다음 10년의 인터넷을 위해 (2026년판)

It’s been more than 13 years since @codinghorror first announced Discourse, with this claim:

Today we announce the launch of Discourse, a next-generation, 100% open source discussion platform built for the next decade of the Internet.

If we assume that this is an evergreen statement, and we’re continuing to build a discussion platform for the next decade of the internet, what do you imagine that looking like?

Consider what the world looked like in 2013 and how much changed between then and 2023[1]. Which of those changes were we really anticipating when starting this project in 2013 and which were complete surprises or came much earlier than expected?

Looking ahead 10 years, what kinds of things can we imagine being true that are not true today? How much should we “expect the unexpected”? How do you expect to be interacting with other people online 10 years from now?

What role to you hope a project like Discourse to be playing in the world of online communication 10 years from now? What are you most worried about happening to Discourse that we might be able to avoid if we work together to shape our own destiny?


  1. Not sure what to reference here to help ground us all, but here’s one link maybe worth skimming through: Timeline of social media - Wikipedia ↩︎

13개의 좋아요

I expect to be talking to people in 2036 the same way I was talking to people in 2013, which is the same way I was talking to people in 2003: typing words into boxes.

That’s the common lineage through Usenet, BBS, mailing lists, IRC, ICQ, AIM, vBulletin, LiveJournal, SMS, MySpace, blogging, Digg, comment sections, Facebook, Twitter, Reddit, Discourse, Mastodon, Slack, Discord, Threads, Bluesky… even media-first platforms like YouTube, Instagram, Vine, Snapchat, TikTok… below the video there are billions of people typing words into boxes.

Plenty of changes were already in motion in 2013, driven by social media, and Discourse picked up a lot of them (infinite scrolling, touch and mobile layouts, live updating). Most of these were things we were seeing in social media and apps that forums simply weren’t adapting because at the time it was hard to retrofit them onto an application that was assuming a single page render was all you needed.

JavaScript pushed the web in a direction where a lot more is possible, despite the early naysayers… client-side processing did catch up, and most people now carry enough power in their pocket to run things that couldn’t work earlier.

Beyond the technology, what was happening then is still happening now and likely getting worse; there’s an effort to consolidate and get as many people as possible into one place as long as possible. This way, between typing words into boxes, your attention can be sold to companies that want to pitch you products and services to buy.

The danger for Discourse and many smaller platforms that don’t keep up with the billions of dollars poured into companies that centralize and resell this attention, is that all the air is sucked out of the room and no one has the money or the time to keep maintaining them as alternatives.

As we’ve seen, there’s a new player that’s doing a very good job at it… AI chatbots seek to shortcut the web entirely. Not only are they capturing attention and consolidating information, they’re also trying to replace much of the reason people were showing up to social media to begin with by seeming human enough to satisfy the need for connection. Who wants to visit a feed where people constantly argue with each other when instead you can talk to something human-enough that is incentivized to constantly tell you that “you’re absolutely right.”

So what’s anyone to do in this environment? The hardest thing is going to be giving people reasons to show up and type their words at all. Without that it doesn’t matter if you sell attention, ad space, information, ask for donations… worse than not having the ability to keep the lights on, there’s no one left to keep them on for.

What I hope Discourse can continue to provide is a safe harbor where you’re not subject to the massive unknowable algorithms that decide who you read and when. Your every action doesn’t have to be scrutinized to determine which brand you most identify with, what your next purchase is going to be. You’re not just sharing information for a chatbot to parrot back to someone you’ll never meet. You can run your own site where people type words into boxes and make it do as much or as little as you’d like… where the people who show up value the exchange enough to help each other keep the lights on.

18개의 좋아요

One way to look at those trends is that human-to-human connection is getting harder, because there are forces trying to monetise and control it.

So the challenge for Discourse is to maintain and promote that human-to-human connection.

Discourse as a way to answer questions, probably not so much.

Another trend is polarisation, and ever more intrusive bad news from the world around us.

I think there’s a role for providing safe spaces where people can enjoy the things they enjoy, without harrassment, without argumentation, without being reminded of horrible scary things. That’s something I want to see.

There’s another view, which is that Discourse could have a role in bringing people together across lines of division, could facilitate peacemaking and de-escalation, could facilitate healthy developments in the real world, like improvements in organising and in governance.

I think those two roles are separate and might even pull in different directions. But moderators of different communities need the tools to shape things as they want them.

Oh, there’s another trend, which is increasing legal constraints or regulations, which make running independent communication platforms more difficult. (I’m personally most affected by the UK’s Online Safety Act, but the trends are the same elsewhere, I think.)

8개의 좋아요

This post is great; I appreciate that you’re opening up this kind of dialogue and enriching it with such insightful responses.

I discovered the internet in the ’90s, and although I’m young (!), I consider myself something of a veteran of the web. I was familiar with its original philosophy, and as much as it pains me, I see its current decline — which aligns with the named points raised in this discussion.

Given how things have unfolded in recent years and with some understanding of the underlying issues surrounding centralization and control over data and information, I can envision two possible scenarios:

A) Discourse with much more integrated multimedia features, a hub format with interconnected communities that go beyond Discourse ID — which I think is an excellent initiative, if it weren’t for the problems one faces into when trying to link to other platforms (SSO, OpenID, SAML, or whatever) authentication systems.

I envision Discourse retaining the purpose and status it has today, as a tool used by both individuals and large corporations that would continue to value spaces for exchange in discussion forums.

B) Discourse that is exclusively part of the hard resistance, cypherpunk by obligation and maintained directly by the community, more decentralized, and featuring peer-to-peer (P2P) connections that ensure its continuity.

As is the case with private chats and forums that evade search algorithms, web crawlers, and the exploitation of human-generated data and content for the benefit of the handful of institutions that receive investments from debt-financed funds, a cycle that seems never-ending.

This may sound sad at first, but it’s actually quite the opposite. Books with purpose and meaning endure for thousands of years. Just like the TCP/IP protocol, P2P file sharing, and the need for human dialogue Both scenarios maintain Discourse as a space of resistance, except that one is transparent/public relatively consistent with its current state, and the other is hidden/private, hardened.

6개의 좋아요

Since moving to Discourse, it has been incredible to see the continual improvement and development that goes on. The platform just keeps growing, it really is quite unique these days.

Recent additions to integrate automation, AI, workflows etc as an example.

Continually being able to adapt to changing requirements is the only way to stay ahead.

While communities have commonality, there are also always differences. Different subjects attract different personalities and that does influence the choices of admins.

For me, the challenges are varied. One is being able to keep the site secure and keep ahead of the bad actors (spam, bots etc). For this I would like to see better tools, visibility and controls.

Humans being humans, there is a small but growing group of people who social media has been successfully grooming to be a tad more entitled and unfiltered than would be considered appropriate when not lurking behind a keyboard.

It isn’t always easy to find those who can’t play nice with others without it being flagged first. Sentiment analysis is still a bit flawed.

Discourse was (my impression at least) built around a philosophy of how interactions were supposed to happen on the platform. This doesn’t always align with some communities. For example, to try reduce the transactional members - those that make zero effort to read, search, understand what the community is about and like to fire and forget - putting “gates” (don’t like that term) in place, or rather, putting newbies through a positive, curated onboarding process, so they actually read guidelines, a brief tutorial to understand what goes where etc - isn’t easily achieved.

A higher resolution trust model with more controls.

Being able to easily, on a per-user (not group) basis, give a timeout (temporary or permanent) for a topic, category or multiples thereof.

It is the fringe users that make the most noise and consume the most time from small teams. Helping manage that group, providing tools to encourage the right types of engagement doesn’t just improve the tone of the site, it helps prevent moderation fatigue.

For the most part, I believe Discourse is already well positioned for the next decade, but at the current pace of change it is a moving target.

4개의 좋아요

하, 내가 이 토론을 시작해 놓고 몇 주간의 휴가를 떠나버렸네요. 하지만 이제 돌아왔고, 여러분과 이 주제를 계속 탐구해 보고 싶습니다.

컴퓨터가 네트워크로 연결되기 시작한 그 시절로 거슬러 올라가 봐도 변하지 않은 것이 무엇인지 깨닫게 해주는 유용한 상기라고 생각합니다.

인터넷이 새로웠을 때를 기억합니다. 당시 저녁 뉴스의 토크쇼 진행자들은 사람들이 인터넷이 무엇인지 이해할 수 있도록 돕고 있었죠. '서핑’과 전자상거래, 그리고 이런 저런 것들 말이죠. 사람들이 아마존에 대해 가장 걱정했던 것이 지역 서점들에 미칠 영향이었을 때였죠.

네, 저는 고퍼(Gopher), 유즈넷, AIM, ICQ 등에서 다른 사용자에게 메시지가 전달되도록 박스에 타이핑을 했지만, 동시에 삐- 하고 신호음이 울리기 전까지 전화선을 사용할 차례를 기다리며 연결을 해야만 했습니다. 그 후에야 메시지를 읽거나 박스에 타이핑을 할 수 있었죠.

지금의 경험은 분명히 다릅니다.

거기서 10년 이상을 되돌아갔지만, 딱 그만큼만 거슬러 올라봐도 같은 느낌을 받습니다. 상황은 달라졌으니까요. 그리고 네…

…이 AI라는 것이 10년 뒤에는 상황을 훨씬 더 다르게 만들 거라고 느끼는 또 다른 변곡점(인플렉션 포인트)인 것 같습니다.

그래서 현재 제 머릿속에서 이 주제에 대해 맴도는 것들은 그런 것들입니다.

AI가 온라인에서 인간 간의 문명화된 토론을 장려한다는 우리가 상상하는 것에 분명히 위협이 되지만, 동시에 우리가 그 목적을 위해 만들고 있는 도구들을 형성하는 데 더 많은 도구를 제공한다는 측면에서는 기회이기도 합니다.

따라서 우리가 Discourse를 어떤 것으로 만들고 싶을지 고려할 때, 1년 전보다 훨씬 더 많은 것이 실현 가능해 보입니다.

흥미로운 시대가 오고 있습니다.

1개의 좋아요

잠을 잘 수 없어 이 스레드를 처음부터 끝까지 읽게 되었는데, 전적으로 동의합니다. 제가 크게 기여할 수는 없을 것 같지만, #community-building:praise에서 말했듯이, Discourse와 Fediverse의 동료들은 사용자의 주의를 알고리즘을 개선하고 사람들을 가두기 위한 지표로 삼지 않는 ‘소셜 네트워크’ 대안으로 자리 잡았습니다.

10년 뒤에도 지금처럼 소통할 수 있기를 바라며, 가능하면 Discourse나 Fediverse의 어떤 플랫폼을 사용했으면 합니다. 저는 인스턴스를 큰 애정으로 운영 중이며, 브라질에서 Discourse를 CMS로 채택한 포럼들을 좋아합니다. 이 프로젝트가 존재해 준 데 대해 진심으로 감사드립니다. 왜냐하면 저는 지금 제가 무엇을 소비할지 스스로 선택할 수 있으며, 그 많은 부분이 여전히 인간적이고 문명 있는 교류의 가치를 보는 사람들에게서 비롯된다고 확신할 수 있기 때문입니다.

1개의 좋아요