RSS FeedJun 01 2009
RSS Feed Issues
I’ve had a couple of emails over the past few days letting me know that your RSS feeds are no longer updating for the site. I’m currently trying to troubleshoot what the problem may be.
Please bear with me as I tinker with a few things trying to fix the issue.
If you discover you’re one of the people no longer receiving updates, I’d love to know what browser you’re using and whether or not you’re in the U.S. or not. (Feel free to comment below.)
I’ve had a couple of emails over the past few days letting me know that your RSS feeds are no longer updating for the site. I’m currently trying to troubleshoot what the problem may be.
Please bear with me as I tinker with a few things trying to fix the issue.
If you discover you’re one of the people no longer receiving updates, I’d love to know what browser you’re using and whether or not you’re in the U.S. or not. (Feel free to comment below.)
The feed at:
http://www.comingunmoored.com/feed/
is updating; it contains the reference to this entry, for example. However, it is not well-formed XML so more-careful software is likely to reject it. For example, the feed reader I use (Akregator 1.4.2 in the UK) doesn’t parse it.
If I use wget (a utility which downloads files via HTTP like a browser would but from the command link (another similar one is curl)) I can download your feed and read the contents in a text editor. When I first noticed the problem I did that and then tried to read the file in with an XML utility (Saxon) which pinpointed the problem directly.
The problem is that the URLs used in the link elements to reference the web pages containing the entries contain ampersands which are not properly escaped. For example:
http://www.comingunmoored.com/2009/06/rss-feed-issues/?utm_campaign=feed&utm_medium=feed&utm_source=blog
In XML the ampersand character is effectively “reserved” and needs to be escaped using “&”. Frankly, whoever wrote the software which produced this must be a complete amateur and needs to be kept away from a keyboard until they’ve learnt to output XML properly. Seriously, this sort of stupidity should have been stamped out years ago and is totally inexcusable now.
As I mentioned in email, this problem is documented on this page:
http://www.clive-publishing.com/2009/01/wordpress-27-rss-comments-link-error/
although the solution given there (% escaping the ampersands, I think) is at the wrong level.
But that’s all rather theoretical. What’s actually changed to break your site?
I don’t know but that query string (all of the URL from the question mark onwards) is not necessary as the page can be accessed without it. My guess would be that some sort of statistics plugin or something is adding those so you can see how people are getting to your pages. Maybe the characters “utm” ring some bells?
Maybe you can work around the problem by turning off whatever is adding that query string thereby avoiding having the ampersands present. Not the “right” solution but a good workaround until WordPress or whatever gets its act together.
Wow, go with Ed…I wish I had taken that dev class last year…
I was just going to let you know that Google Reader is picking up your feed without a problem. I just scanned the archive and haven’t found one post that it missed.
Jesse
Just in passing, I notice that when Twitter e-mails with a notification of a new follower the URLs it gives to the new follower’s twitter page have these utm_campaign, etc, query strings (which are also not needed for access). I don’t know but it might be a clue as to where the ampersands which are tripping up WordPress are coming from.