 All right. Thanks, everyone. Sorry for the problems. I just had to restart the computer, and that usually I was fixed as it, so it wouldn't connect. Thanks for coming to my talk. I feel a bit trapped back here, sorry. Talk is Drupal 8, the schema.org, how everything is connected. Essentially, what I want to talk about is how you can use, I guess, some of the latest things out there in search land to make your sites more relevant and more useful. I've done a lot of talks in the past on semantic web, which I tend to really enjoy, but they're very esoteric and somewhat more academic, and so I think semantic web in a lot of ways is something that we'll never quite get to, but where I was going with, whereas schema.org is something that actually is useful today and is something you can put in place today and see real value with. My name is David Peterson. I'm a senior solution architect with Acquia here in the Asia Pacific region. This is my Twitter handle, and I've got the Sunshine Coast office with Acquia. So me, myself, and I. So what I want to cover is what is schema.org and why you'd want to use it, some use cases, and then I will show some demos and things like that, so how you can actually use it in practice. So there's a lot of, Drupal 8, of course, has some really great tools for it. Schema.org is built in. There's also a tool called RDF Builder and RDF UI, which I'll show you as well. So I just want to run through a few example sites that are out there. So this is one of my favorite movies for comedy. So I mean, we look at this, and it's quite easy to see that it's a movie. It's got a 90% rating. It's called Raising Arizona, 85% audience, sort of average rating of 7.6. So again, I think most of us can look at that and just know that it's a movie. Same thing here. There's, sorry if it's cut off, but this is a pizza from an SPS site. And again, we can see that it's a recipe. It serves two, 20 minutes prep time, 35 minutes cooking time, and it's an easy skill to put together. Same thing here. There's just a commerce site. I tried to use Australian sites as much as possible. So you can see it's a cannon, how much it costs, the delivery, things like that, 12-month warranty description. So again, we as humans can look at all these things and instantly understand what they are. Of course, machines don't have that same flexibility. Google has gone a long way, and lots of search engines have gone a long way, to use their own artificial intelligence engines and all sorts of things to make sense out of this soup that's out there. But they're also driving forward the notion of semantic data or marked up data, which actually makes them, helps them do their job better. And so kind of coming back to the origins of what schema.org is. So back in 1992, when Tim Berners-Lee was first sort of formulating the very early parts of the web, basically I think this was the third web page that was ever, the fourth web page that was ever created when he was at CERN. He talked about actually having link types. So they actually dropped it. So we ended up having this href equals whatever. So if it's a document, if it's a person, if it's a video for 20-some-odd years, we basically hyperlink was just something to somewhere else. But back at the very beginning, they actually were thinking of using sort of a semantic type web of actually saying, is this a person or is this a document? He got shot down. They didn't pick that up, unfortunately. So it's taken about 23 years to get back to where we started at. And here's some interesting, these are diagrams that he also put together. So here's, again, a traditional link to here, to here, to here, to here, to here. It's all in terms of how a computer sees it, it's all just essentially a page with a hyperlink from here to here. So it's pretty flat, pretty boring, not very interesting at all. Here we can, angles are always good. And here you can see that actually these links were talking about many different things. And again, sorry for the, it's all a little bit anti-aliased and things like that. But here's a house, here's a person, there's an idea, there's a piece of clothing. So each one of these pages was actually a thing. And they wanted to describe those relationships. But again, it didn't quite work that way. So why did it take so long? What was the reasons for that? A lot of it was because the semantic web was driven by academia. And I think that's a good thing, because there's a lot of research that's gone into it. But it's also at the same time, the ability to sell, the ideas, the ability to implement the ideas were never really quite there. And for commercial enterprises and for businesses, there never really was a reason for including this kind of extra burden, if you will. But so here is, I guess, you can actually see what you can see today if you do a search for pizza. So this is kind of showing, well, why do you want to do this kind of stuff? So you actually get rich search results. You see, you do a search for cauliflower crust pizza. And you get images. You get ratings. And you get reviews and all sorts of things. So it's very compelling. And for sites that don't have this, which I'll show you later, there's a natural inclination for individuals to not click on things that aren't maybe as pretty. And also that maybe don't have as much information. I know when I'm looking at stuff, the ability to quickly see your different ratings, it gives you an idea, well, this might be more interesting or might taste better than other things. You see, there's 75 reviews, 94 reviews. So there's actually a lot of information in that little small area. And as we go through, I'll show you how we can do that. And here's some more ones. So these are Thai restaurants in Melbourne. So again, you can see some that don't have it. These ones just, this doesn't have it. But some of these other restaurants do. So they have ratings. Again, you can see TripAdvisor is doing a good job. But when we get to non, maybe the bigger players, they don't tend to have those sorts of things in there. And lastly, the other reason, and this is getting even further than schema.org, Google is building the knowledge graph. And that's really kind of taking it. That's been a multi-year project. There's billions and billions and billions of connections they have. And literally everything in the world has a Google ID behind the scenes. Some of those things that they share, and some of them they don't, they're actually going to be releasing a new API in April. They're deprecating the freebase back end, which is currently what helps them drive this. So you can kind of think of freebase as if you're familiar with Wikipedia, it's sort of a textual representation of the world, but it's not data. So freebase essentially took the textual representation of the world and turned it into a graph of data. And then Google bought them. And now that's why when you search for Melbourne, you get these rich search results on the side, which allow you to actually explore the graph all within Google. Google wants to keep you on their site. Facebook also has the social graph, and LinkedIn as well as the graph. And so if you click on, if I go back here, like if you click on National Gallery of Victoria, which I did, you're essentially traversing that graph all within Google and you don't even really, you don't even actually realize it. So as soon as you do that, you get this top bit as well, which actually allows you to see other points of interest within Melbourne as well. So you can kind of just keep on going, keep on going, and you can get, you get your traditional search results here, again with any ratings and things. But as well, you get these rich graph representations over here with reviews and all sorts of things. So really schema.org, I think one of the key drivers to that is really search engine optimization. And also, you have the biggest players are all participating. So I mean, really, as long as Google is doing it, that's kind of all that really matters. But I mean, we have Bing and Yahoo, which isn't really as relevant anymore. And Yandex is more of, again, more of a Russian search engine. But essentially, you have the big players going. And Google alone has, this is as of last year, but 15% of the crawled pages that Google indexes, which you can imagine would probably be hundreds of millions, at least for here, have rich schema.org terms in it. And they have over 5 million sites that they index actually have schema.org. So it's used a lot. And I was reading in the interview that I was reading through, previous to schema.org there was 10,000 sites along the web that used different markup. So you can see there was a massive increase from 10,000 for basically 20 years to the last three years to having 5 million different sites. So in terms of those rich search results at UDC, these are the different categories that you can actually, that currently have those search results. So we have things like product, recipe, review, event, software application, video, and article. So not everything yet. There's not visual artworks and things like that, but they're getting there. And here's a representation of all those ones that I just showed. So again, here's reviews. Here is recipes. Here's another review. Actually, there's a commerce item, sorry. There's an event. And you can actually see it even breaks down all the sub-events. There's a game, videos, and then there's news articles. So each one has a slightly different representation. But they're all, again, much richer and much more interesting than the static traditional blue headline with two or three lines below it. The other reason that it's quite interesting, as well as the knowledge graph, which I kind of touched on before when exploring Melbourne. So this allows you to traverse your content. And again, the reason that Google can do this is, first off, they have lots of money and lots of time. So they're investing all these resources, but they really want us to participate in this knowledge graph. And the way to do that is, again, well, before I get onto that, I did a search for Henry Matisse. And you can see y'all's artwork. And you can start exploring that. So it is pretty interesting. Recently, we're looking at dogs, for instance. And you can actually sort of choose one breed of dog. And you can actually choose other breeds that have a similar temperament. And you can see how many people are also liking those sorts of things. And I did a search for Australian War Memorial, which I'll touch on later. And as well, they have a rich markup for their site. The way that you can actually participate in this is by marking up your site with schema.org terms, which I'll demonstrate later. So previously, the connections were pretty minimalistic. You basically had one thing connected to one other thing with, essentially, one term. And there's about 400 different terms that you can attach to things. But now, Google is really pushing this to graph notion, because, again, they want to keep you on their site. So what they've done is they're allowing you to further explore this notion of a graph. So here, they're actually saying that if there's a Joe Montana, who is a famous athlete for the San Francisco 49ers, he has a role. And he's Joe Montana, who's his name. And then his start date is when he actually started playing his 1979. So you can see all of a sudden, we've gone from very simple connections to even more complex connections. And this starts expanding over. And I think Google has hundreds of billions of these connections in their knowledge graph. So getting to use cases. One of the key use cases is commerce, of course. So here's the URL for the schema.org slash product. And there's a very deep hierarchy to all different things. So everything has a root of thing. And then from that, we have product. And again, coming back to the Canon 700 camera, I did a search for that. And you get a bunch of rich results back. And so you can see that, actually, JBiFi, Kogan, Expert Reviews, Harvey Norman, and so on, some images. So again, this is all powered through Google. And these are the newer type things. But the interesting thing is you can actually see that Kogan actually gets rich results. They get ratings and things like that. You'll see that JBiFi actually doesn't. They just get the hierarchy at the top, the breadcrumb trail, but they don't actually get the reviews. If you actually look at their content, which I did, they both actually are trying to do the right thing. But for whatever reason, JBiFi didn't check their work. So here is JBiFi. You'll see that Google gives you this nice checker. So you get this big red flag that says Offer. It says Property Offer, blah, blah. You don't have to worry about that. But effectively, what they've done here is they've just doubled up on Offer and Offer. So Google essentially can them from rich search results. Even though they've done all the right things everywhere else, they've gotten 99 out of 100 score. But because they missed that one thing, they don't actually get to have a rich snippet. So it would be a pretty good idea if JBiFi would check that. Over here you can see as Kogan, they have Offer right here. See how it's actually showing a currency price and then the availability. And there's no warning flags at all. So if we go back to here, that's why we get Kogan with the rich search results. And JBiFi doesn't, even though JBiFi thinks they probably should be doing that. That was just, I just checked out about an hour ago as I was finishing up my slides. The next one is food. That's a very popular one with recipes and cooking and things like that. So again, there you can see the URL. Everything has a URL, so slash recipe. So now we've gone down a little bit. So it's thing, creative work, and oh, that's the wrong one. Sorry. It should just be food. Sorry about that. And here's an example of it. And again, we don't have to go into too much detail. But again, you can see slash recipe. You can see date published. You can see prep time, cook time, calories, fat content, ingredients. So effectively what they're doing is it's sort of almost like putting a class, an extra class of things. So item prop equals ingredients. So you're using all your regular HTML things. So it does bulk up things a little bit. But again, there's quite a good reason to do it. And there's lots of automated tools, which of course Drupal 8 is one of them. So I did a search for pizza on the SBS site. And that one looked actually really good. I'm really hungry right now. So again, the other interesting thing that I found was actually that's the title of that pizza recipe for SBS. But for what a reason, I wasn't finding it in the top search results, even if I put in gluten-free, which these ones actually don't even have. So the reason I think that is, is that SBS uses a domain name, as I scroll down. There's still no SBS recipe site or SBS recipe on there. And the very last result, the recipe shows up. But you see it shows up in the news. Again, there's a slight. They're being categorized as a news article, not as a recipe. So they're automatically dropping out of the recipes and showing up on the bottom. So even though they have amazing markup and they've got amazing, I checked their stuff. Everything clicks. But because they use sbs.com.au slash food, they don't actually have a different domain. Google has determined that sbs.com.au anything is news. And so therefore, no matter what they try to participate in, they're always going to be marked against that. So there's some interesting things that have to kind of keep in mind. Google is being a bit too smart now. It's just assuming that SBS is just news. So it's a bit of a double-edged sword, if you don't kind of watch it carefully. Because again, SBS is doing all the right things. But unfortunately, their recipe, even though actually searched for that exact title, did not show up on the very bottom. So they're actually getting marked against for you doing the right thing. So I think that might have to do more with how they rank their actual domain names. And maybe if they would add food.sbs.com.au, they might actually propel themselves back up. I don't know. I don't know if anybody said SBS is in the room. So here's their actual markup. And you can see they've got no tick marks. Oh, there is two errors. But they're minor ones. They weren't actually involved. But that one page that I showed you, it's recipe, name, cuisine, image, image, description. So they've got all this rich markup available. And they're doing the right thing. It's the sign in sbs.com.au is news. And so that's why I gave it a news headline. Even though it's actually giving them. And actually, you'll see that. They don't even get the recipe markup. They get the news markup one day ago. So why would a recipe have one day ago? So they're being marked up as news. So yeah, there's unfortunately their. Yahoo. Yeah, I didn't actually check on Yahoo and Bing it, actually, but it would be interesting to see what they've done with that. The other area of interest as well as the glam sector. So if anyone's not familiar with what glam stands for, galleries, libraries, archives, and museums. Of course, that's a large sector with a lot of information, a lot of really amazing connections. The schema.org tool set, if you are vocabulary, keeps expanding probably every two or three months. They introduce new terms and things like that. Just recently they added visual artwork, which is a huge big win for things like galleries, archives, and museums. And so one of the examples I wanted to show was the Australian War Memorial. They've been working with Drupal for a long time. And they've been doing, marking up their data in very semantic ways for a very long time. But again, they don't yet get those rich search results because Google hasn't released those lenses, if you will. But again, when you talk about the knowledge graph, the thing which actually is essentially even more important because it's all about discovery, the Australian War Memorial and other institutions like that are essentially building that knowledge graph for Google and essentially getting better discoverability. And the example I want to show with them is just slash person. So again, it's a thing and then person. So it's a very, again, a very small hierarchy. And here's an example of one person that I did a search for, which I'll show the data shortly. But here's a nurse from World War II. And within the memorial, they have a rich ecosystem of things and rich connections. And so this lady here, this is her dress, which is on display. There's a photograph that she appeared in. So all these things, this graph is actually being marked up within the memorial. And what the memorial is doing is they're actually exposing all these connections to Google, which we have never have known that all these things are related. So again, you can imagine that this will help power even more things in the future. So things like the glam sector, it is sort of a bit too early almost because Google isn't really doing the right thing by them. But at the same time with the knowledge graph, it's helping to power that. And again, things like with Bing and Yahoo, they're all essentially sucking in this information and essentially what you're doing is you're exposing your schema or your database, all of your fields and all these amazing connections that you have behind the scenes and then get all dumped into straight text on your website. That's kind of a sad thing. You know, you actually have all these amazing domains and things like that. And then you end up spitting out text and HTML and it might look nice, but you've lost all these beautiful connections. And what schema.org does, it allows you to actually populate those rich connections all the way from the database, all the way up to the website. And essentially what your site actually becomes an API because all this information can be scraped, which I'll show you as well soon and turned into JSON, turned into XML, turned into whatever, essentially automatically for you. So yeah, let's get to work. So kind of the key tools that I'll show is I actually have Drupal 8. So I'll show you about Drupal 8 and then also Google has some really awesome developer tools which I'll touch on as well. So of course, with Drupal 8, what could go wrong? Well, that's a bit of a leading question. So here I had my site set up, great, fine. I think it's beta seven is the latest one that's out there. So I had it in place and it was working fine. So I thought, great, I'll just grab the RDF module, which allows you to do these mappings and then put it in place. So of course you get the white screen of death and I was trying to figure out what happened. And of course there's some weird PHP fatal error, like something has changed. So what I had to do is I had to go back to beta four so with beta four, I'm happy again and I was able to get my content types going. So the demo I'm going to show is with beta four so it's a little bit out of date but the RDF modules that I want to show you, which are used for the schema.org mapping are only currently work with beta four but they don't work with any of the other ones as far as I know. Well, I tested them and they don't work. So with that, I'll switch to a demo. All right, so here's that same site. Pretty plain. You'll see there's a very basic thing here so we can just kind of ignore that. So again, we've got a very basic site. Everyone's favorite sort of garland theme. Looks pretty boring, but I'll click on structure and this is where it gets more interesting and I'll click on content types here. And now we get a new option which is add schema.org content type. So it's actually really cool. What you can do is of course you can continue to add content types as you would normally but you can actually build the entire content type of all the fields just by clicking and you get all the mappings done for you. So if they click on add schema.org content type you just get this nice little search box with this little annoying thing. So any ideas of what I should search for? Any sort of content thing? Recipe. So if you start typing, there we go. Recipe. So if I, as I clicked on here, I'll maybe video, medical. So you can see there's all sorts of different nouns that are in this system. So I'll click on, I'll type out recipe again. And then just click next. So it's a little bit overwhelming but you get all of the fields from schema.org sent back to you automatically. And you can tick a little box if you want to include them and it gives you the name of the property and it does a pretty good job of matching it to the actual Drupal data type as well over here. So some of them are a little bit wrong. You can see associate media file, it's good. Headline, audience, things like that image. So it maps up pretty well. You can correct it. So if I scroll down, you can see that. So if I scroll down, you see there's quite a bit, you know, there's rating, there's audience, audio, author. So these are all things related to a schema.org definition of a recipe. Cooking method, copyright holder. So what I'll do is I'll just quickly click on a few of these things. So I'll maybe click on about. I'll click on aggregate rating. Let's see cook time, that sounds good. Cooking method. Description there. And then maybe we'll find image, there we go, image. That should be good. And we'll just do a quick check to make sure. So image, that looks good. Description, we'll go actually text formatted long. Cooking method, text plane, cooking time, text plane. All right, so that looks good. So there's probably, I don't know, 100 fields here, of course, I don't think we'd want to include all of those. I'll click on save. So now what it's doing is it's going in, it's actually creating your content type with all those fields. And you'll see content type recipe has been created. And you get the different fields here, and then now we're back in Drupal land, essentially. So you can go ahead and the first thing I want to do is I'll just leave that, so I'm happy with that. So I'll go to content, and click on add content, and click on recipe. So here's my, again, my traditional Drupal thing. So I'll go with pizza, without cooking pizza. A rating will go five. Cook time, 10 minutes. Oven, and stuff about pizza. So anyways, just regular content type, which allows you, taps into the whole Drupal. And of course, you can choose an image, which I'll just leave for now. And I'll click save and publish. And there's my beautiful recipe. Of course, there's lots of styling, there's lots of extra things you want to do on here. This is very ugly and very basic, and no one would want to cook this ever. But you can see that, effectively, you've got all the things in place for that. So again, there's nothing magic about this. I mean, you could just create the content type and create your fields and you're done. But what it's done is behind the scenes is actually put in all the mappings for all these different things, for all the schema.org terms, which I was talking about, the things like SPS are doing. So if I click on these, and I go to inspect element. Okay, hopefully, so you probably can't see that. But we have things like schema recipe, and then everything that's aggregate rating, and things like that. So all those things are marked up in there. Yep, that's a good question. So what you can do is you can actually go to existing content types, and that's where you can go to, I can go to the article content type here and go to manage fields, and then I can go to the RDF mappings. So that's what allows you to now map, do the reverse of what I just showed you. So of course, you may have already your own content types created, which is most likely the case, because that's what you've already had. And you actually want to be able to then add those mappings to your content types. And so you would just do the opposite of what I did. I actually created the content type and everything automatically. You would just add the mappings to it. So all the same things are there, and all the same lookups are there, but you just have to do it manually to go for that. The module was created by a Google Summer of Code person that was working with SCORE. So Stefan Kulesk, who's based in Boston, he's been brilliant, I mean, he's been pushing a lot of stuff for years now. He had a Google Summer of Code student, she's from India actually, and she did this over the last few months. And so as far as I know, it's just schema.org, but yeah, I'd imagine it wouldn't be fairly easy to actually tap into your own vocabularies for universities, for public sector, for literally anything. Thanks, Stefan. All right, thank you. Yeah, is there any other questions on anything about here? So I've also got a site that's in the cloud. And the reason it has to be in the cloud is so you can of course check it. And this is a visual artwork. So this is one that I did earlier, and it's just a demo site that's in the cloud. The reason has to be web accessible so that you can actually use the Google, different Google tools to check the syntax. So if I just copy that URL, and if I go to the Google developers test tool, so, no, that's not good. Let me try reloading that again. I have screenshots of most everything, because okay, great. So Google's got some really great tools. Actually, let me just go back to the presentation and I'll just give you a quick rundown of those tools before I jump into that. So yeah, I mean, effectively, the only thing that I was doing is I was taking Drupal beta four and the RDF UI module. So it's just RDF UI with no space or no underscore. You can go there, you can download that. Just one module, it's got a sub module, so it's got the RDF builder, which I was showing you. So you just install that, and then effectively you get all the things that I was showing you. So it's kind of nice, it's just one module. And that was just stock Drupal. I didn't even turn on any modules at all. So out of the box, stock Drupal is using schema.org terms for the built-in content types and for things like that. But of course it would have no idea if you're actually creating a recipe or a commerce item or a car or whatever. And that's where the RDF UI builder come in as it allows you to define your content types. There is schema.org module for Drupal 7, which does very similar things. Yep, but there is, because of how Drupal 7 was implemented, it was implemented well before schema.org was out there, and it had sort of an RDFA implementation. So there's some issues with the markup that it produces and it's not as good as with Drupal 8, but there's workarounds for that. So yeah, Drupal 7 can participate. The only reason I didn't include that is just for time. But I was gonna actually, if I would have had a bit more time to show a Drupal 7 implementation with the schema.org module. Again, with Stefan is also a score, has put that together. So essentially anything semantic web or schema.org knowledge graph related in Drupal, it's score who's involved in it in some way or the other. And so here's some other tools that I'd like to show you. So the other ones that I want to talk to about are the structured data markup helper. So I'll show that one first actually. That one's really, really, really cool. And the other one is the testing tool, which is the one that I was just gonna show you. And then lastly, Webmaster Tools. So Google Webmaster Tools, which probably everyone's familiar with and has used before, now also has, if you go to Webmaster Tools, site appearance, structured data. You'll actually get sort of a 10,000 meter overview of actually all the structured data in your site and the different content types that it's found and maybe any errors, things like that. So it's a great way of letting you see sort of your whole pool of data and your pool of rich content for your entire site. So again, Google is really, really pushing this and they're really creating a lot of tools for it, which is great. So let me go to the structured data markup helper. It's a monthful. This one is really cool. Let me just find it. There we go. So hopefully we can see that. There we go. So this is nice, because what it does is it allows you to sort of step through the process of marking up existing content to get a better understanding of what's happening and how you can mark it up and doing it the right way. So it does, it lets you, and there's also things like email. So Google also, schema.org term, also you can put in emails. And if you use Google, you actually get those little buttons for actions. So you can RSVP things, whatever. That's all schema.org as well. So if you're booking a flight and you maybe wanna check in, if you see that little check in button on Google on Gmail, that's all through schema.org and you can check it here. But for instance, there's things, these are the major categories. So we have articles, local businesses, restaurants. So I'll choose restaurant, because that's an interesting one. And then any examples of Melbourne restaurant that I could try? Anyone? Anyone local? Any restaurant? Gringo Vibes? Vibes. That one? Great. All right, so there's Gringo Vibes. You can see it's a nice looking site. I wanted to be nice, I wanted to be nice. All right, I have to find my tab again, sorry. Where are we at? Was my tab, sorry. It's crazy. Okay. Sorry, I lost my tab. Did I override it? I just wanted to come over here. Okay. Okay, here we go, we're back. Restaurants, and I'll take Gringo Vibes. Again, I've never done this before, so I have an example one that I did use earlier. So I'll just paste that in and click on Start Tagging. So what Google is doing now is it's actually going out and grabbing that site. It's loading it in. This is a really cool tool. It's really cool. So here's the site, and here's on the side, structured data. So currently there is no structured data yet. So what you do is you actually start navigating, hopefully there's some text on here. See if it's any. Oh no. All right. So there's no text. So they've already lost the battle for Google. So we can do low, we can do image. There we go. We can get image. We can get a, it's not gonna like this. I'll try a phone number. I don't know why. It's gonna, oh yeah. It's not gonna like this one. So they didn't do too well. So let's try a site that actually has a little bit of text in there. S-A-R-T-I. Is that this one? Or this one? Sartita.com.au? This one. Great, there's some text here. I think I can work with that. All right, let's try this one again. Live demos never work. So I'll go back to enter page. Keep the restaurant. Start tagging. And this tool I think is quite new. I had actually never even heard of it before. I was just doing some research into different tools. So great. So hopefully, there's some text here. There's some text. Perfect. All right. Here we go. So I'll highlight that. And I'll say that that's the address, the street address. You'll see there it's showing up over there. I'll click on Melbourne, and I'll choose locality. I'll choose that, and I'll go to telephone number. I'll choose that, and I'll go to email. That's not the full email. Anyways, you get the example, you get the picture. Email. There we go. I'll fix that later. And then maybe we'll use that one for their image. So what it does, it lets even people who aren't semantic web heads and Google data individuals kind of participate in. It's not perfect. Let me see if they've got it. Yeah. Yeah, looks like it. I picked a random Thai restaurant, and it worked perfectly, of course, so I ask everyone. So I'm going to go to that random Thai restaurant that I picked, and I had no idea if it was going to work or not. And it worked really well. Where is it? Anyways, I probably don't have time for that anyways. But you can get the rough idea of how it's working. So you get telephone number, email address, and things like that. And you can click on create HTML. And what Google then does is it gives you back the exact HTML with what you actually need to add to mark up. So where are we at? There we go. So wait. It's a very terrible HTML for this site. You see that there's only a small section. So what Google has done now is it's actually added the exact schema.org markup that this company needs to add to their site. So at least to say that their address, their locality, their telephone number, their email address, and things like that. Of course, you would have to clean it up a little bit. They can further add that they relate what sort of food they offer and all sorts of things in their opening hours and closing hours. But it's pretty cool. And you can also then click on the JSON link data. And you can get a JSON representation of that exact same content. And then you can, of course, download that. Or you can enter that into your own site. So it's a really helpful tool. And the Thai restaurant that I randomly tried worked way better. And next time I won't ask any questions. I'll just use the one that works. And then the testing tool. I'll show that one really quick as well. I've got some of those preloaded. So I got all the URLs. But what I did is here's the JB Hi-Fi one. And here's, I just copied and pasted in their URL. And over here you can see that it's a product. So it's already knows that it's a schema.org product. And you can see that there's the one error. And the really cool thing is you can actually click on each one of these sections. And it goes to the HTML representation on the side. Sorry, you can't. This is gray. It's not, this projector is not very good. If I click on offer, it brings you right to the exact error of why they're not getting the rich snippets. It's because of this section right here. If they fixed that, they'd get rich snippets. So you can copy and paste any HTML in here. It was amazing. Literally, nine out of 10 sites that I pasted in here have rich data. Here's the memorial, for instance. Here's Frank Lee Beckett. I'll just put his name in, or the URL for him, that the memorial has. You click on fetch and validate. And that's not a very good one. Here we go. Again, stick to the script. Works better. Here we go. Captain Vivian Bullwinkle. She's the one that I was showing earlier. If I just, I've already, yeah, here we go. If I put her name, the URL for the representation of her on the web. Click to fetch and validate. You can see that person. You can see it. There's a few warnings and things like that. But there's actually an amazing amount of data that's in there. That's all about her. There's image, person. You see her birth date, death date, description. All of this is available as JSON. You just can take that URL, put it into a linting service. And there's also a nice, there's a lightweight one. So here's a structured data linter. You can use this as a service to get JSON out. So you can just put it in as you can get RDFA, JSON. So here's an example of what a rich snippet would look like if Google would finally put glam rich snippets in to put an image and information about her. And you can see here's a ton of structured data that it's pulled out of that. So the Australian Warmer Whirl has done a massive amount of work. All this is available. You can just grab that as JSON. You can start creating all sorts of things. And with linked open data, which is a whole other area, you can then follow that graph and walk around it and do all sorts of interesting things. So again, that dumb web page here actually has a lot of amazing real structural data that's in here. And these are all the same terms used around the world. So there's no ambiguity. That's the other beauty of it is that this death place, this death date, the birth date, is the same birth date as they would use in the UK or in India or anywhere else. So that's what also creates that global graph of information. And I think I'm running out of time. So I'll, sorry, I didn't leave any time for questions. There's Markup Helper testing tool. And there's a whole bunch more tools, which I didn't get to run through, Ruby command line things, and questions. Yeah, there is. I was watching, the question is what happens when there's a difference between what I'm saying about my site and what Google thinks they have of your site. So I was watching a video on that and the product owner for the Knowledge Graph said that they've got complex systems in place to fix that. The reason that's there is so you don't game the system and say, you know, like that you're a Thai, Mexican, you know, whatever kind of restaurant, and actually you're a Thai and you've got a one out of five rating or something like that. So they do check, but they are still wanting, you know, this rich data. So they were, again, they were very open about everything else, but for that one thing, they just basically kind of said, you know, we have complex systems in place to prevent gaming system. Any other questions? I realize we're at one minute over. If you're using Drupal 7, you download the schema.org module, put it in place and start adding these terms which has very similar look up and things like that. It works really well. Add it to your site and then use those checker tools. Use, and even if you don't want to do that, you can actually just go to that Google tool that I showed you because everyone here would create better sites than these rubbish ones with lots of text, which is good. And start adding the right markup to your site, testing it. So Google is great because it lets you check all your stuff and know absolutely if you're doing it the wrong way or the right way. Previously, there was no tools like this. There was no tool like that structured data one which took you to the exact line and said why you're doing it wrong. That's there now. You can fix that really quickly. And because you're working with a template system, you fix it once and you might have 10,000 pages that all start working better. And as you see, Google webmaster tools as well is also giving you that bird's eye view of your content so you can actually see the health of your site and how much content is being indexed. If, for instance, SPS is getting the recipes marked up as news, that's a problem. You can actually see that in the Google webmaster tools. You can see that your content is all being marked up as news, whether it's sports or whatever. So doing those little things allows your site to better participate. And if you use some of those existing 12 different ones, you'll get rich snippets usually within a few days and sometimes it might take a bit longer for other sites. And then they release new tiles over time based on popularity, essentially. So the more sites that are using this stuff, the more Google sees it, and the more likely they are to release a smart tile versus just dumb results. How so? Yeah, that was also one of the questions that I tried to follow. And they didn't really give a good answer to that one either, so I don't know how it works. I've done everything with English-based things, so I don't know. The question was, how does it handle multilingual? There's four different titles or something in different languages. I'd imagine they'd use some sort of disambiguation in terms of knowing that the canonical URL is this, and then all the different language settings would allow it to be searched better for other languages. So yeah, sorry, I don't know. I just thought English is the only thing I ever worked with. All right, thank you.