However, I couldn’t immediately make use of the new domain because Discuit is more than just a regular website; it’s also a Progressive Web App (PWA), that can be installed, almost as a regular app, on user devices. And migrating a PWA to a new domain is a tricky thing—for starters, it’s not too difficult to completely wreck all the existing PWA installations.

In fact, since PWAs are in part identified by the domain name that they point to, this cannot technically be done. The only way to make it sort-of work is to:

1. Keep the old domain working as it is for the existing users (that is, for the existing PWA installations and others who’ve visited Discuit before), without redirecting them to the new domain when they visit the old one.

2. Redirect new users to the new domain when they visit the old domain.

Simple as that may sound, getting all the technical details correct was still not altogether trivial. Now, finally, all the necessary changes are made and Discuit is officially from now on at discuit.org.

As I mentioned, if you’re an existing user (or if you’ve only just visited Discuit before) the discuit.net domain will continue to work for you. But I would still encourage going over to discuit.org and accessing Discuit from there in the future.

If you’re an existing user and you’d like to be automatically redirected to discuit.org, then simply clearing the browser cookies for the old domain, discuit.net, will do. Before you do that, however, make sure that you’ve saved your password, otherwise you might get locked out of your account (and it’s a good idea to add an email address to your account, so that in case you lose your password, you can reset it).

Patreon link: www.patreon.com/discuit

All "Supporter" plus Patreon members will receive a "Supporter" badge on their profile. Thanks a lot in advance to all donors.

On monetization: Discuit will be a non-profit going forward

Back when I launched this site, I knew that the path that previous generations of social platforms took—being funded by venture capitalists who expect exponential growth and huge returns on their investments—is invariably going to lead to the same outcome, if any new platform tries the same approach, no matter how good the intentions of the founders are. This realization, as I explained in my first blog post, left me with the conclusion that, for a new social platform to not end up where Reddit and Twitter and Facebook are today, it has to be set up either as a private company (non-VC backed) or a non-profit. It was unclear at the time to me which was the better option for Discuit.

Having thought about this in depth, I made the decision a while ago that Discuit is going to be a non-profit, modeled after organizations like Wikipedia and Signal, funded entirely on user donations. This means, most importantly, that there won't ever be any ads on Discuit, nor any user data will ever be sold to any third-parties. Having no investors—nor any owners, for that matter—will ensure that with growth Discuit will not, in the end, turn out just like other platforms. We're already open-source, and I believe being also a non-profit would make this platform truly a community project.

(I'm looking into how to register Discuit as a 501(c) organization. Hopefully, the fact that I'm not based in the US wouldn't be a barrier. If any of you have any experience with the legal side of these things, I'd love to talk to you about it.)

Another important upside of not being a VC backed company is that there's no pressure to grow as fast as possible or else go bust. Discuit will be here, at this domain, as long as I'm alive (and hopefully even thereafter). And also, generally speaking, it seems to me that certain non-profits are more long-lasting than for-profit corporations. I'll bet that Wikipedia will still be here decades after Facebook is no more.

Subscribe now

If this is the first time you're hearing about Discuit, we're a community platform—an alternative to Reddit—that launched a few months back during the API controversy of Reddit. We've been growing slowly ever since, and I'm happy to say that we're now home to a small but dedicated and awesome user base.

If you're tired of the repeated anti-consumer behavior of the major social platforms, and hate how Meta, Alphabet, Reddit, etc, all want to make you addicted to scrolling and suck up all your data, then come and join us and be a part of an attempt to create an alternative (and fun) space, free from all of that. Let us together fight the attention economy and the destruction of the internet commons!

Technology

The backend of Discuit is built in Go and the frontend is a React application; MariaDB is used as the main datastore, and Redis is used for transient data (sessions, caching, etc). I chose to build most things from the ground up without relying on a ton of external libraries and frameworks—well, aside from React—to keep things fun and interesting (just ignore the occasional partially built wheel, please).

If you'd rather jump in and start playing around, just clone the repository and follow the instructions in the README to set up a development environment on your local computer.

License

The Discuit codebase is licensed under AGPLv3. This is an important choice because this ensures that the code that runs Discuit will always remain free and open-source. And people can contribute with the guarantee that their work won't be made private in the future. And if anyone else launches a site using this codebase, the license requires that they make their additions and changes open-source as well.

I've been building a Reddit-like platform as a side-project for the last two years, which I initially began as a service to run in my home country, Sri Lanka. Now, with Reddit going dark and seeing people looking for an alternative to migrate to, I thought of repuporsing all that work I've already done to create a general alternative to Reddit.

It isn't just that Reddit makes inane management decisions in how they treat their users, but as a software platform, they've been stagnant for many years. When was the last time they built a feature that actually helped their users? And their software has been getting buggier and slower every year.

And there's a lot I think could be done to make online discussions (and communities) better, partly by building great software, partly by good governance. I go into my vision for the future of online discussion platforms in a bit more detail towards the end of this post.

This is not a federated platform; it has no crypto nonsense; it's a good ol' centralized, Web 2.0 site. Nothing more.

Link to the platform: https://discuit.net

Why not federation

Federated social platforms, I don't think, will ever become mainstream. They may work in certain niches or perhaps among a subset of tech savvy people, but I'm highly skeptical if regular people will ever migrate to them.

Without arguing for my position in detail, I will simply state why I'm skeptical of federated social platforms ever going mainstream:

• Federated platforms are too confusing for normal users. I could simply go to twitter.com and create a new account to join the platform. With a federated platform, I have to first know what federation is, then choose an instance (perhaps among hundreds of viable instances). This process is too complex and confusing for most people. Even for those who understand all this, this is likely to induce decision fatigue. One could, of course, have a flagship instance and promote that, but in that case are you not compromising the ideals of federation?

• It is likely that a few instances will become bigger than all the rest combined, which would, again, compromise the ideals of federation.

• Federated platforms are too difficult and complex from a technology standpoint. And therefore, they are bound to be slower and buggier than their centralized alternatives.

• The whole of the federated network will not be accessible to any one user (from one account), because no instance will federate with all the others (even if one instance wants to).

• The issue of trust: One has to trust the admin of one's instance (and to a lesser degree the admins of other instances one interacts with). Contrary to what some might think, people trust companies and organizations more than random strangers on the internet. Security and trust is one of the big upsides of walled gardens.

Some of you may disagree with my take on this, and that's fine. This project, then, may not be for you. There are a lot of people who believe in the federation model, and some are actively working on realizing those ends. And I totally get where they're coming from and I wish them the best of luck.

Content moderation

There have been previous attempts at making alternatives to Reddit, and most such platforms that managed to gain users nevertheless failed, in the end, because of a lack of moderation. Those platforms that were released as free-speech alternatives to Reddit were soon overrun with people with extremist political views, conspiracy theorists, trolls, and the like, rendering those platforms completely unappealing to everyone else. Freedom of speech, as a principle, is absolutely crucial for a functioning democracy, but for an online discussion platform, it's just not going to work, especially not in the beginning.

So, with regards to moderation, these are the rules I'm proposing:

• No adult content.

• No racism or any form of bigotry.

• A soft rule of: Don't be an asshole.

In the enforcement of these rules, there will certainly be a degree of subjectivity (what is considered adult content or when is someone being an asshole in a discussion, for instance?). And occasionally there will be cases where the moderators are in the wrong, no matter how careful they are. But perfection is not possible. The best we can hope is that such errors will be small in number.

These rules are not set in stone; they will change, when (if) the site gets big enough so that it can't be overrun by a small group of fringe users.

Monetization

Frankly, I don't know what the best way of monetizing a platform like this is. I can see only two viable options:

1. Donations: Wikipedia seems to work fine this way.

2. Advertising + subscriptions: Which is how Reddit is monetized today.

The first option is perfectly fine for small(ish) communities, granted that people chip in, but I don't think it scales well. Wikipedia works, but it's mostly static content and it's barely growing compared to social platforms. Could Reddit, as it is today, be funded this way? I rather doubt it.

When I think about advertising, it isn't advertising per se that I dislike. It's that on most platforms the ads are intrusive and there are too many of them. And the drive for profit results in ever increasing dark patterns on the platform to capture as much attention as possible from the users. The real evil here seems to be taking enormous amounts of VC money, on which the company has to make a return on, and therefore, sooner or later, will come a time when every possible dollar have to be extracted from the users. But does it have to be this way? Is it not possible to run a moderate amount of ads and still be profitable without making the user experience horrendous? Craigslist is probably a good example here.

As I said, I am undecided on how best to monetize. The only thing I can say with certainty is that I won't be taking any investments from anyone whose values don't align with what I'm trying to build. Enshittification would be avoided at all costs.

Not hell-bent on growth

Some of you may have seen the documentary The Social Dilemma (2015) or otherwise have heard critiques of social media; of how these platforms are making us addicts, shortening our attentions spans, driving political polarization, and so on. I am concerned about all these problems and thus have no desire to contribute to them.

As such, there will be no dark patterns on the platform. No requiring registration simply to view the content; no popups asking users to sign up every time they land on a page; no making the mobile browser experience horrible to drive traffic to the app.

(I’m afraid, the front-page of Discuit is currently infinitely scrolled, but soon users will have the option to disable it.)

A vision of the future

Social media in general, and discussion platforms more specifically, I believe, could be way better than they are now. My vision of well-designed social media is about giving agency to users. That is, giving them more options and more control over their social media experience (with sane defaults, of course).

Imagine that you could:

• Enable/disable things like infinite scrolling, suggested content, etc.

• Filter out all memes/funny/low-effort content (or filter out everything except these).

• Tweak the recommendation algorithm to your liking.

• View top posts within a certain geographical area.

• Pay a small monthly fee to get rid of all ads.

• Have multiple feeds (like Twitter Lists) with different configurations.

• In general, configure just about everything (especially the feeds) to your liking.

When it comes to discussion platforms, specifically, there are many things that could make them better. Such as:

• Moderation logs for building trust.

• Option to have to complete a questionnaire to join a community (set per community).

• Option to require a certain amount of karma before voting (set per community).

• Ability for long-term members of a community to remove power abusing moderators by a popular vote.

• Better moderation tools (group chats for internal communication, bulk editing, etc).

• Option to have invite-only communities.

• Option to have image replies (set per community).

• Limiting the number of communities one person can moderate.

• Tracking user karma by community.

• Option for more experienced users to help with moderation using a flagging system.

• Option to filter insightful comments from the puns and vice-versa.

As of now, Discuit only has the most basic features that it needs to function. None of the aforementioned features are yet built, nor are any of them fully thought out. If we gain enough users to sustain a community, at least some of these features will definitely be built in the months and years to come.

If all this sounds good to you, please come and join us. At this stage, this is really just an experiment. I have no idea whether enough users will sign up and come back to get the platform off the ground.

If you want to keep up with this project, you can subscribe to this newsletter (or you can follow Discuit on Twitter or join the Discord server).