The Importance Of RSS Feed Autodiscovery
By Rob Safuto on Nov 2, 2007 in RSS | Tags: auto discovery , Feeds , jaiku , lifestream , RSS
I recently joined the lifestream/microblogging service Jaiku thanks to an invite from my pal Eric Rice. You can now follow me over there at http://robsafuto.jaiku.com. Be warned that my Jaiku presence contains info from just about everything I do on the internet. The fact that I can easily aggregate all of the things that I publish (including videos, photos and podcasts) while also keeping track of everything my friends publish makes the Jaiku service a very useful tool to me.
Jaiku makes it easy to gather all this info thanks to the lovely technology of RSS feeds. To me RSS is the most common sense protocol we have for sharing data on the internet. At the same time there are little nuances to the technology that can make or break the experience for users.
With Jaiku you can enter either a link to a page or a direct RSS feed so the service can track what you publish. If you enter a link to a page then Jaiku scans the page for a valid RSS feed. Here’s where it gets tricky. Sometimes we as publishers don’t realize that our pages are not optimized for services like this to recognize the proper feed associated with the content. This isn’t the end of the world. After all if you enter the page link for one of your own websites and no feed turns up in Jaiku then you can always enter the RSS feed address (which I’m assuming you know) directly. But what if you’re using another service to publish?
In setting up Jaiku to aggregate channels on YouTube and Blip.tv I found that neither provides good publisher support for RSS auto-discovery. YouTube returns no feed at all. And Blip.tv returns a feed of all the recent videos published to the site. This isn’t good news for publishers.
Current versions of Firefox and Internet Explorer will autodiscover RSS feeds. So that means visitors who land on your profile page on YouTube or Blip.tv won’t get the benefit of the easiest discovery of your feeds.
I also found during this process that some of my blog sites were displaying the wrong (or no) feeds for auto-discovery purposes. This is likely to be the case if you use a service like FeedBurner which creates feeds at a different address than at your blog domain. This issue can be solved by placing a line like the following in between the <head></head> tags in your blog template. Be sure to replace my blog feed title link with yours.
<link rel=”alternate” type=”application/rss+xml” title=”Awakened Voice” href=”http://feeds.feedburner.com/AwakenedVoiceBlog” />
You should also remove any similar links if you don’t want applications or web browsers to discover them. I highly recommend checking any sites where you’re publishing to make sure that the RSS feed auto-discovery feature is finding the feed that you want people to subscribe to. If it isn’t then you could be scattering your subscribers across feeds or even losing some potential subscriptions.
For the moment there’s probably not much you can do about 3rd party services who don’t present the right feed meta data in your page except to let them know to correct their header tags.
Mike Hudack | Nov 2, 2007 | Reply
Hey there,
We actually offer individual RSS feeds for each blip user, complete with enclosures — even multiple enclosures using MediaRSS.
It does look like the RSS autodiscovery in our header is wrong, though, and I’ve added a bug to our tracker regarding that issue. Hopefully we’ll get it fixed soon!
Yours,
Mike Hudack
Co-founder & CEO, blip.tv
Rob Safuto | Nov 3, 2007 | Reply
I doubt that most people take notice of little things like this. But every bit of visibility helps for content producers. Thanks for monitoring and putting the issue in the queue.
Mike Hudack | Nov 3, 2007 | Reply
Hey Rob,
Turns out this was a really quick fix. It’s already been resolved in our new maintenance release, which is scheduled to go out the door in about two weeks.
Yours,
Mike