I like it. I'm not sure I'd use it much, maybe for languages, but it could be cool for answering machine stuff.

The feedback link doesn't link to Tangler

I like the idea, although I think the barrier to participation is still somewhat high given the need for a microphone. The most interesting thing to me was feeling a much stronger connection to the global community than through text alone.

I think Skype headsets help here!

There is more of a sense of connection, although I feel as if it would get boring and possibly obnoxious at times (if people abuse the system). If it were to ever become popular it would be awfully difficult to have real conversations on it. I think it could be interesting for research however.

 For instance, lets say I'm doing a report on how people hear and understand languages they have not learned or cannot recognize. There is a topic on Chinswing about this, and I could include clips of these discussions in my presentation, video, or report. I think it could be very helpful for that sort of thing.

 Of course, this also brings about the terror of horrible audio on the net (but, of course, podcasting and youtube helps with that as well). Many of the participants on the site had very annoying pops and audio artifacts that would have been vastly improved with a higher quality mic (or probably even backing off the mic a bit). 

Interesting concept. After listening to a few discussions I certainly felt a better connection with the users, mainly because of the voice element. For some discussions, it works well having it run in the background (i.e. you don't need to be 100% focus like you do with reading text)

One thing I miss though is the ability to quickly scan a discussion. With text, you can easily do this but with voice it's a lot tougher. Plus you can only consume as fast as the person is speaking (and there's also the danger of not understanding what's being said)

I like the way you've laid out the discussion thread, simple and straightforward. One suggestion would be to auto scroll the ticker once you get to the last displayed audio message, so you can see who the next messages are from