| 11:50 pm on Oct 3, 2006 (gmt 0)|
Run the site through Xenu LinkSleuth and look at the report that it generates.
It lists all pages on your site that have links with errors - by pages with errors, and then again by error URL.
| 12:20 am on Oct 4, 2006 (gmt 0)|
Tried that first, Never found the links
that Google says are errors. Not one.
| 12:28 am on Oct 4, 2006 (gmt 0)|
They must exist on other sites, external to yours then.
| 2:23 pm on Oct 4, 2006 (gmt 0)|
No they must exist somewhere within my site's framework, based on the link itself, besides, if they existed on someone elses site, would they be showing up on a Google webmaster's tools report for my site?
for example here is one of about 5 with a similar error
showing as "not found" in the report
| 2:43 pm on Oct 4, 2006 (gmt 0)|
That piece rel= looks suspiciously like a page that wouldn't validate with the w3c validator. (Just guessing.)
| 2:45 pm on Oct 4, 2006 (gmt 0)|
I know what I am about to say is kind of dumb, but i immediately thought of a thread several months ago discussing page rank and a particular page rank prediction tool being used to get some xml information from that tool. I can not remember what it was, but i remember i looked in to it and i was seeing that a web site that i was checking had bogus links in the R N="2" L="2" value.
Its kind of off topic, but i think that if you ever had a page like that, it may be trying to look for it again if you had already deleted it, also I have also noticed it that R N="2" L="2" value that there were bogus pages that never existed.
I think what i am trying to say is that there is a good chance that Google adds bogus links to a site for some reason that has to do with PR and also algorithm.
Thats a guess of course, but [webmasterworld.com...] is when i first noticed that google did that kind of thing. granted its not sitemaps, but i do not see why google would not apply it to other things search and site related.
| 6:55 pm on Oct 4, 2006 (gmt 0)|
You simply have a missing "quote" right after the URL in a link, and the space before the rel= gets converted to %20 each time.
That's all the error is. Find the page with that non-valid HTML and run it through [validator.w3.org...] and fix everything found.
Do HTML validation errors affect indexing? Yes they do. This is a typical example.
| 6:59 pm on Oct 4, 2006 (gmt 0)|
>>>>Find the page with that non-valid HTML <<<<<
bingo...that would be my initial question...find the page....
| 7:15 pm on Oct 4, 2006 (gmt 0)|
There might be more than one, if you use copy and paste a lot!
That's why I always say to people to run a few pages of your site through an HTML validator. It easily finds silly problems like these.
Are we sure that Xenu LinkSleuth doesn't see it? I would be surprised if it did not.
| 7:22 pm on Oct 4, 2006 (gmt 0)|
If there is a group of errors that has a common string like '.htm rel=' and if you have a utility like grep, then you could search the directory for all the files that have that problem (e.g., missing quote after htm).
| 7:29 pm on Oct 4, 2006 (gmt 0)|
Xenu LinkSleuth found other errors, though not major ones such as what it interpreted to be image link errors that in actualality were not, but these were not found.