 So it's December, the last month of 2017, and usually we make plans for the next year, which is 2018. Yes. And so I thought I would give myself two minutes to talk to you about my plans. OK. What are your plans? So my plan is the template proposal from Apple. Oh, Apple template proposal. The Apple template proposal. So what have they proposed? So we've already got templates, things like moustache and handlebars, that sort of thing. Yeah, also like templating engines. In this case, we're talking about the template tag in HTML. That's already shipped. That is shipped, and it's there to where you can clone and instantiate partial dormitories really, really efficiently. But otherwise, I've never really used it. I just use it as a parser hack. Exactly. And that's kind of what they want to augment. They want to add the capabilities of adding bindings, like we know from moustache. In this particular position in the template, I would like to use this value from a variable. So we already have a template language on the web, because we've got JavaScript's template language. So what's the difference? The difference is that in this case, this would be baked very efficiently into the browser. It would basically do things like we talked about HyperHTML and LitHTML, do these kind of things. Remember which bits are actually mutatable. And then you can just update the instance by giving in a new state object, which like rip out the value that it needs, put them in the right positions, and you have an updated DOM fragment. Oh, so it is very much like HyperHTML, LitHTML on the web. It seems to be very similar from, I don't know how the implementation would look like under the hood, but from the developer's perspective, it is very similar. But in this case, you could just write it in your HTML markup rather than to wire up a imported library and to use the HTML tag string template. Literally. Literally. So are other browsers interested in this? As far as I know, the reactions were fairly good to it. It was discussed at TPAC and the proposals out on GitHub and everything looked fairly promising. There's like some things that have to be figured out, but other than that, it seems like browsers are very much interested. Sounds like a nice high-level thing to augment like web components and stuff. For example, yeah, Shadow DOM, all these little things can make good use of templates and I think it's going to be interesting in 2018. Good timing. And the thing I'm going to describe to you is a flat map and flatten. What? Good question. These are methods that are going to go on arrays.