- From: Guha <guha@google.com>
- Date: Sun, 25 Oct 2009 16:06:00 -0700
- To: public-html@w3.org
- Cc: martin@weborganics.co.uk
- Message-ID: <ec9e7c50910251606r688188bl5dcfcce3707d696f@mail.gmail.com>
Here are a couple of cases for itemid that we are working on and absolutely need itemid for. 1) Aggregating data from different sources about an entity: We are working with a small number of agencies (govt. and non-govt) that publish environmental and census related data about different locations, towns, professions, etc. We are depending on itemid for uniquely identifying the various entities (locations, professions, etc.) so that we can aggregate the data properly. 2) In the context of search: Google & Yahoo both use structured data in documents to decorate search result snippets. It would be great to use this data in other search contexts. E.g., if Yelp reviews Acme's plumbing, in addition to using the Yelp review data for the yelp search result, if we could reliably associate that data with Acme's website, the structured data becomes much more useful. Yes, we could, on a case by case basis, invent new attributes that are equivalent to itemid. But given how frequently the need for it crops up, we would very much like to have one as part of the microdata spec. guha
Received on Monday, 26 October 2009 10:35:44 UTC