For repositories like Youtube we can't download the content. We need to download the URL to the content and store it somehow, either as a plain URL (processed by filter later) or as some parameters to an Object tag or something ...