Reading Lists or Grazing Lists?

Adam Green has scripted an OPML Reading List for Tech Memeorandum which is dynamically updated, hourly, with feeds for the blogs that have made the frontpage of that memetracker. This is a terrific use of OPML but I think the results are more suited to a Grazing List (glist?[1]) than a Reading List.

I think of Reading Lists as small (10 or less feeds) lists that change slowly (weekly or less frequently). But Adam's is a high volume, fast changing list. And with over 50 feeds currently I think its a little overwhelming as a Reading List. Joshua Porter seems to be thinking similarly[2] -

"The blogs got there because of some really interesting post, because they’re somehow related to the top stories of the day. In other words, the blogs themselves may or may not be interesting to me other than their one, attention-getting post. So OPML might not be the best solution at this level. So the question is: are reading lists dynamic? Or is it simply news headlines that are?"

Well I think that OPML still is the best solution but the means of accessing it needs to change. I'll continue to use BlogBridge for subscription to Reading Lists because aggregation of small, slow changing lists is still essentially subscription. But for access to large, quickly changing lists I think its more appropriate to use a Feed Grazer like Taskable or OPod, which allow you to quickly scan numerous feed items from the blogs with which you aren't familiar before deciding to subscribe (with a regular aggregator). It's a subtle difference but nevertheless an important one.

Better again, imagine if Adam's script could generate a multi-level OPML hierarchy with the feeds for the original story in the top level nodes and the referencing blogs leading off those as sub-nodes. Now that would be getting close to the "evanescent, biotic OPML hierarchies" I spoke of yesterday.

[1] I noticed that some bloggers, like Matt Terenzio for instance, are using the term glists to, apparently, refer to multimedia reading lists but I can't find the origin of the usage. I previously suggested the word Beed (Bundle of feeds) for the same purpose but whatever sticks is fine by me, as long as we have a term that doesn't limit us to thinking of dynamic OPML lists as text only.

[2] Amy Bellinger is also feeling feedinundation and thinks that reading lists should have between 4 and 6 feed. She suggests making groups of lists if you need more. And what better way to represent lists of reading lists than with OPML hierarchies?