"ara.t.howard" <[EMAIL PROTECTED]> writes: > i think the real question is: what are you spending more time doing, > creating data or reading it? with relationships and properties > separated creation is always a snap and, while reading might be > slower, the api is always the same for every object. maybe the real > question is how it would scale. my gut says having an index on all > the keys should mean it scales fairly well, the key is the definition > of 'fairly' ;-)
That's a great question, and you're right, most of our time is spent doing development related tasks, and not data mining stuff. I think this does scale pretty well, especially if there is a way to fetch all the properties at once and cache them in the object. If, however, you fetch them one at a time, and I need to access more than a couple of them for a single object, there would be a lot of back and forth with the database. -- Peter Jones, http://pmade.com pmade inc. Louisville, CO US _______________________________________________ Bdrg-members mailing list [email protected] http://rubyforge.org/mailman/listinfo/bdrg-members
