Discussion Platform Client

anychan is an open-source web client for online discussion platforms like imageboards.

Demos

Operation

anychan is purely a frontend application and runs in a web browser.

Since it runs in a web browser, it's subject to CORS restrictions when hosted at a 3rd-party domain, so it requires a proxy in order to function when hosted at this demo website. A default proxy is provided for demo purposes.

In some future, anychan could also be distributed as a standalone mobile/desktop application built from a web application using Electron.

Feature Requests / Bug Reports

Feature requests and bug reports can be reported in the Issues of the GitHub repo. The main repo for the project is still hosted on GitLab because GitHub is known to be a notoriously opressive ultra-liberal organization owned by Microsoft. They've banned my user account before, erasing all my repos from their website, so they can't be trusted.

Feature requests and bug reports can also be reported in Telegram.

Hosting

anychan is distributed in the form of a zip archive, the contents of which should be "statically" hosted somewhere.

It could also be hosted at the original social platform domain itself: such cases are detected automatically using a known list of domain names for each supported social platform.

When hosted at the original social platform domain, it doesn't require neither setting up the default "data source", nor setting up a proxy. When hosted at any other domain, it does require setting up the default "data source", as well as setting up a proxy in order to function.

Why

When I first discovered the "imageboards" subculture in 2015, I was fascinated by it. Then I also discovered that "imageboards" have been continuously losing popularity over the years to non-anonymous thematically-focused history-preserving services like Twitter or Reddit or Telegram "channels" or Discord "servers".

I first thought that maybe it's because the user interface of the existing imageboards is so clunky and old-fashioned. More recently I came to an understanding that it's not the actual reason, but then, in 2018, I set off to rewrite the imageboards' UI from scratch. By mid-2020, I've already finished the first working prototype, so even when the realization came, I still felt like getting a closure anyway, even if it was potentially pointless.

I also decided to maybe adopt this software for use with non-imageboard data sources — like Reddit or Telegram — in some hypothetical future. This client then could be used as a quick and simple UI solution for someone attempting to build their own experimental open-source social network backend.

Side Thoughts

While solutions like the Fediverse are definitely valuable, the history of the Internet has shown us that the main principle is "simpler and faster is better". This means that a product has much more chance of survival when it's lean, agile, flexible, not overengineered and focuses on solving one specific problem, rather than being a swiss army knife that doesn't do anything specific particularly well. Much like they pioneered it in the old days with the Unix-way which has proven itself throughout the generations.

Humanity has discovered that even making a seemingly simple messaging service like Telegram involves so much complexity under the hood — if you really want to do it perfect — that even companies having billion-dollar budgets struggle with the engineering complexity, let alone the open-source community consisting of seasoned part-timers and temporarily-enthusiastic undergraduates.

So the aim of this project is providing a basic, yet convenient, open-source "client" for using it with some abstract backend implementation written by someone else. Imageboards are just the simplest most-convenient and realistic-enough ground on which this client could be field-tested. Therefore, this project is not limited excluseively to imageboards. It could support anything that fits a vague definition of a simplistic text-focused social network.

Additional thoughts.