On 9/1/06, Jakob Lechner <[EMAIL PROTECTED]> wrote:
We consider the primary use case of smart tags to be to link words (that are recognized by an external library) in a document with actions provided by this library. The user usually doesn't add patterns manually, but another application provides the patterns. For example if you have an inventory system, it could register inventory IDs with OpenOffice and so have any documents containing inventory IDs be possible gateways to the inventory system.
OK, that helps clarify. Thanks. As I at least am envisioning enhanced metadata, the location of the representation is largely irrelevant, and mostly just a question of convenience. So think about the ctiation case: in text, you have a URI ID. That URI could be some resolvable URL, it could be a more abstract URI (like a URN) that a tool can resolve somewhere else, or it could also be used to identify the embedded metadata description as a way to link the two. E.g. if I am understanding you right, maybe there is room to consider both the same kind of use case? You use the smart tags here to enable a piece of document content to be in some way associated with some data-related action or representation. In the inventory case, you're somehow connecting to a record somewhere (an online database? maybe tying into XForms? I dunno) to accomplish some task (user wants to see or edit the record maybe). In the citation case, we need to allow tools to grab a data representation in order to process the output for display. Bruce --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
