This usually happens if Zotero does not recognize the page as WorldCat (most likely because of proxy), but does pick up I some metadata embedded in the page. What URL are you importing from (exactly as you see it)?
translators update separately every 24hs, they're not tied to a Zotero version. You can trigger the update immediately by clicking on the "Update now" button in the general tab of the Zotero preferences.
When I click the book icon in FireFox's toolbar to "Save to Zotero (Open WorldCat)", it creates a webpage entry.
This one imports correctly as a book: http://www.worldcat.org/oclc/636120100
I think the issue is that whenever I try to import e-books from WorldCat, Zotero considers them webpages.
Importing using COinS doesn't have any of these issues.
Let us know if there are further issues.