Jan 03, 2006, 01:02am EST
I wrote the following a while back on the flockstars mailing list:
…the more I think about it, the more I think that users shouldn’t have to copy/paste a feed URL directly. Instead, whatever you’re copying into should be smart enough to find the
<link>tags automatically. …
The distinction between a feed url and a site url feels more like a technical distinction, not a semantic distinction.
As Bill pointed out, this is only useful if your aggregator knows how to follow autodiscovery links. So I decided to see which aggregators were able to do this. I decided to use this site’s URL because it has two autodiscovery links (entries and comments) and I wanted to see how aggregators handled that as well. I tried almost all the aggregator urls found in the log files for this site.
|found none||found main||found both|
So it looks like most aggregators understand autodiscovery. And because most understand autodiscovery, it’s not really necessary to have an orange feed icon as a clickable link. A feed isn’t for human consumption, so why make it clickable?
Alternately one could have the icon go to a page describing what a feed is or perhaps point out the autodiscovery icon in the user’s browser.
And yes, I realize I still have the link on my site. I’m thinking about what to do about it.
Jeof Oyster at Jan 04, 2006, 03:15pm EST
It amy also depend on the format fo the feed. I use intraVnews (a plug in for Outlook) which has no trouble with autodiscovery for RSS, but can NEVER find Blogger/BlogSpot s, which are always the URL+ atom.xml and are in the Atom format.