A couple of weeks ago I set up four RSS feeds for the search term “journalism weblogs” using Feedster, Blogdigger, Bloglines, Google Alerts and Google News. The idea was to find out which of these sources would produce not only the most results but also the most quality and relevant results to my aggregator. Again, I know this isn’t an earth-shattering idea, but I’m thinking that at some point I may aggregate more search feeds than individual Weblogs. I haven’t collected any hard data on this, but I would guess that probably only about a third of what I collect from the edu-Webloggers that are in my Blogroll is really interesting and relevant to me. What if the search feeds I read were, say 75% relevant? Wouldn’t that make it more worth my time? I’m still working through this idea, but I plan to do more and more study and comparison with these types of feeds.
Anyway, I took a look at the posts from each of the five feeds listed above for the past week and here’s what I found:
Feedster — 75 total entries, 58 unique. All in all, the best results came from Feedster. I was actually surprised how many more hits Feedster returned, and I was also surprised how much good stuff I found there that the others ignored.
Bloglines — 38 entries, 22 unique. There were only a couple of posts that Bloglines found that Feedster didn’t. Just about all of the posts were relevant.
Google News — 31 entries, 21 unique. Google News is obviously different from the others in that it searches newspapers for its results, not Weblogs. In fact, I would say that to get the best results, and search should be done with Google News in tandem with Feedster.
Blogdigger — 11 entries, 9 unique. Ironically, just in terms of relevance, the results in Blogdigger were all great. It just seemed to miss a lot of the other good stuff out there.
Google Alert — 0 entries. Google Alert searches new sites that are added to the Google database. Nothing new this week, I guess.
So what does this mean? Not sure yet, though it would be cool to have one search that combines blogs and newspapers. Is there one out there that I’ve missed?
Hi Will,
I’ve been toying around with Feedster and Pluck. Neither of them, IMHO, were good enough. Pluck slowed all of the page loads in IE6 and Feedster misses too much of what gets posted at the sites in my blogroll. Yesterday I downloaded Mozilla Firefox 0.8 for PC along with the aggreg8 extension. It seems to be working very well, and it’s *free*. It’s a browser-side aggregator with a search function and customizable notifications. Like your entry here, for example; I could tell weblogg-ed was updated because the blog’s name has a red asterisk next to it. Your review of the other RSS feeds is helpful; it will relieve me from rushing off to experiment with any of them.
Derek…I decided to set up aggreg8 with Firefox and I really liked the way it reads and feels, although it’s a bit more effort than Bloglines. And, of course the deal killer is that I’d need to set up all my feeds on all my computers. That’s why I love Bloglines.
I understand. I’ve been flipping back and forth between browser-side and server-side feeds. I forgot to mention earlier that I was able to set up Firefox with a variety of non-blog feeds: NPR’s Fresh Air, A&L Daily, Slate, National Geographic News. I suppose it’s possible to do this with any aggregater, but I thought Feedster missed too much stuff. In a few years, all of this fuss will probably look silly; I’m sure all of it will improve tremendously in the months ahead. -DM
Hi Derek,
I wanted to let you know that I looked through all of your feeds and only a handful of yours weren’t indexed by Feedster (and those didn’t have feeds at all). It is possible, however, that when you looked at us that we weren’t indexing some of the feeds in your blogroll and that’s what caused the problem. The one or two feeds in your blogroll that weren’t in Feedster and did have feeds, I added a few minutes ago. If there’s anyway we can help you — or you continue to have problems with Feedster, please let us know.
Scott
Scott et. al.,
Mea culpa! I was unfairly deft in my dismissal of Feedster. It stands as the best server-side aggregator I’ve tried out, and although that’s not many, I shouldn’t have criticized it the way I did before taking it through a more comprehensive trial run. Knowing that Will is setting up aggregation on multiple machines, Feedster does seem to serve that purpose better than competitors. My critical note was a misstep for which I humbly apologize. -DM
Will, thanks for doing this informal study, it is very useful. I am a bloglines user and lover, and find the search feature extremely powerful. I am a complete newbie to feedster though. Based on your post, I signed up and gave it a quick whirl, and was left puzzled by one thing I couldn’t figure out how to do. In Bloglines, the power of the search feature is not only in being able to constrain the scope of the search catalogue (blogs I read vs. all blogs vs. all blogs except ones I read) but also in being able to save those searches and look at them daily in one’s aggregator as another type of feed. Can one do this with Feedster? Can one turn a search into a regularly consumable RSS feed, so that one can read it/aggregate it elsewhere? Sorry if this is a simple question that I just overlooked in my haste, but it’s that feature, the ability to regularly syndicate search results, that has for me turned bloglines into nearly a perfect search bot rather than simply a search engine with a well-constrained catalogue. Cheers, Scott Leslie.
Hey Scott…yes, you can get the RSS feed for a Feedster search and then put that into Bloglines. That’s the cool thing about all of these searches. You can do it with Google News even, which I find really helpful for my journalism students. I agree that Bloglines is a great tool, and I especially like that it can search just the feeds I’m subscribed to. But that includes any other search feeds that I bring into Bloglines which is doubly cool. I just need more time to tweak the results.
I’m sorry to hear Blogdigger find so much information, but I’m glad to hear the stuff was relevant. I think this is a function of the size of our database. All I can say is that we are getting bigger everyday, so don’t write us off yet!
Thanks!