Well, the best place to discuss this is the K2 forum for one:
http://www.joomlaworks.net/forum/k2
As to what you say, I don't consider this a bug. I'm just gonna add a comment, but if you wanna continue this discussion, please head over to the K2 Forum. If there's a better way to do this (in terms of router logic), I don't have any objection to update the URL router.
Now, what you're doing is visiting fake URLs right? Cause there's no way to visit
http://www.utahsymphony.org/support/adv ... ven-wagner from your menu or modules, right?
The same goes with core Joomla articles. And you don't even need to specify a different existing menu alias. You can place whatever you want before the article id. And whatever you want after it:
This is the live URL
http://www.joomla.org/announcements/rel ... -here.html but I can easily make it
http://www.joomla.org/announcements/WOW ... -here.html and still see the same content.
Now let's have some fun with the article alias:
http://www.joomla.org/announcements/WOW ... WOWOW.html
So, it's kinda how Joomla's router logic works in general. When a URL is processed in K2, we search patterns, starting from the item, moving on to the category and then tags, users etc.
K2 may generate double URLs in some cases, especially with tags and categories that don't belong to some menu item. But again this is a side-effect of how Joomla works. When you're inside some K2 item, all tag links within this item will have the item's URL path in front of it. But, again, this is normal when you build a router the Joomla way. And frankly I'm not aware of any other way to fix this.
But at least in K2 you can easily solve this problem by creating a new menu in Joomla that doesn't get rendered in any module and in there creating a single menu item to all K2 categories (just don't select any category and K2 will assume you want them all in the menu settings). Give it a nice menu alias (e.g. "content" or "site") and all double tag URLs (for example) get normalized to domain.tld/content/tag/x or domain.tld/content/tag/y. In the upcoming K2 v3 we actually have a better mechanism for this and you can define prefix paths for separate content entities in K2 (tags, users etc).
Now, the reason why Google scans this URLs is probably from your own browsing history. If you're logged into your Google account and browsing the web and got Google Analytics on your site too, know that
everything gets reported back "home". Otherwise Google simply cannot crawl URLs that don't exist. In your case, you're creating the URLs by navigating to them.
For more, let's continue this over at the K2 forum.
