I'm in the process of build a gogle desktop search-like software. Is for indexing corporate data & files.
I have 3 diferent sub-systems, the search engine, a config/admin tool and a spider. I wonder how balance the metadata and the content. I'm thinking in put the "link" information in a separate database. The "link" information is for example, path, size, author, user name, number of views, bla bla. The scoring is calculate here, and apply in search-time against the index. I'm rigth in this approach? I have a problem, because if I put this info outside, then how leverage this when the user search? I know I can build a link-id and that, but I want to know what problems I can expect if separate this information... -- Mario Alejandro Montoya MCP www.solucionesvulcano.com <http://www.solucionesvulcano.com> !Obtenga su sitio Web dinámico!
