DMOZ clones are more uptodate than the original

[kaled]

A couple of months ago, a page on my site got a dmoz entry. Unfortunately, they used the wrong url. (The url works but is likely to break). I requested a change to a more generic url (that won't break in the future). I repeated the request yesterday, however, I just discovered that several DMOZ clones are using the correct url. Presumably, this means that the editor changed the url but the listing displayed by DMOZ is outofdate.

Has anyone else noticed clones being more uptodate than the original?

Of course, this would go some way to explaining why Google likes DMOZ so much - they are both seriously flaky!

Kaled.

[motsa]

>>I repeated the request yesterday, however, I just discovered that several DMOZ clones are using the correct url. Presumably, this means that the editor changed the url but the listing displayed by DMOZ is outofdate.

It could just mean that the downstream users decided to update the URL themselves. A note: if you're trying to change the actual URL for your site to a redirecting or framing domain name, you're not going the get the URL change as the ODP lists the underlying URLs, not the covering domains. You can always ask about the status of your update request in the ODP public forum.

[choster]

There is no longer a single dmoz.org; there are several mirrors in a pool as well as at least two official mirrors in Europe. While they are all supposed to be synchronized every few days, in practice the syncing may fail on one or more of them. So yes, it is possible that a clone which scrapes ch.dmoz.org might be more up to date than another which scrapes de.dmoz.org, and both might be more up to date than one or more of the servers which represent dmoz.org itself. The canonical version (editors.dmoz.org/editors/editcat.cgi?cat=Category/Path) is only available to editors, however.

[kaled]

motsa,

The change I asked for was in line with guidelines for editors i.e. chop off the pagename.html part. I agree, it is possible that the copies were adjusted, but of three that I checked, all had been changed to ...dir/ rather than ...dir/pagename.html. None had been changed to ...dir (which was what I actually requested).

Seems odd to me.

Kaled.

PS
Pagename was of the form a.html?b.html so it seems unlikely a simply filter did this.