Msg#: 4255134 posted 12:52 pm on Jan 19, 2011 (gmt 0)
This could be important for SEO!
I was researching a 400 error (Bad Request) reported by webmaster tools. The link reported by WMT looked fine! But when I clicked on it, sure enough, I get a 400 error from my server and the link produced by WMT is corrupted. So repeating; looking at the WMT page gives you no hint at the problem, BUT, when you look at the WMT web page source you discover what could be a significant problem for you, another webmaster, or Googlebot itself. Below the actual code extract examplified (Hmmm a new WebMasterworld Term):
The link should be: www.example.com/example-link.htm The WMT reported bad link looks great as well as the link in the status bar when you hover, BUT The text for the link in the source file is: www.example.com/example-​link.htm And the WMT generated href is: www.example.com/example-%E2%80%8Blink.htm
Which of course my server doesn't like too much. I may be able find the source website in my logs, but I may have to look back for months to do so.
The ​ is, I believe, a zero width space probably in somebody's website somewhere, BUT, the gotcha here is: The link displayed by WMT looks fine!
The moral of the story is; You may want to view source in WMT when investigating problems reported by WMT.
Msg#: 4255134 posted 8:59 pm on Jan 19, 2011 (gmt 0)
I have few of these too. In my case these are dynamic URLs that also look fine in WMT and in status bar when hovered, but when you click on them, you get ? & and = encoded to %3F %26 and %3D that is, URL