. geographical areas, marginalized groups, and various large fandoms) each have a variety of subreddits (and corresponding Discord servers) with their own unique roles: memes & humor, meta memes & humor that elevate obscure in-references to a type of absurdist canon, serious support groups for various situations, places to gather and coordinate social events and remote gaming sessions, places for non-members of the group to ask questions and be answered seriously, etc.

Each of these "villages" have different subcultures and distinct moderation needs, especially to the extent they are targeted in different ways by bad-faith actors and repost bots. And it's good that they have different needs! No single moderation team should need to maintain all those subcultures with one single set of policies, nor do they have the same on-ramps and levels of traffic. Together they form a web of meta-communities. Web platforms can easily allow overlapping spheres to coexist; the dimensionality of the platform, so to speak, is practically infinite.

But part-and-parcel with this is that the number of potential moderators in any group is spread across those sub-groups. If Reddit wanted to continue to grow, it would make sure the barrier to entry to maintain these communities was low. It is doing quite the opposite at this precise moment.

Yeah I agree with you. The thing that makes a village a village isn't (necessarily) seeing the same people over and over, it's seeing the same types of content over and over (which is something "seeing the same people" tends to do, so I can see how OP got them conflated)

To me, on the internet, I mostly recognize "people [I've] seen before" via the username. Most of the good internet villages I venture to, a username is secondary to me. I care about the content first, which can come from lots of people, and the users/faces/people second, which are mostly just the extremes of my (dis)favor.

I think mod tools are huge. I also think voting on content is big, though. Finding a village that aligns with my value of content is more important than recognizing specific people online.

Reply on news.ycombinator.com