So what if the unread counts go up? I have a feed with 543 unread items, and many others with 100+. Sometimes I scan the titles of the latest dozens of items, then mark all as read.

For me, that's the advantage of RSS over Twitter: your unread counts are separate for each feed, so that blog that rarely updates but is always worthwhile to read doesn't get lost in the pile of cheap content.

Lol and I have a feed with 3000 unread items, actually several. I don't pay attention to unread counts. Well actually I do, they're there for reference, but that's it. I do the same thing, I might open 50-100 interesting-looking links and then mark the rest as read.

I find the key to RSS is that you need to subscribe to actual content, rather than just a feed. A lot of the stuff on the internet is, frankly, just a barely curated feed of information pushed out a firehose. Unless you feel like you need to, don't subscribe to feeds from news organizations unless you're willing to skim and blast away lots of unread stuff. Most of that you simply won't read.

With RSS, I like subscribing to journals and organizations that publish substantial content (lots of text rather than ads) at steady but not explosive rates. RSS is pure consumption, so if you're using it to "follow a feed" I think you're doing it wrong. Though YMMV, of course.

I've gotten into the habit of organizing my feeds into two groups. People who post infrequently with high quality articles get put into one group. Sites that generate loads of articles (NPR, Guardian, HN) go into the other.

I generally read everything in group one. In group two, I start the day by just marking the entire group read. If something neat comes into group two during the day I might read it when I have a spare moment. Otherwise, it gets flushed.

I am constantly producing tabs on my laptop, and every now and then saving the session somewhere so I can get some fresh air after ~200 open tabs. But the real problem is on Android, where chrome must have +1000 tabs by now (the last time I counted they were +700) We really have to find a way from this enormous amount of content.