where do the 404s in the webmaser central indexing stats come from

[Oliver Henniges]

The webmaster central console is a very helpful tool to get one's site clean and search-enginge-friendly. However, from the very beginning, I was wondering about the error reports.

By the end of january my sitemap-account reported zero errors for a short period of time, because in the previous months I had used the data to ftp some 301-dummy pages, which should redirect external broken links to the appropriate page. But recently I submitted two new pages (residing on root level) and a few days later I found exactly these two pages as errors, the spider expecting them under wwww.mydomain.com/folder/pagename.html

I checked as careful as I could, whether this is due to any internal broken link, but I found none. Neither google nor yahoo have yet indexed these two pages, so I doubt any scraper directory has a broken external link to one of these two pages so far.

Additionally, as for this morning, the webmaster central console reports a third error, containing some abbreviative dots in the url not found. I have seen this several times before; it seems as if the (abbreviated) anchor text of the google page displaying the data in some cases gets mixed up with the urls themselves somewhere in the databases.

Where do these errors come from?
What can we do to help improve the displayed results?