Am I the only person that doesn't find this to be much of a problem-solver?
I read a lot of regularly-updated websites, from Slashdot and Metafilter down to some more obscure stuff, and pretty near all of them offer RSS feeds. Instead of having to go site to site, I can just read them all in once place! It sounds like a great idea.
So I tried syndicating them here on LiveJournal. That didn't work out very well, because some of them (Slashdot, BBC Newsworld) get updated so often that I end up having to go back pages and pages, and missing all of the ones that update once daily (which are usually pretty high-quality) in the noise.
So I decided to only read the low-traffic ones that way. It still didn't work out that well -- I'd forget to check, because it didn't update that often, and then the sites' posts would be interspersed with one another, and so on.
So I tried a standalone RSS reader (well, one built into Firebird), which let me read one site at a time via RSS, and that seemed even more useless, because reading one site at a time is what it's meant to avoid, or so I thought. If I'm going to read one site at a time, I might as well bring up the site and see the original!
It seems that RSS readers miss out on all of the things that make the original sites interesting to read. For instance, on Slashdot via Alterslash and on Metafilter, a lot of information is conveyed in the number of comments on a post; on Kottke.org, the interspersed reviews and "remaindered links" break things up nicely; on BBC Newsworld, the size of articles' headlines conveys useful information, and so on.
In the end there are three things that I use RSS for that I find useful, or that I understand the value of:
- Reading other people's blogs -- but not news sites -- on my LiveJournal friends page. That lets my friends page work whether or not someone happens to use LiveJournal, and is thus good.
- Adding modules to infobots without having to send someone a patch. Infobot has had a "read RSS headlines" feature for a while, and for the most part it's never used for actually reading headlines anymore, but as a general RPC mechanism. There are a gazillion nicer RPC mechanisms that could be used here, such as the ever-popular "plain text in response to an HTTP request" one, so this seems only incidentially good.
- Slashboxes and their equivalents -- listing the headlines (only) from a related website in a sidebar.
Am I unique in not having this itch to scratch?