Is there a place to file Google bugs? I couldn’t find one, so here is a report I uploaded to the cloud. 
Description of the problem: Google returns feed URLs in search results. These are rarely helpful to searchers in general, but downright mystifying to those who don’t know or care what feeds are in the first place. When using the main Google search interface, I’d imagine that feeds are not high on the list of desired results for most users.
Steps to reproduce the problem: Google for “wordpress mysqli”, and look at the second primary result (your milage may vary, considering Google’s various customisations of search results).
Actual results: Feeds appear in Google search results, sometimes even prioritised over their html document parents! I took a screenshot to illustrate:
Expected results: Removing the feeds altogether might be over the top — it depends, of course, on the context of the search and the feed. But where they might be useful, Google could provide one of its signature unexpectedly awesome features. Maybe something like this (though I probably should’ve snipped the summary in this case):
How often does this happen? All the fucking time!
But I still love you guys.
9 Comments
Perhaps you have to be logged in to get this form?
http://www.google.com/quality_form?q=foo+bar&hl=en&safe=off&pwst=1
“Is there a place to file Google bugs?”
anywhere. the eye of google sees everything
Or bether yet (with some googling ;))
http://www.google.com/support/bin/request.py?page=troubleshooter.cs&problem=irrelevant&contact_type=irrelevant&stage=fm&contact_typemaster=prob&bugs=irrelevant&Action.Search=Continue
I think it shouldn’t show the feed as a result whatsoever. However, it should show the main site as a result with the ‘feed icon’ beside it (like in your second image but without a second result) with a link to subscribe to the feed.
As the feeds are just an other version of what exists everywhere, and often in a less informative way (worse for the first encounter, evaluating the feed contents), they shouldn’t really be shown at all imho in any search results.
I’ve noticed the same thing recently with xml for a html document. The assistant on my floor was trying to find articles for my professor and found the xml and didn’t know what to do with it. I deleted a few items off the end of the URL and we found the html file. She shouldn’t have to know how to do that.
Maybe the problem is also that the feed seems invalid. At least also Firefox does not think it’s a feed when you go there.
But Google could of course detect/classify it nonetheless.
Just do what you did
Blog about it ! I did it last week about the Reader and Gears combination and I had a comment from a member of the Reader developer team in less than a day asking for more details.
The feeds should obviously be excluded from the search results. Anything not being “text/html” or “application/xhtml+xml” should be excluded or at least marked up properly as not being an HTML result (like PDF and PPT files are). When clicking a feed search result, Google should do as it does with other non-HTML results; show an alternative “view as HTML”, so it’s possible to view it in the browser.
One Trackback
[…] Waugh wants to report bad search results to Google, but where? His commenters have some good suggestions: # […]