My friend Rich pointed me at daytrotter.com, and it looks like they have a really nice collection of live recordings by a bunch of interesting artists. All the files are available for download on their site, so using some trickery, I created a txt file of all the mp3 URLs so that you can use the same old wget trick to download them all. Watch out, there’s 869 files in there, so I’d expect over 4GB of music…
I’ve grown impatient waiting for 2008.sxsw.com to release their torrent of mp3 files. I’m wondering if they’re going to do it at all.
So, I decided to just suck down the whole site and scrape out all the URLs to all the mp3 files and download them. It was very straightforward.
First, scrape the site by doing something like this:
wget -nd -nH -r –no-parent -nc http://2008.sxsw.com/music/showcases/alpha/0.html
Then, do something like:
grep mp3_download *.html
(Yes, they were silly enough to use a CSS class for all their mp3 download links named ‘mp3_download’.)
Then, you’ll have a file with a bunch of raw HTML links. Pull that into something like emacs and do some replace-regexp commands to trim it to just the URLs themselves. (There are 740 of them). I then took the resulting list of mp3s, split it into 2 files, and am running two copies of wget in parallel to suck them all down. Here’s a copy of the list of all 740 mp3 files.
Send me an e-mail to my private account if you’d like me to hook you up with a .tar.bz2 of all 740 files. I wonder if they’ll release that .torrent soon? 
UPDATE: The download completed overnight, and the resultant files are about 3.4GB.
My Microblog