What happened to Open Community Captions Forever?
Open Community Captions Forever (OCCF) is a project we started some half a year ago. Our goal was to provide a community captions system where everyone benefits. Our method would be to allow captioning any video, anywhere, as well as the ability to "fork" captions: create a new revision based on an existing one. Let's talk about how it all went wrong, and what that means.
The first barrier we met was licensing. Not of the captions - they're not our problem - but the code we wanted to use had general licensing issues that needed to be fixed. And they did... in a development branch. This generally wouldn't be a problem, but due to the requirements for getting a WebExtension signed - a process required for publishing a firefox/chrome extension - we're not allowed to use development branches, and we failed to arrange with the developers to solve this issue.
The second barrier we met was lack of web development skills, particularly when it comes to making user interfaces. This is entirely on us, altho it does take some time to learn those skills. This is a minor issue in the grand scheme of things, but it's still an issue.
The third barrier we met is that auto-generated captions are increasingly better, in a way. Ppl have come to simply accept them, which is an issue when your project is built around that not happening. We are too late.
The truth about artificial intelligence - and auto-generated captions - is that it alienates the user to the labor relations. With OCCF, you'd know we make OCCF itself, you'd know everyone involved in making your captions, etc. With auto-generated captions, you don't. You don't even know who developed the model, who contributed training data, etc. It's labor laundering.
There are additional issues, too: Platforms aren't gonna allow auto-generated captions on music videos. Platforms are gonna censor swearing in auto-generated captions. These are real issues, and they make content inaccessible; shouldn't we be fighting back?