Originally from ticket #22421.
Great question. So the way the YouTube connector works is that YouTube has its own API that doesn't follow any particular standard or specification, YouTube just came up with it themselves. The connector gets data from that bespoke/custom API and then translates that into a learning industry standard format, specifically Tin Can statements. Any events need to be present in the YouTube API in order for the connector to take them and translate them.
If the event isn't a part of the YouTube API, there's a few other approaches you can follow:
1. Convince Google to adopt Tin Can on YouTube and allow learners to send their activity on YouTube to an LRS whenever they're logged on to YouTube. To be honest, even if you were very convincing and had contacts with Google's YouTube team, this might not be the right approach as many YouTube users do not login and might not configure their YouTube accounts to send data back to the LRS. It also only works for one website; what about Vimeo?
2. Have the user browse the web in a frame or pop-up and have the parent window track pages they visit. This approach is used by some social/curation learning platforms such as Curatr. This is useful where you want to direct the learner to specific content.
3. Give the user a tool like the bookmarklet (http://tincanapi.com/bookmarklet) and have them self-track their experiences. This allows the learner to filter and track what they find interesting.
Does that help? Which of those approaches seems most relevant to your situation, and do you have any follow up questions about that approach?