Since this is quite a conceptual question which might involve a bit of discussion, I've made it a community wiki
I'm working on a site that will work as a fairly simple aggregator of content from various sources including several Twitter, Flickr, and Facebook accounts. Updates will occur on a periodic basis, at which point feed data will be fetched and updated in a local store. A single page on the site will display the latest set of content; other static pages will be updated occasionally.
I'm building the site on top of WordPress to facilitate easy management for a non-technical editor.
My initial prototype stored details of each feed item's content (e.g. title, link, description, image) in a custom built database. However, since beginning the WordPress side of things, I'm wondering if it wouldn't make more sense to use the existing WordPress database for this purpose - i.e. store each feed item as a WordPress post (in wp_posts) and additional data in wp_postmeta. The category/tag model could be used to differentiate Twitter items from Flickr ones, etc. This would give the advantage of using just a single database, and also making all WordPress functions available for the management of individual post items.
Does anyone have an opinion on this, particularly any 'gotchas' or possible pitfalls that may arise from such an approach?