I haven't really gotten around to making that blogroll I was promising. But, for the hell of it, I tried cleaning up an OPML export from my aggregator to remove my private feeds and for-pay feeds—which, incidentally and unfortunately were some of my most interesting feeds.

But, that said, here's my feedroll. It's the result of some more URL-line magic:

• Take one OPML file: feedroll.opml
• Grab this XSLT: opml-to-hblogroll.xsl
• Run them both through the W3C XSLT Service to get this: feedroll.xhtml
• Got some OPML on the web? Paste your URL here and hit return:
  Note: Doesn't like OPML that's invalid XML.

If I were less lazy, I'd've included some unobtrusive JavaScript to give this thing expand/collapse handles. And, I'd also like to work on making it less of a cluttered and raw mess. So, it's not quite yet what I'd like to share as a proper blogroll.

But, for what it's worth, view source on that feedroll. I came up with a tentative "hBlogroll" microformat for it, based on XOXO and some OPML semantics. I haven't written the code for it yet, but I imagine that this should be pretty easily parsed.