next »

[webinfoguy25]

How do you code your site to prevent google bots from marking directories broken when href= something that starts like this: (../) 

[perkiset]

That's a relative URL. That means "Move up one dir, then back down into (whatever follows) or that item specifically means get the index.html page one directory up.

My recommendation: Do not use relative urls. Always start URLs with / so that Google knows precisely where everything is. Relative URLs can be really hard to debug in a large site, and there's no reason for them at all (without looking at really advanced implementations and black stuff).

[isthisthingon]

Related link.
Another related link.

Not sure if ranking is your priority or simply fixing the Google-break.  Links such as the following should be the same to Google:

Absolute URL:

href="http://www.anysite.com/directory/somepage.htm"

Relative URL:

href="/directory/somepage.htm"

Absolute links potentially add unnecessary weight to your pages which, if performance is effected, may negatively impact your ranking at Google.  That said, most people still prefer absolute URLs and in your case it would also fix your problem

[perkiset]

It's been kind of assumed that anything that starts with / is an absolute rather than relative URL, because you're claiming the dir hierarchy from the root of (wherever your web pages are). If you start with / then the domain is assumed. On the other hand, if you start with anything OTHER than / then it is certainly a relative link.

It's like the difference between these 3 statements:

rm /adir/another/file.txt -- remove file.txt in (root)/adir/another
rm ../adir/file.txt -- remove file.txt that is one a directory one up from me then into adir
rm adir/file.txt -- remove file.txt which is in the directory adir below me

URLs work exactly the same way. This immediately looks to be cool. It is not, because it makes page position in a website dependent on current position. It's particularly troublesome for spiders and scrapers. Net-net: I'd not use relative URLs unless it was absolutely necessary. And I've never found it absolutely necessary

[isthisthingon]

I truly have no opinion either way since SEO is not in my bailiwick.  However, the points on both sides are interesting

[perkiset]

Actually I'm speaking purely from a management perspective.

I've created several sites (a great long while ago) where I used relative URLs and created logical loops that spiders got caught in and they never went away. There is no benefit, as far as I can see and IME to relative URLs (from an SEO perspective) so I don't think it's a good idea to use them.

Not disagreeing with you, just trying to clarify.