Google Reader: Big, blue, chunky water-wings
Say hello to Google Reader, the newest addition to the world of web syndication feed aggregators.
It's slow as hell at the moment, but I'm guessing that's due to it just having been introduced and maybe not fully deployed throughout the Googleverse. We'll see how fast it is once they've had a chance to actually get it stabilized. On the bright side, it accepted my metric ton of feeds in an OPML export from NetNewsWire without complaint.
Seems pretty neat and whizbang so far—ooh, ahh—but it's just too clunky, chunky, and cute for me. It's got big, fat, blue toy borders around everything. And, despite having nice keyboard accellerators, it spends an awful lot of time animating the scrolling strip under the "book magnifier".
(You know, those things they sell to people who want the Large Print edition of Readers' Digest? I'll probably need one someday.)
But, then again, I never caught the Gmail buzz either. Too much AJAX, not enough IMAP. To be fair to this lens thing, though, I'm an admitted info freako outlier who'll probably be approaching 1K feeds before the year's out. That said, I suspect I'm just ahead of my time and more people will be where I am, eventually.
Anyway, what I look for in a feed reader is how well it enables speed skimming: I'm going to ignore 70-90% of what I see in feeds, so I don't want an aggregator which helps me carefully and methodically pick my way across the headlines.
No, I want something which lets me scream through feeds as fast as my eyes can move and a finger can slap the space bar. I'll queue up what I really want to read in browser tabs or some other temporary storage, but I'm not going to want to spend more than a split second on a headline.
The other thing I want is more machine intelligence. For example, for my book (which you should buy), I built a popular links feed generator which skims links out of all the entries from all my feed and ranks them by number of mentions. This is the first feed I read each day, because it's the most valuable.
From there, I read feeds produced by sites like Technorati and Feedster, followed by feeds on topics I've currently prioritized. If I have time, I eventually wade deeper into my feed list and even reach the end some days. If I had more machine intelligence helping, I might never actually have to make it through the whole list to catch what I really want to catch.
Basically, what I look for in aggregators is assistance in attenuation. I want to push as much as possible out of my front-and-center attention, let other processes deal with things, and discard with wild abandon. This is not email, where every postcard that falls through the slot could be a message addressed directly to me. I don't want to manage these pieces of information in onesies and twosies.
No, my subscriptions are the rapids in the river of news and these water-wings don't help me.