As to why I prefer reading in an aggregator over visiting weblogs directly, I see four reasons:
1. An aggregator substitutes one discipline for many. It brings information to me when there is information to be had. I don’t need to remember to cycle through a blogroll.
2. An aggregator makes more efficient use of my time by taking over the task of polling and collecting information for me.
3. Ideas come to me in pure form and on an equal footing. Moreover they show up in a consistently readable format. My aging eyes don’t tolerate small fonts or oddly-colored backgrounds well. While there may be design asthetics I am missing, it’s a price I am willing to pay.
4. I find the juxtaposition of ideas arriving in my aggregator stimulating. It promotes serendipitous connections I would not otherwise make.
My typical practice is to scan through my aggregator in several passes. In the first pass, I quickly look for items to delete. I use a pair of bookmarklets to toggle checkboxes on or off depending on my mood and how many items are backlogged in the aggregator. If there are lots of items (> 200 say) I toggle the check boxes all on which presumes I will be deleting most items. As I scan, I click off the checkbox for items I want to come back to. Bad titles and boring leads mean an item is likely to get axed. If I miss something good, there’s usually a high probability of someone in my subscriptions list bringing it back to my attention.
In the second pass, I still tend to focus on material to eliminate based on scanning the first few sentences or paragraphs. More stuff gets deleted.
When I get down to a few dozen or so posts, I start to read more carefully. Some items I post away to categories I maintain locally strictly for my own purposes. Backup brain kinds of things.
Finally, I’m down to items I want to think about and likely comment on or use as a launching pad for my own ideas. Those might well sit in my aggregator for several days to a week, sometimes longer depending on what else I’m up to.