Drinking from the fire hose
I’m a little backed up on…well, life at the moment. Everywhere I look it seems like I’m behind on something. Anyway, one place I’m perpetually behind on is catching up on my feeds in Bloglines.
I’m currently subscribed to 215 feeds. Some of those are one-a-day type feeds, things like webcomics. Others are utilities like my co.mments feed or the comments feed on my personal blog (I subscribe so I know when Akismet has let spam through or when there’s something I may want to respond to). That being said, I’m still tracking too much “stuff”.
What I’m struck by is how unhelpful the tools, like Bloglines, are at helping me deal with this cruft. For instance, I know there are dormant and uninteresting feeds in my list. The problem is, it’s difficult to know what’s dormant or uninteresting at any given point in time. What I need is some sort of view that says, “the following feeds haven’t updated in over a month” or another view that says, “the following feeds haven’t posted anything you’ve found interesting in over a month.” I bet that feature would allow me to quickly eliminate as much as 25% of my subscriptions.
Even worse are my search feeds. I have search feeds on Technorati and Yahoo! News Search. I have several searches set up with both to look for people talking about Yahoo! Mail. I use that to find out what people are saying about us. Unfortunately, both Technorati and Yahoo! News Search prove to be extremely noisy channels. I get a ton of results that have nothing to do with Yahoo! Mail, mostly because it finds blogs where people post by sending email to their blog software from Yahoo! Mail…the tagline on the message trips up the search engines. Even worse, I’ll log into Bloglines and get 20 new items…if I come back shortly afterwards, I’ll have 20 “new” items again that really turn out to be the same 20 I saw earlier. I’m not sure if that’s a problem with the feed or with Bloglines. Either way, it’s a ton of wasted time. I was recently reading Scoble’s “The ‘RSS Tap’” article. I don’t see how he manages to call his a “tap”. Mine always seems to be more like a fire hose.
The worst search feeds are, admittedly, the text searches. I also have tag searches set up in Technorati and Ice Rocket, but that requires people to tag the posts in order for them to be found. I have both text and tag searches for Yahoo! Mail in Technorati. The text search frequently turns up valid hits that the tag searches don’t. That tells me that a tag search alone isn’t enough. So I’m stuck with the text searches, which (as I mentioned) get tripped up by the Yahoo! Mail taglines that show up in many blog posts (although I bet it does wonders for the SEO of Yahoo! properties that get promoted in the Yahoo! Mail taglines). You can check out this example to see what I mean. The post has nothing to do with Yahoo! Mail, but it still turned up in my “yahoo mail” search on Technorati.
Anyway, I’m on the hunt for better tools and methodologies for handling these issues. If you have suggestions, I’m all ears. While I do enjoy web tools over desktop applications, at this point I’d be willing to entertain switching to a desktop tool if I thought it would outweigh the benefits of web applications (centralized data, available anywhere I have a browser, etc). I could always build my own tools, but that gets back to the whole “behind on everything” problem I mentioned at the beginning of the post.
January 1st, 2007 at 5:09 am
Oh well I was actually waiting for your personal feed reader. Man you told us you’d build your own!
Anyway, from what I”ve seen, the googlle reader is now a complete rip off bloglines so you may wanna try it
January 1st, 2007 at 10:56 am
Yeah, I kept starting and stopping. I haven’t had a lot of time to develop on my own, personal stuff. Things have been too crazy lately. I’m thinking I should just start with an existing aggregator and add my features. That might have a higher chance of success.
Google Reader is interesting, but yeah…the feature set is similar to Bloglines. I actually checked out Reader last night thinking, “hey, those Google guys are supposed to be smart…maybe they solved this.” Sadly…no, the features I want aren’t available on Reader.
Someone told me to hook up with the SearchFox guy at Yahoo!. I remember looking at SearchFox a while back and it felt a bit like gear grinding. But it sounds like it had something similar to what I want (although, not exactly).
January 1st, 2007 at 11:27 am
Yea, I have the same problem. Marshall Kirkpatrick “open sourced” the way he used to consume massive amounts of RSS data when he used to work for TechCrunch. I’ve tried some of his ideas and I think they’ve helped a bit, although they have encouraged me to subsribe to many more feed so I’m still overloaded Anyway, check it out:
http://marshallk.com/open-sourcing-my-techcrunch-work-flow
January 1st, 2007 at 7:35 pm
Hm, I’ve been thinking about ways to trim back my subscriptions. My reader is starting to show its limits and I’ve been thinking about adding some features to it.
Some of them are just basic stats to help me manage feeds — like dead feeds, feeds with high error rates, feeds that haven’t been updated for a while, that kind of thing.
I have a lot of photo feeds, and they’re just *different* than text feeds — and should probably be treated differently. I’m not really sure how best to do it, though, since each photo feed has a different idea of how to put “here’s the photo with a caption” into a feed item.
Anyway, your post describes some useful views that I plan to rip off shamelessly. Too bad my feed reading software is unlikely to be released. It sure makes my life nicer, though.
(It’s not, technically, a “reader”, since it’s an expanded form of the RSS-to-IMAP script I posted about a while back. I’ve been using it for over a year now, though.)
January 3rd, 2007 at 4:10 pm
As luck would have it, we just added a stats/trends page to Reader. After you’ve used it for a while, it should have the kind of data you’re looking for. This blog post has more about it:
http://googlereader.blogspot.com/2007/01/i-like-big-charts-and-i-cannot-lie.html
Mihai Parparita
Google Reader Engineer
January 4th, 2007 at 1:13 am
Mihai, that’s pretty cool. It’s close to almost exactly what I want. It does have the inactivity monitor so I can look for feeds that aren’t updating. And it has a notion of which feeds I think most frequently have good stuff. But I can’t flip that and see which feeds I tihnk least likely have good stuff. If the “Reading Trends” part of the report allowed me to see the “bottom 10″ instead of just the top 10/20/40, it would be perfect.
That being said, Reader is much better than the last time I gave it a look. I’ve imported my OPML and I’ll be doing a little bit of a “Pepsi challenge” between it and Bloglines this week.