DMOZ mirror sites

Why so many?

[twilight47]

Does anyone know why there are so many site that take the literal ODP index and just duplicate it?
Why would someone use their index instead of going straight to the source?
The only good thing is these other mirror indexes appear more up to date than the Google directory. ;)

[choster]

So you can provide a web directory to your users, but from within your site, with your look and feel, offering your features or enhancements for browsing, advertising your products or services for sale, or to otherwise pass off a big database of sites as your own.

Why provide a web directory? Well, it depends on what you're building your site for. Perhaps you are an ISP and want to provide a customers' home page with pretensions of portalhood. More likely, you are a hobbyist or specialist in some niche subject or field, and you want to promote your hobby or specialty by offering a reasonably authoritative listing of similarly themed sites -- without the trouble of digging excavating them from search engine spam yourself. A great many ODP licensees only use the listings for some small branch, rather than the whole shebang.

And then of course, there are a few who will simply because it's free, and they can, the same reason why moreover.com headline feeds, weather.com icons, espn.com scoreboards, etc. appear on personal homepages everywhere.

(technically speaking, of course, we're talking about ODP data licensees-- currently the only two true mirrors are ch.dmoz.org and de.dmoz.org)

[g1smd]

Some sites using ODP data haven't updated for 3 or 4 years, and look quite sickly now (low amount of content, large number of links no longer working and so on).

There are a core of sites that update several times, or more, per year, using the RDF file.

Lastly some sites actually dynamically read the ODP pages, live, reformat them and serve them as their own. These screen scrapers are always as up to date as the ODP is itself, but tie up resources on the ODP servers. These latter types are requested to screen scrape the mirror at [ch.dmoz.org...] instead.