I've had an idea brewing for a while, and I've decided it's the kind of idea that I'd rather get semi-public feedback on than try to perfect on my own, even at "risk" of having it stolen. Frankly, I'd rather buy it anyway.
I call the idea: Herddit.
As an avid user of reddit, and regular commuter who uses podcasts to ease the commute, I've always wished there was a way to bridge these two loves.
Reddit's power comes from the decentralization of its content, and podcasts are popular (at least with me) because it's the most action-packed form of content I can currently receive through my ears.
Why not create an audio forum, whose structure is modeled after reddit's comment system, and is designed for people whose eyes are occupied? It could even be an ideal application for the blind to participate in online forums.
I'm not talking about a text-to-speech or voice recognition layer for Reddit, although someday that may be practical. For now I'm talking about having an interface that allows people to post their own voice as audio clips to reddit, and have a simple interface to navigate during their commute. Ideally, the interface would also be compatible with any "remote" hardware, from car stereos to headphones.
Here's a screenshot of the current version:
The basic usage of this app would be a lot like listening to any digital device: You listen to a station (or "Sub-Herddit") that you like, and whenever you start hearing something you aren't fond of, you hit the skip key.
You might notice there are two kinds of skips: The small skip would skip a single branch of comments, and the big skip would skip a whole topic of conversation. This is akin to browsing reddit, and collapsing a comment that doesn't interest you, so you can continue reading the relevant discussion, or just skipping the topic if it doesn't appeal to you altogether.
Upvote and down-vote would work as on reddit proper. A record button would allow the user to record a response to the currently playing comment, and a list button allows a more visual representation of the discussion, for users who have the luxury of using their eyes.
As far as infrastructure goes, I currently have the application building on top of the regular reddit, but adding a prefix of "HRD" before any subreddit, so that these subreddits are independent and have no overlap with the regularly trafficked reddits. The current version does not support the creation of new sub-reddits, although it does feature the ability to follow newly discovered ones.
As far as hosting all these audio files, the best solution I've heard so far came from Sterling Cooley, who suggested using SoundCloud. Being the YouTube of audio, it seems the natural choice, they even have an API for doing all kinds of audio interaction. Right now, this is also my technical hang-up, because streaming audio is a new field for me. I already have the reddit interactions working, but as far as playing soundCloud links goes, I'm hoping to give some serious time this winter break.
Anyways, I'm curious what you all think, any feedback or ideas are certainly welcome. I'm especially interested in hearing people's thoughts on open source software. I haven't done anything open source before, and it feels like a real leap of faith, that I haven't been ready to make yet, to throw your work to the wind, and hope something good comes of it. As someone who is not working as a software developer, I don't know whether it would be better for my job prospects to open source this and work as a team, or to churn it out myself.
With your help, we may be able to continue the conversation even when our eyes are otherwise occupied.
Thanks for reading,
-Dan Finlay