When I first started working on RSS Bandit I wanted an application that looked and acted as much like Microsoft Outlook as possible. Two years and over a hundred thousand downloads latter I realize that there are a number of drawbacks to using this model for reading feeds [or any information for that matter]. Mike Torres describes some of these reasons in his post Why I dig Bloglines, he writes

Part of the problem for me is that applications that look and feel like Microsoft Outlook tend to make me feel like I am working, and I am immediately in "information overload" mode (we get hundreds of pieces of email each day at Microsoft.)  Catching up with friends, reading Scripting.com, or checking out Engadget shouldn't be tedious.  But for some reason, it was.  Until I switched to Bloglines.
...
Anyway, here is what I like about Bloglines:

...

• I can scan dozens of feeds in less than a minute.  With NewsGator for Outlook and other Outlook-style interfaces, it just simply took longer.  Probably because Bloglines shows me the feed in the way it is supposed to be presented - reverse chronological order on a single page.  Not as individual messages that I have to click through. 

...

This is the bane of the current information viewing model paradigm favored by email and newsgroup readers which many RSS aggregators have decided to inherit. The major problem is that the Outlook mail reading paradigm has a fundamental assumption which turns out to be flawed. It assumes you want to read every item you get in your inbox. This flawed assumption leads to the kind of information overload that hampers the productivity of lots of people I know at work. I've met several people who seem to always have hundreds unread items in their email inbox. For this reason I always have to learn who's easier to reach via IM or swinging by their office in person than sending them mail.

Most people I know get four classes of messages in their information aggregators (I am lumping reading email, reading news and reading RSS/Atom feeds into a single category). These are

1. notifications (checkin mails, comments to my blog, etc)
2. headlines (email newsletters, feeds from news sites, etc)
3. messages sent directly to me or that is similarly relevant
4. messages sent to an interest group I am a part of (XML-DEV mailing list, comp.text.xml newsgroup, etc)

The problem is that the typical Outlook inspired information aggregator treats all of the above as being of equal relevance. Even though Outlook does provide mechanisms for managing assigning relevance to incoming messages, they are either hard to find or cumbersome to use.

This is definitely one of the areas that needs to be improved in the world of information aggregators in general and RSS/Atom readers in particular. There are a number of features that I'm working on for the next version of RSS Bandit aimed at making it easier for people to consume information from various sources in a flexible manner according to what relevance they place on the information source.