It’s just data


Items in my comments.rss2 feed now have dc:creator information.  This should help NewsGator users.

You're welcome, Greg.  ;-)

Thanks Sam!  Very nice.  :-)

Posted by Greg Reinacker at

In brief: insomniac edition

A little of everything.  I mean, why not?  I'm up anyway.... [more]

Trackback from dive into mark


While you're in a generous mood, is there any chance to have the comments feed with a different title than the main feed?


Posted by Ziv Caspi at

Ziv - done.  All comments.* pages (including comments.html) have a new title.

Don't be shy.  Are there other things I can do to make my site or feeds more useful?

Posted by Sam Ruby at

A LINK tag on each individual entry that points to that entry's RSS feed would be nice.

Posted by Mark at

Mark: what namespace should that link tag be in?  If it weren't qualified, wouldn't it get confused with the link back to my blog entry?

Posted by Sam Ruby at

I meant an autodiscovery LINK tag in the HEAD of each individual entry's HTML page.  You have one, but currently it just points to the site's main RSS feed, not the entry's RSS feed.

Posted by Mark at

OK, I've updated my template.  It will propagate out as entries are updated or my server cache entries expire.

Posted by Sam Ruby at

Since you're updating your comments feed anyway, how about a <a href=...> link back to the original article from within the <xhtml:body> of the comment?

It would allow SharpReader users to view the comments as a thread off the original article in your main feed.

Also, how about making the comments feed include trackbacks as well? The latest version of SimpleComments will allow you to do this easily.

Posted by Luke Hutteman at


My comments feed includes trackbacks.  And pingbacks.  And excerpts.  And email.  And web services.  Oh, and comments.  [Example]

As to your first request... it seems odd putting that data inside the xhtml.  Wouldn't it be better if there were a separate element in the RSS feed to show in which thread(s) this item participates?  A trackback would have two such "parents".  Thoughts?

Posted by Sam Ruby at


Apart from SharpReader's threading feature, I think the advantage of doing a link back to the original article in a comment-item is that it provides context. I think this is useful information no matter what aggregator someone uses (though SharpReader's threading makes it even nicer)

That being said, a trackback-like tag may be a good addition to rss (under some namespace of course - not in the core spec) as it would allow for a much easier way to provide context compared to finding links in the description.

By the way, since a-href tags are allowed in your comments, but ampersands are auto-encoded, how could I have had the a-href visible as a tag in my previous comment (instead of showing up as <a href=...>) ?

Posted by Luke Hutteman at

I agree with Sam that this information should be metadata in the RSS feed item and makes little sense being included in actual content be it xhtml:body or description.

There's nothing stopping newsreaders such as SharpReader from providing visual cues such as threading the message or providing hyperlinkss to original posts to indicate that a particular RSS item is a comment, pingback or trackback.

Posted by Dare Obasanjo at


<a href="...">

If it is not well formed, it gets escaped.

This comment is in reply to Luke Hutteman in the thread in response to the dc:creator blog entry.

Wouldn't the above paragraph get old after a while (like the second time you saw it)?  On the other hand, some tasteful visual clues (like the curved up and down arrows that sharpReader uses) would be excellent.

Posted by Sam Ruby at

Ok I agree that is kind of verbose... Especially when shown on the html-page instead of just in the rss ;-)

However, you currently end the items in your comments feed with

posted by XXX

which could simply be changed to

posted by XXX in response to YYY

For a less verbose solution. This is what I currently do in my comments feed.

But following your logic, I guess the clean way would be to remove the entire line and add an rss-tag for the link to the author(<ns:authorlink>?) as well.

The advantage of separate tags of course is that aggregators can be more flexible in how they choose to handle this information.

The disadvantage is that users of an aggregator that does not know about this namespace would miss out on this information, and that not many feeds would initially use this new namespace. But of course that goes for any new namespace, and should not be an obstacle to add useful features to rss.

So how does one create a new standard namespace anyway? and can you think of other useful tags for this namespace?

Posted by Luke Hutteman at


As any URL can be a namespace, creating one is easy.  The difficult part is making it a standard.  To do that, requires getting the cooperation of both vendors of tools and producers of content.  It also makes sense to explore prior art.

If you were to pursue this, you would have my cooperation.

I see trackbacks as connecting two blog entries, so separate elements for each connection would seem to be appropriate.

Posted by Sam Ruby at


Thanks for the link to prior art. I see there are already two candidates for this information: mod_threading defines a <children> element (though in a somewhat awkward double-nested rdf format) while dcterms defines both <references> and <isReferencedBy>.

the dcterms elements seem to fit best as it allows for linking in both directions and does not contain any unnecessary nesting of tags.

I did not see any element for a link to the author btw.

Posted by Luke Hutteman at

I'd say dcterms is much better than mod_threading (which would let you say that an entry has comments and trackbacks, but not that a comment or trackback refers to an entry, which would be mod_annotation despite a certain someone misusing it as an equivalent of RSS 2.0's comments element), but be sure to be ready for hamsters like me, and look for the URI as both an rdf:resource and as content of a simple XML element.

The only thing I can think of for an author link would be FOAF, either foaf:homepage or foaf:mbox, depending. And as much spam as I'm getting lately at the email address I use in RSS feeds ("rss_errors, want a bigger...?"), I'd say just skip the mbox entirely. At one point Ben Hammersley was doing a foaf:Person in a dc:contributor for comments in his individual entry feeds, but he seems to have stopped now. Dunno why, though I have to admit that it was one of the straws that broke my attempt at parsing RSS 1.0 with an RDF parser, having no flipping idea what I was supposed to be looking for without parsing a few dozen schemas. It may be wrong, but I really like a few simple elements where everyone agrees what their content should be, and nobody nests a bunch of unexpected stuff in them.

Posted by Phil Ringnalda at

Phil, you are a magpie not a hamster.

I like magpies.

Posted by Sam Ruby at

Mmm, pie.

Posted by Mark at

Setting a few things straight

Help me find better choice, and I will gladly choose it.  Until then, I plan to continue to respect the roadmaps.  Both of them. Dave is providing feedback.  Here's my responses:  I have a policy. This weblog will follow this... [more]

Trackback from Sam Ruby


Add your comment