Cleaning out my RSS Closet

I've been in the process of cleaning up my feeds lately. (And it seems I'm not the only one).

First, I killed almost all the feeds that do not provide full content. The only survivors here were feeds that either provide unique content I can't get elsewhere or that I truly consider 'must-reads'. Needless to say, there were not many survivors.

I've also unsubscribed from most aggregated feeds. I realized that, generally speaking, I was already subscribed to most of the bloggers who I was interested in already. The aggregated feeds added little value, as most were just duplicate posts I've already read, and any others got lost in the mix. I would love to see my aggregator of choice automatically filter out duplicate posts based on the GUID, but I wonder if that might not add too much computational overhead.

I figure that if someone new is blogging that I should be reading, I'll hear about them through one of the many other channels I read.

The aggregator of choice I mentioned above is NewsGator. I was previously using intraVnews, but I was having some issues with it, so I eventually switched to NewsGator when I realized this model worked for me. Unlike Scoble, I'm not doing this for offline access - rather, I'm using this to pull it into my Exchange server.  This allows me to have my news (like my mail) synchronized across my various machines and allows me to keep up through Outlook Web Access.

Of course, there are two main drawbacks to this: 1) I don't get new RSS feeds unless Outlook is running on my primary machine and 2) I can't subscribe to new feeds when I'm not on my primary machine (ie through OWA). Ultimately, I hope to replace this with a server-based solution that I'll have time to work on this summer.

This also means that I can make extensive use of the search folders. I can make folders for legal reading, for all unread tech blogs, and for truly news feeds (like Reuters, CNET, etc). I can even make various 'channels' based on content keyword search - sure, it's not perfect, but it's a step closer to organizing individual posts into categories. So, for example, I have one search folder that shows all posts that mention 'Longhorn' or 'Yukon' or 'Whidbey'.

All in all, it means I actually get a lot more out of the feeds I'm subscribed to, instead of just subscribing and falling behind in my reading.

Engineering