So, where are we? So, here's our canonical set-up. There's a doctor on the right, interacting with a computer which gets patient data from the EHR. Gets data from the Physician, maybe their preferences or some other observations, and then has to use the knowledge in the knowledge base. So last time, we talked about the knowledge is in the knowledge base. This time, when you want to talk you about how's the knowledge get into the knowledge base. You could see that there are basically three ways. There are three sources of knowledge that we'll be talking about in this module, expert knowledge, scientific literature, and the enterprise data warehouse, the EHR and similar data sources. Almost anything you think about maps into one of those three. So, let's start off with the expert knowledge because no matter what software development you're doing, no matter what the intervention is, you need to talk to the client. You need to figure out what they need, you also have to figure out what they know. Now, it used to be thought that all you need to do is talk to the expert, they'll tell you the knowledge and you are done. But it turns out that, that step of speaking with the expert and gaining knowledge is in fact the major bottleneck for most of these knowledge-based systems. So, as I said, you think that it's in their head, they can speak it out, they give it to you and you done. But no, that model is wrong. In the next talk, we'll talk about how it's wrong and how we deal with that.