A few weeks ago, I read in Monkey X about the lack of emergent prioritisation for blog entries: "There's no way for me to know what's worth reading and what's not. A combination of the author's own prioritisation (like email) plus a network indicator of importance would help me filter out the important items when in a rush or with an excessive backlog." True enough. With my GoogleGuy Says blog, I have taken a crack at the former. Each entry is assigned a High, Medium, or Low importance level. I have also created filtered views of the blog -- you can view all entries, or just High, Medium, or Low importance entries. Last, but not least, I created RSS feeds for each importance level. This enables the reader to subscribe to only the high priority/importance entries.
It seems to me that using Technorati or Google's API you should be able to determine the popularity of each of your blog posts, and list the top X of them.
Btw, I saw your post on GGSays earlier, and I like the priority issue. I've been reading that blog for a bit, and am somewhat annoyed when his comment is what he had for breakfast or something similar.
I've been doing an 'editor's picks' at the top of J-Log for a while. I'm also playing around with a 'blog customization' thingy My J-Log that will let users help decide how the content is provided to them.
Prioritisation is based on some sort of indication of a *value* judgement. I've um-d and ah-d a bit about this a bit on my blog in a posting called Web Logs: Dynamics and Value, but I'm sure that there's a lot more to it than just this. I guess the easiest way to get a quick and dirty view is to have some form of recommender system (Rate this page), but it would be great if referrer logs could be exploited to create some sort of Amazon-style: "people who liked this blog also liked..." in keeping with blogging's network roots.
Mark, great post on "Web Logs: Dynamics and Value". There are a lot of interesting questions surrounding value, trust, reputation, linking, and discussion...