RSS is one of the most useful communication tools ever invented.
Having a tool that automatically brings you information you’re interested in as soon as that information is published on the web has a terrific productivity benefit.
No hunting and pecking on websites; just relax as everything comes to you.
And if you publish content yourself – from a blog, for instance – that content gets out automatically to people who are interested in what you have to say.
One flipside, though, is when you find out that a feed doesn’t work properly and it’s tricky to find the fix.
A case in point is the feed for comments on this blog.
All my feeds are channelled through FeedBurner. A very useful element of FB’s service is an alert if any problems are encountered with a feed.
I got one such alert this morning. Multiple alerts, in fact, saying this:
I’ve experienced errors like this before with this blog’s main feed, usually relating to text that contains a non-breaking space code:
Such code gets inserted from offline editors like Windows Live Writer, BlogJet and ecto. (Why do they do this? I wonder.)
Fixing that means going back to a post that has the offending bad characters, manually editing the post and republishing.
A bit of a pain but what else can you do?
But this current problem relates to the comments feed, containing content other people have written and not to a feed containing the content only I’ve written.
So I run the comments feed through Feed Validator, and see this:
The problem is clear – the content that results in an invalid feed is in a trackback from another blog.
No problem with that: I get lots of trackbacks. But this one is in another language, not English; in this case, it’s Swedish.
Yet I’ve had trackbacks in Swedish before without this type of feed error. Maybe the cause of the error is the text encoding on the originating blog? I don’t know.
Maybe I need to try and figure out UTF-8 encoding. But do I really need to do that? What would I learn from that to apply to my blog?
Other than going into individual comments to edit them (I really don’t want to do that), there must be a simple way to address problems like this, and before they happen.
Any ideas or suggestions?
If you are able to wrap the content of the description tag with CDATA labels like in the content:encoded section, that sometimes helps with problems like this. These XML validation issues are always a pain to deal with, especially when it is funky characters in question.
Neville, this is an issue with our Feedburner RSS feed for our Monster feed. Affiliate sites pulling in the RSS feed sometimes have trouble if a stray character has caused it to invalidate.
A bit of a kerfuffle.
I’ve come across an issue with the CDATA tag before, Chip. But that looks almost impossible to prevent given that the tag is inserted by an application.
Bryan, agreed – stray characters. Easy to correct when it’s content you’ve produced; trickier with a trackback to content elsewhere, as is my issue right now.
I’ve fixed it although it’s a manual hack involving editing the comment to convert the offending characters to HTML code. Now when I check the feed through Feed Validator, it validates correctly.
There has to be a better way than this. Time for a question in the FB forum.
The next big thing in RSS = PredictAd
Take a look at the demo for NevilleHobson.com:
*simply start typing a search query in the search box such as “thailand” or “sun” and PredictAd will display AutoComplete suggestions with contextual advertisements. Click on the Live RSS tab and you can instantly search rss feeds as you’re typing!
Try it out – there are only a few feeds, but you get the gist.