We need to identify valid username links. It will be much better for the experience if we only turn the cursor into a wand when the user is hovering over a link that we know how to pull a username from. Youtube, for example, has many links that look like usernames, but are actually g+ redirects, channels, etc. If we don't know how to get a username from a link, we should reject it to avoid confusion when tipping doesn't work.
We need to be able to get the username from a link. Usually the actual text of the a element will work well, but who knows. For instance, on Facebook, the username is actually a part of the url, and the text is just a display name.
These will need to be different on every site we support, so I suppose we'll need to come up with a good strategy to manage that as well. Later, if we want to get crazy, we can write some script on the serverside that actually follows the link and extracts a profile photo etc. Not essential right now.
aelement will work well, but who knows. For instance, on Facebook, the username is actually a part of the url, and the text is just a display name.These will need to be different on every site we support, so I suppose we'll need to come up with a good strategy to manage that as well. Later, if we want to get crazy, we can write some script on the serverside that actually follows the link and extracts a profile photo etc. Not essential right now.