Re: [FeedValidator] Re: Fragment Identifiers
"Bill Kearney" <[email protected]> Mon, 14 Mar 2005 10:05:13 -0500
| Newsgroups | gmane.network.syndication.rss.devel |
|---|---|
| Message-ID | <[email protected]> |
The point about prohibiting relative URLs is that the source URL might not be 'available' to the program processing the RSS. Either then and there during the download or (more likely) well after the fact. Were this RSS-1.0 there'd be some merit to using fragments as rdf:ID and rdf:nodeID attributes. But in all the non-RDF flavors of RSS it's not supported. Perhaps a follow up question is in order, what is it you're trying to accomplish? Are you trying to make HREF links from one RSS item to another? Or are you trying to apply some sort of semantic identifiers on blocks of text? One thing that's always bugged me about RSS feeds (regardless of flavor) is the inability to refer to a specific item from a given 'instance' of a feed document. As in, item 6 of the feed published on 3/14/2005. There's just no way to do it. This is not a bad thing, per se, as the feed itself is really only supposed to be a set of descriptive pointers to the main content. The content's not /supposed/ to be in the feed (although many folks seem bent on doing just that). I suppose something like a p2p link could be assembled based on a hash of the feed instance and the item ID. In RSS-1.0 this would be simple enough as each feed item is required to have a feed-unique rdf:about attribute. Granted, this is again rather pointless as the feed itself isn't the content; the website is the content. So the inability to make relative references to an item from a given instance of a feed really isn't all that important. It's one of those 'gee it would be neat ideas' that doesn't hold up to practicality. -Bill Kearney Syndic8.com ----- Original Message ----- From: "Bill Kearney" <[email protected]> > It's still a relative reference and thus not valid in an RSS feed. > > ----- Original Message ----- > From: "Jacques Distler" <[email protected]> > > > refers to id="foo" of the *current* document. No relative (or absolute) > > URL resolution is involved. > > > > So what's "invalid" about it? Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/rss-dev/ <*> To unsubscribe from this group, send an email to: [email protected] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/