I love using my aggregator to stay up-to-date with all the goings on at the various websites I read regularly. Unfortunately, many many websites cross-post the same items. I've recently begun reading Gizmodo (I went to high school with the editor!), and I've been increasingly annoyed at the number of items first posted there that then proliferate on to BoingBoing or Slashdot. Both BoingBoing and Slashdot have enough new content to make it worthwhile to continue to subscribe to them; but I really wish that this Web 2.0 craze would find some way to keep me from reading the same story on three different sites in my subscription list.
Signal to Noise
4 Comments
On Carthik added:
On skippy added:
I keep glancing sidelong at Gregarius. Every couple of weeks I'll install it, re-import an OPML export from FoF, and try it for a couple of hours. Then I get annoyed at how ugly it looks, how much it gets in my way, and ultimately delete it again.
I don't want to fiddle with the themes. I don't care for tagging (yet). And I hate the default decision to show me the last set of titles I've already read if there are no new items. If there are no new items, I want a blank page! I see the SVN version of Gregarius has an option to permit that; but all the other cruft is still there to get in my way.
I like FoF because it shows me the contents of my feed subscriptions, and stays out of my way. It was easy to tweak the stylesheet to use a friendly, large font. It's easy to add and manage feeds. Yes, FoF suffers from some bugs, but for my usage patterns, it's far more useful.
However... if Gregarius could alleviate my concern outlined in the post above, I'd happily switch, and jump through any hoops necessary to style it in a way that doesn't completely suck. At least for posts on BoingBoing that originated from Gizmodo, you could filter based on the via Gizmodo attribution...
On Bertil Gralvik added:
Yes, multiple postings of the same content is a little bit web 2.0 crazy.
What in fact happened to the old web 1.x way of doing things? I mean, after all linking is what html is all about (?)
A little redundancy may not hurt, but too much is too bad.
Cheers though!
On Bertil added:
Giving it a little more thought, the redundancy of the web could be less. Not only by the HTML inherent way of linking.
Every page or article or post on the web has its own identity or URN and the idea of syndication could be used a bit more clever.
As matters stand, syndication is normally done at the client side using some feed reader. If the ID followed the content, equal posts from different sources could be filtered out. Or maybe even better, a syndications server could do this for us.
We would of course have to convince people not to cite other peoples work in extenso, but excerpt, comment and link. The concept of the semantic web is just about that: a way of telling the web what you want to know and the let it collect relevant information on you behalf.
A long way to go, no doubt, but a promise made in the blue for the future to expect.
Something I have been wanting to avoid for quite some time now, too. This gets painful with Technorati search feeds and the many regurgitating websites that contribute little original content. The Linux news websites are worth special mention here. Stupidity!
Maybe I should write a function or two and plug it into gregarius, the feed reader I finally settled on.