 Okay, so This is building cementing content models in Drupal 8 This is Stefan Corlos get I'll let him introduce himself Hi everyone Stefan Corlos get score on Drupal.org have been a long time contributed the Drupal project in Both core and contrab In the realm of semantic web RDF RDF a schema dog And I also happen to be a member of the security team and I work for Acquia. We both work for Acquia So and my name is Kevin O'Leary. I'm the director of design at Acquia on the Octo team I you might have seen designs that I've made like the Drupal 8 toolbar edit in place whizzy wig responsive preview several other Acquia products as well and Score and I have been working together for a while on collaborating on these you know problems around content modeling and and and Semantic content and link data and kind of bringing together the two worlds of UX and You know and link data and schema.org So this is this is a quote from Oliver Wendell Holmes, which I thought was really interesting. It's not really about Semantic content or about the internet or or any of that But I thought it was really fascinating in in in relation to what we're talking about because it's about thinking in things You know and that's essentially really what? What schema and and an RDF and semantic is all about it's about Attaching the meanings to the things and making sure that they that they are that they are not lost And that we always are rooted and grounded in the sort of in the realities of what people are actually Doing and what and what their real needs are and that we don't sort of let the language get pulled apart from or separated from either Physically, you know or or semantically from from what it's what its true meanings are so With that You know, I'm going to talk a little bit about Content modeling in the semantic web. So these two worlds. I don't know how many people saw Nazar Bena's presentation yesterday Okay, so so a lot of you saw it that was Naz made a really awesome You know presentation about you know about that the way that this whole world is changing and I think you know one of the overarching things that we saw in his presentation is this sort of this overlap and sort of marriage in a sense of of of content content strategy content modeling and semantic web and link data so You know what we're hoping to do is to try to take that further and to talk a little bit more about how we do this with Drupal so So we're going to talk about Kind of two main buckets. The first one is semantic content model creation And then the second one is semantic content creation So on the one hand we're talking about essentially kind of field UI ish kind of stuff and the other hand It's kind of when you're in the whizzy way when you're when you're inside of the body field So Stefan is going to talk a little bit about the history of the projects that we've been working on and you know Some of the things that that we've accomplished with that So you might have seen the video that Dries projected Back, I think it was in Austin. I did in his keynote. This was Guha giving his stake on schema.org and the adoption of schema.org by Drupal so just to recap what happened here before is When we started to work on Drupal 8 schema.org was launched in 2011 that's a few years back already and We decided this this was the future. We realized that that was the future of the semantic web and how you Annotate your structured content that you publish on the web. So search engines were behind this initiative. So we thought We definitely have to adopt this standard schema.org and Guha is the creator of the schema.org initiative. He works for Google and We had an interview with him. That was that's on YouTube So we'll have the link in the slides if you want to watch later, but most of you might have seen it already in Dries's keynote So schema.org it's a it's like a dictionary or a schema to tell you what terms and how to express your content, but it doesn't Tell you what format to use. So you have to pick a format what we use in Drupal 8 And we've been using that format in Drupal 7 as well It's called RDFA and its additional attributes that you add to your HTML output that tells search engines How to understand your HTML? How to break it down? What type of? Content that is described as an event or is it a news article and then all the values inside it What are those at those dates or those people authors? So this is describing your content really well to search engines And then they can pick those up and display your pages as rich snippets and make them more attractive and search Engine result pages. This is an example of an offer from using the schema.org Vocabulary and it's got a name a price for for for example, this is another offer a bit more detailed it has Reviews only one here for the sake of simplicity review rating best rating etc. How many how many rating got and schema.org has a lot more types and Properties than just reviews and offers it covers a very wide range of topics That most of you probably need on their site and most people need on the site. So like recipes Books news articles Physical items like places location countries Offices building like government buildings public buildings. So any anything that People published on the web and anything that Google is interested in understanding and surfacing and search results In triple seven we we didn't launch triple seven with schema.org support simply because The launch of triple seven happened way before triple before schema.org was launched, but We we still managed to backport some of those or to integrate some of those Vocabularies from schema.org using contrib Using a few contra modules. So the first module is is called schema.org and it's a distribution called schema kickstart But this provides you with the UI where you can go and map your existing fields and Existing content types to schema.org types and properties So if you have a lots of content on your site, but you don't have those mappings You can go and install those modules and have the mappings out of the box for All the content that you have on your site if you map things properly And we also have another member of the community who built a module called entity field builder And that was a nice initiative from his part to come up with To come up with an interface allowing you to Spin up and instantiate content types from scratch Reading the definitions on schema.org. So say you want to you're starting a new site You want to have events on your site Instead of going and clicking around and creating every field after creating your event content type You can instead go in in this interface and say I want an event Content type on my site. I don't have anything yet Please just create me an event content type with all the relevant fields and all the rip choose the field types for me Etc etc. So that was a very Bold Idea, but it's a great idea. We think so we built on that and we have a demo of what's working for triple eight Coming up in a little bit So So on the you know on the on the other end of the spectrum, you know, we saw we've seen over the last few years This is diagram from from Karen McGrane from from the MPR content model We see this sort of the world of Content modeling, you know coming into play at the same time and and over the last sort of five or six years This is really sort of exploded into a whole a whole new A whole real world of content modeling for the web and Basically the the central idea is this idea of content Create once publish everywhere, which is quite similar in a sense to the to the dry concept of don't repeat yourself that we have in you know in in development and You know you what you really want Is you want each atomic piece of content to have only one canonical version so that if I have You know if I have an author or a name or a or a place or you know Any individual piece of content that that exists one place and can be updated one time and that that will You know then flow out to all of the you know all of the places to which it's linked So having these You know having these well described and well structured You know content models that Drupal enable enables us to do as a real sort of accelerator for this, right? so Stefan is now going to show you a little bit of how You know the semantic content model creation Happens in Drupal 8 but essentially what we're talking about is you know these like I said before We have these two these two personas on the one side. You have the site owner Or this could be a sort of an editor or a site owner or a site builder But essentially it's the person who is responsible for modeling the content for for for creating for creating the content model And then you know making it a reality in the site and then on the other height on the other side We have the content author and that's what we're going to talk about a little bit later on and there's some interesting modules that have been That you know we've been working on for that, but on the site owner side You know this this is a busy individual who has lots of things that they have to do in their day and lots of sort of management responsibilities, so What we want to do is really kind of Make things much more efficient for them You know they don't want their time wasted and in fact currently in Drupal You know a lot of the a lot of the things that we we do take a lot of time And there are a lot of sort of excessive numbers of steps and UIs that you have to go through to do things One of the things that we realized is that if we can if we can save people time in the content modeling and content sort of Content model creation But have the RDFA sort of built into that process then then the link data comes becomes almost automatic And that's really kind of the the underlying goal of this So you know so people don't have to think about the RDF They found a good place that they can get an easy way to create a content model and and then the RDF comes along with it for free so Stefan is going to get into a little bit of this This this problem space and you know essentially the first problem is when I create an entity I want to start with same defaults right and The second problem is you know I want to I want maybe a whole content model essentially more simply give me a Semantic entity template and then give me a semantic Entire content model template so we're going to talk about the first one first, which is what we've really worked on and So this is the Drupal 8 RDF UI you want to yeah, we have a video for this Well, we had a video Oh There is okay, so this is all work that has Been done. Sorry. This is what I'm going to show here the videos I'm going to show is actually done has been done by a student from the Google Summer of Code over over a period of two years So the first video was done last last year summer 2014 and So this is This is it starting here so we're going to skip this step here, okay, so she's installed all the models etc and Creating she's creating a content type in this case it's of Type person and we can see her as you create your new content type You can map it directly to what it corresponds to on schema.org using auto complete Save your content type and just to show you There is this is equivalent to the person on schema.org and there's more there's a lot of properties that belongs to a person a lot of Properties you can have for a person such as the age did a state of birth Friends name first name etc So there's here we pick affiliation we create a field for the affiliation of a person in Drupal just using their regular field UI and in the settings of the field at the very bottom of the form We just click save so we just continue doing doing this in the Drupal server version. We had We had a slightly different workflow So here you just create all of your fields and then you go to a separate page dedicated to the mappings of Your content of your content type so it's called already of mapping in this and this is by the way This is triple eight. I just want to remind you There's an equivalent modules for triple seven as well the UI is slightly different But this is the newer newer version So we have the list of all the fields that we have in the person content type but the affiliation job title and We are going to map those fields which so far we're just local fields and Google has no idea what what they mean We will map them to something meaningful that Google will understand based on the schema dog Vocabulary and definition So Baldy Baldy will map to description those those values by the way in the drop-down are all coming from schema dog So you you know that if you pick one of those you got it, right? There's no chance for typo here job title So now we save the form everything is set and if we go and create Content of that content type just gonna get that step here. We can look at the Markup So this is the normal page the page that Google will see when it crawls the site And I don't know if you can see here. Well, but there are some some specific attributes that were added such as Schema affiliation. I don't know if you can read here, but this is the mapping we chose in the UI So all of a sudden all of your markup will have those additional attributes Which are really useful for Google and any search engines to understand your content and Pick out the values that they like and they want to reuse Same search results like this. So here we have the affiliation The affiliation all the values that we have here affiliation job title All of that is extracted by the Google parser and Google can do Many things with that. So now there's like the next step after this Instead of creating all those fields manually. Why don't we have a model that does that for us? We just pick a type we pick a few properties from schema.org and then we we go ahead and we let Drupal do all the hard work for us so Instead of clicking add a content type like we normally do we just add a schema.org content type and this is the list of all the types available on schema.org and Here we will choose something Sport related so sports event Just to make sure this exists here that there is a sports event type on schema.org. This is the one we just pulled Okay, so those are all the fields that we downloaded from schema.org Those are all the options all the fields we could choose to decide to include in our new brand new automatic content type We're gonna pick image Location here in the drop-downs on the right you can choose a different field type if you don't like the one we picked for you Start date for an event that makes sense That's also in the list of the fields. We want to create automatically for us and there we go So we have all the fields here machine names We used a little namespace in front here just to avoid any Conflict with existing fields that you might have on your site already All right, and then the order of mappings are pre-filled already. So that was very quick We we built one new content type and a bunch of fields and just a Few seconds, I guess it was a bit longer here just to show you but You have to do that every now and again Very often. It's very quick to do it. All right. So everything is set And then from there everything is back to like what you've seen before we can go and create a new sports event Here there's no difference. We're just gonna use an example From Wikipedia Okay, let's keep that part Copy paste. Okay. The node is there It's been created And it's ready for indexing by Google we can check that the markup is there Exactly what we wanted type of schema sports event There you go and all the fields and if we check on Google Just put our page. All right gets updated Date is there and location. So we're good. Everything is picked up All right So that's the end of the first video I'll switch back to Are you caving to believe? So So this this is all the results of you know a lot of really hard work by both Stefan and and and Saccini and you know We really Our goal was to make it as simple as possible. So we've really made the process of of entity creation much much simpler you know, you know in Drupal and The this way of being able to automatically download an entire entity and then and then set it up is kind of what we were Striving for and one of the key things about that is creating as Stefan showed you, you know The default field mappings so that if you wanted to you don't have to reconfigure anything You can just go click click like these three fields and done Because every field now has a default field mapping which maps to pretty much an 80% use case You know what that field would would generally be used for and it's this concept of you know Of going with the 80% use case in the same default and getting something into the site as rapidly as possible That makes this an accelerator for the site builder and again Gives them an incentive to do this even if they're not really necessarily interested in linked data or schema.org Or any of that although they should be So but what about the entire content model right what we'd really like to move to Is an entire you know content model? Template system, so this is an example site. You might have seen this in Andres's keynote. We use it for a Music site a sample music site It's a it's a it's a it's a sample content model that I've been using as an example for a while So imagine you have this music review site with You know with music reviews and it needs a music review content type obviously But then it also needs you know a several other content types So it needs all of these you know like album review concert artists venue song album genre and then you know in Drupal Obviously these these all need to be These all need to be You know different types of entities so some of them are nodes and some of them are possibly going to be views Some of them are going to be maybe even you know a media file or a taxonomy term And you need to make you know what again are the sort of sane default choices for for what those things are right now to do this You know it requires really an immense amount of work to get all of this all of this this stuff done and You know this is assuming of course This is assuming that you didn't use a you know a pre-created schema org You know content type that you that you Imported but even if you did you still have to create all of the entity reference views You still have to create and then configure all of those entity reference views and then you know essentially You know link together all of these things and connect all of the dots to create this entire content model Right and if we could make that simpler and and go that even one step further Then we have even more of an accelerator for for users for the site builders to come in and begin to create You know fully semantic content types So this is this is not a good process right now and we can and we can improve it and in doing so You know make it even simpler and and get those people a better experience So part of this part of the idea behind this is is the content model as a feature So if you know if we do all of those things, right? We can we can begin to You know take these schema org linked content types and bundle them together using features module so You know, I've whenever I mentioned features module a lot of people say this You know, this is actually a quote from the features module project page on the on the Drupal 8 version of features And I hear this all the time. Why do we need it? You know, we have CMI now, but the the original goal of features module was to create features so This is what we're trying to do with it now We want to create an actual feature not just a bundle of configuration that I want to transfer from my dev to my stage my prod Instance of my site I want to actually make a feature that's shareable with other with other people that I could put up on a feature server going back to that original dev seed kind of apps kind of idea that that it's a real You know a single thing that has that can be that can be downloaded and installed it kind of in one click and run and You know as a as an accelerator for content. This makes a lot of sense because you know It's it's more of a You know, it's more of it's more of a template then sort of a then then a thing that has Business logic embedded in it or or or some particular Industry vertical like a distribution would we're talking about just the 80% use case content model for a sort of domain like music or Film or art or you know, whatever this the particular domain that you're talking about is that shares this sort of 80% use case Content model and out of this you can create a kind of a content package That's a feature that could be could possibly be hosted on a feature server and download it as a Sort of one thing that people could start start from right so features module in Drupal 8 has actually come quite a long way and I I'm not sure whether they have a beta release yet, but but they're but they're doing a lot of work on it and it's going to be Really great. I believe it's going to have the ability actually to To not that you can download a feature and that you don't have to have All of the dependencies on on the site and it can immediately run the feature without without any further You know downloading and installing of other modules So, you know the future of this actually You know when we think about it, there's there's there's more really that we can do there's more that we can do with content creation and and adding to this whole sort of experience of You know allowing the the the content model Person, you know the person who's creating these these you know these these these content models to do not have to go through All of these steps because you know each each one of these boxes here represents a new UI in Drupal that you had to go to That's ridiculous You should be able to do this in a single in a single action or in a in a simple kind of in a simple UI That allows you to do it all in one place So when if you look out on the web and you serve search around on content models and you know content strategy and you know And and modeling tools and diagramming and and things like this This kind of stuff has been around for a long time data diagramming databases and entity Relationships and there's a lot of different ways that people do it, but the overarching You know concept here is that when I model my content I can see the whole thing I can see it at a 10,000 foot view I can see all of these entities and how they relate to one another But field UI in Drupal takes you through this process where it's like this I go to this UI I go to that UI I go to the other UI and there's no single place where you can create a mental model of your entire content You know structure of your site so that you can so that you can really easily create these things So one of the things that we're moving towards one of the next projects that we're hoping to build Is something that looks kind of like this. This is an initial sort of first-pass design where you can see that It's bringing together, you know a UI with a You know with a with a content modeling tool so that you can literally grab these entities and then bring them in and then And kind of drag and drop sort of you know the connections between them and then determine whether that that connection is going to be You know a one-to-one one-to-many is this going to be? You know a child of you know all of the different kind of ways that you want to you want to make these connections, you know In the directionality of them So or I should say cardinality of them. I don't always get the developer terms right But you know and then another thing, you know that that we really You know need to start to think about also building into this kind of a tool is a more automatic way of Making of making these links to External APIs right so if I have my music, you know genre List sorry, I'm gonna go back to if I have my music genres. I couldn't make that as a taxonomy I could go in and you know have the people in my and my content team You know start typing in you know jazz classical pop et cetera et cetera et cetera et cetera But why would I want to do that when I when I could go to you know a standardized? API like this one echo nest for instance that I can get a developer key to and I can get and I and I can send You know a rest call to them with rest in Drupal 8 that we have now and I can make a call to this API and immediately go and get Based on artists give me the genre You know or based on song give me the genre based on you know a number of different criteria So there's there's lots of different get requests You know that I that I can make and I can just get that thing pulled directly into my site into this You know into this piece of content now, you know having those kind of things baked in that's an even more powerful accelerator because now my taxonomy is not going to get stale and out of date and And again, it's dry and it's and it's and it's and it's create once published everywhere I'm taking that John from one place where where somebody else is maintaining it I'm offloading a maintenance task on to a you know a public API and essentially really this is available for all of these things So then my content model becomes much much much simpler to maintain in the long term Because all I have to do all my content creators have to do is create album reviews because that's what my site is all About I'm getting my artists from ASCAP. I'm getting my albums from Spotify my songs from Spotify tickets from Ticketmaster Venues from Foursquare etc etc etc, and this is true in any domain There's there's API springing up all the time that are serving up these kinds of You know of enumerations and and and and you know You know bunches of content that you can that you can suck into your site and really use So that's kind of where we see the the future going with that now We're going to talk a little bit about the content creation side And this is again. It's about the content author and getting into that You know into the whizzy wig So, you know, this is the person responsible for you know researching and writing the article and You know they don't they want things to be automatic, you know They don't want to have to think about it And the other thing is that they don't want to be taken out of flow, right? So they want to when they're in when they're in their editing experience and when they're actually typing They don't want to have to go out of that browser window into another browser window type something start researching something and then have to come back And if you look at a lot of the enterprise level tools that are out there What a lot of what these tools are doing is that they're bringing the research functions in context They have search windows on the side where you can literally start, you know typing things and researching or even having your Your your your your content parsed for entities, you know that are that are that are real things like Marlon Brando or you know Frederico Fellini that thing becomes something that gets automatically parsed and and and that you can link So Stefan is going to tell you a little bit about some of the background in this area And then we also have a video to show on that So about the background the previous work that has happened in this field really there isn't that much of it a lot of the focus that we've Had in the past many years was about content model not about the actual instance of your content So in other words, we were able to tell Google that this taxonomy term is a place or a city But that's really all we were able to say we were able to say the name of the city is say London Google might assume it's London in the UK, but really there's no way to tell the search engine that no it's not London in the UK It's London, Ontario in Canada So that's the problem. That's a challenge. We're trying to address here We haven't done a lot of work in that field. We we can't map the instances of your notes to a particular entity in the in the real world and that's because Google wasn't really as far as I can tell they were they were not as interested in that they were In the past years it has it has changed over the last year or two with the knowledge Knowledge graph announcement and then the week free base acquisition of by Google and then the wiki data movement from the wiki media foundation and wikipedia people so this aspect of Annotating your content itself not just what type of content it is, but what is it really? What if you have a name, you know, it's about disambiguation So the only thing we could find in the past that has been done about this is this early FA content editor It's a whizzy wik plugin for tiny emcee where you can go and Tag annotate your strings And that's really all there is to it in in durable so so But at the same time there's some in there's been some interesting work going on, you know in other areas On the on the usability side of this in terms of being able to while you're in flow while you're in an You know an editorial experience to be able to To be able to you know get get into a place where you can actually search, you know for other things now Yeah Okay, so this is this is something from the MIT Media Lab. It's called MIT. It's called fold It's actually a it's actually an online magazine so I'm in the actual content creation flow here and You can see what I'm doing is I'm creating this little card where I start to type something and then I can hit a plus button I have this tool over on the side where I can immediately search things and add them to You know to my to my article right so I'm I'm I'm writing an article on Bob Dylan So I'm one of I saw first I found a video and now I know that Bob Dylan was born in Duluth So I'm gonna link to the map of Duluth. I'm gonna link to Immediately again put a map there I can go on and I can do things like I can you know, then I then I Annotate that as his as his birthplace and then you know do other things like you know search for images or or even even Animated gifts so you can see this last one is You know, there's all these there's all these little animated gifts of Bob Dylan singing So, you know, this is this is really fun And it's a it's a way for sort of searching for things and adding things to content in context without getting out of context But it's not necessarily it's not necessarily semantic web. It's not really I don't know whether all of these things that are being added are necessarily You know marked up with RDFA or whether they're really they're they're really You know going to be discoverable apart from the fact that they that they're connected By by by virtual being in the same piece of content But so it's it's it's like we have these two areas of You know of of the link data and the and the user experience sort of moving towards one another and and we're and we're Trying to pull those things together. So what Stefan is going to show you next? is Yeah, so again This is work that was done by the same student this year at this time. So here. We're just going to Enable a module called the link data tool and what this module will allow you to you eventually to do eventually is to map and give a node or a given term of your site to a particular ID from a well-known authoritative bank of data or Tizoris such as wikipedia's wiki data or Google's freebase So that when Google comes across that page and sees that this is really the same Entity as the one that I know already in my system then Google or any search engine can map those two and say they are the same thing And that's useful again for avoiding ambiguity on the web So here. We're just creating a field creating a new field to to our content type and Again, I repeat this is triple eight. This is a very prototype level and But we're still working on it and polishing it So we have the field now. It's just it works like a regular field and when we create content a new article Say about Drupal Just gonna fill it up fill up the form Add a tag. So that's just a regular tags field. So don't worry about that. What's interesting is the last one here At the bottom. So you type Drupal and it's it does a another complete request to freebase and There is your tag. So not the ID here q1 It's already gone, but q1 Something that's that's the idea of the concept about Drupal in in wiki data We support two providers of tizari right now wiki data, which is the next generation wikipedia if you wish Which is here. So here's the page about Drupal on wiki data and you see the oops The idea here. So that's really the end of that first video so this was about again to to recap this was about simply Mapping a node or any entity on your site using a new field type to an ID on freebase or wiki data so so so So obviously, you know the the the experience is something that we're still sort of sort of moving towards right now We have we have one field where you can where you can map these things to to the wiki data or to to freebase But ideally what we want to be able to do is is have You know sort of more more of these search tools sort of built right in kind of alongside your your editorial experience So that's the next step You know with this tool is to is to do things where where the user can begin to in in a similar way to fold start To search for entities You know and then and then actually and actually pull them into the into the content and add them things like videos or You know or audio or or whatever? You know they may want to add to their sites And and have content that is you know that is that is rich and actually richly and properly linked to all of its canonical you know sources and You know when if you you know, I've done a lot of you know talking to people in the publishing industry at for instance at time Inc. And and others and this type of ability to sort of Immediately be able to research and link and and and create these cross references In your editorial experiences is a highly sought-after thing. So we're really kind of you know Going going in the direction that the user sort of is driving us and again creating an accelerator for For link data by by making the user experience of creating content and doing research easier So so the future in this area. I think is is very interesting because you know Essentially really what we want to do is we we don't want to force authors to to mark things up, right? So, you know when we talk about getting into the body field, right? When we when we bring things like what we're talking about like this like this entity browser into the body field where you know Where a where an individual? You know entity in the middle of a paragraph, you know sort of Marlon Brando is written in the middle of a paragraph Can be can be selected and then that can be immediately linked to you know to to its wiki data source or freebase source That's that that's that's kind of a kind of automatic power that you want it We want to give to users and if you look at you know Google sheets For instance, you can already do similar things to this with If you type a link if you have a cell in a in a Google spreadsheet and you and you type a link into the cell If you type a name into the cell it will pull up into the into the link field It will it will pull up the website that it thinks that you are trying to link to and you and you can add that in That's the kind of automatic Automagical kind of functionality that you really want in these kind of situations so that so that everything that the author Needs to do to make their their their content semantic sort of is kind of provided for them Right right where they are and indeed all of the other tools of making the content semantic like creating Sections and you know and citations and headings and etc. Is also all you know kind of Automatic so this is some you know work that we've been doing on improving the authoring experience and You know it's kind of trying to bring together these two worlds of sort of seamless smooth You know distraction-free editing like you might find in a tool like medium with with with with semantic tools that surface themselves when you need them so Don't just give me sort of an you know a effortless authoring experience Give me an effortless effortless authoring experience that exposes the structure of what I'm doing So that I don't have to dive into the markup Or have the knowledge to dive into the markup to to make that happen and And if we can do that if we can keep authors out of the markup It makes the job of of of site owners much easier because then we don't have to worry as much about things like filtering and And all of the problems that we run into Inline styles etc when when authors are looking at the at the HTML view so You know and ultimately going forward, you know, I believe that it really should be the responsibility of the CMS to In a sense interpret the intentions of the author, you know, so that if the author adds And you know, you see this in a lot of markdown system if systems if the author Types a one and then a period that automatically becomes an ordered list That's the kind of sort of automatic You know sort of interpretation of an of an author action as a key to what the semantic structure should be Right and I believe that you know as we move forward, you know We're going to get closer and closer to this ideal where where what the author does, you know Is actually underneath the covers Creating a semantic structure that that makes sense out of what out of what they're doing rather than having to sort of Impose it later, you know or have some other kind of structure to you know to impose that and it's really kind of the smart CMS Which is you know, what we'd ultimately like to to move towards so you know I the takeaways are kind of you know labor-saving tools Can be can be a catalyst for for creating You know this world of linked data that we that we you know believe will really be powerful for everybody And you know as as long as these things can can become kind of automatic that that that people don't have to think about it That that's the way that that this is going to gain traction And that we have the the first level of tools which you've seen and soon hopefully with that with some more more time and effort We're going to have some more powerful tools as well You know and and this future of you know of a smarter CMS and a more you know a more Automatically semantic CMS so The last video oh you have another video yeah, so we took a stab at what Kevin has just been describing where Instead of mapping a whole form you can go in the body field and Map each of your strings or concept to something that search engine will know and I will just show this demo now I didn't even know you had that video Was a surprise out of his hat surprise for the I was just working on it here. No So again credits goes to Sachini From Sri Lanka for working on this during the last Google summer off code. Here we go So we have the this is just the regular Text format configuration interface that we have here and You can see some buttons here Those are two new buttons that we will add to the triple eight core whizzy wig interface So one of them is for tagging and Piece a piece of string and the other one is just to remove The link so this is the way the UI works You have a list of options and you drag the one the ones you want to use In your whizzy wig editor and all the htm attacks will get updated automatically for you So you save that and then you can go and create the article of your choice I'm just using some content here. This says judge judgment of Paris. So this is just a text that I took from Wikipedia but what's What's interesting here is that I can highlight some strings and Map them to Wikipedia and it's actually automatically detecting the string and querying Freebase for the ID matching that string. So those are Those are this story is picked up from the Greek mythology and those are Personalities from the from the Greek mythology. So this is Paris. So this is actually a Concept of Paris in the Greek mythology, which is not the city It's just a fictional character from a play and here we could actually Say when we type our text and when we annotate the string Paris that it's not the city that you might think of It's actually something completely different that you might not even even know about so We've tagged all our text here We're just doing another one. This is Apple So again, not interesting thing here is Apple could mean anything could mean a lot of things, sorry And here it's the fruit. So we made sure that this piece of text means The apple as the fruit. So we're going to save that note now and those strings all appear as links and As we will see shortly those links will go to Freebase, but let's first take a look at the whizzy way after if you want to edit your note You still have the same interface. So you you can clearly see this is a link data Tool and here in the markup. We have links to freebase in the whizzy weak source so let's just go back and click on one of those links and See that it indeed brings us to Wicked freebase, sorry So this is the page for those on freebase and It has a link to Wikipedia That has more information. So this is all connected once you connect your Your body field and the pieces of your body feel to those concept Then people can all of a sudden jump in and learn more about it You don't have to look for the link. You just type in or just Select what concept you meant from the drop-down and that's all So that's the view Again, this is all Drupal 8 if you're interested in playing with this approach in Drupal 7 there is actually some work that has been done by some of the folks in the back of the room here from the Finnish broadcasting company and It's called yield why I LD on Drupal.org and We just had a good meeting over lunch today Just to agree on the next steps They have a different approach. They're using taxonomy terms in Drupal 7. We're using a field in Drupal 8 But eventually we want to merge and agree on the best approach. So we're just trying different things now So all of those this session is really Inspirational it's not meant to be definitive those wireframes are still kind of being worked on So those are just ideas and I guess we want your feedback on whether you think it's a good direction for the community to go Towards or if you think it's not the right direction or if it's maybe completely useless I don't know. So I guess that's why we're opening for questions. No, right? Yeah, any questions would be would be welcome And there's a there's a mic over there. Do you have any other things that you built? No This one question coming Hello, do you hear me? Okay? I'm not sure if you If you said it, but I'm not sure if this is Custom module or if this is already a part of I don't know core or whatever Especially the last two demos. So what is this actually? It's already it's it's everything is on Drupal log so if you look for It's so the last one is a sandbox still because it's very early And if internet works well, we can put it up No, not this one. Sorry So already a few I is the first the very first video that I showed in Drupal 8 it's here and Okay, there it is The video is here if you want to watch it again Sachini is the the maintainer their student that worked on this for two two years ago two summers ago and then the other module Yes, the other module is also entrepreneur gets a sandbox, but we can prepare the next question if you if you like I have another one LD tool that's where it is. I'll link data tool. So okay. Is it possible to integrate with some? custom Data endpoints or not only use wicked data or previous or so if you have your own Vocabulary or your own tessori internally and actually the guys from the finished broadcasting company have that use case and they have their own ontology, I believe and also their own Freebase like system. So the the design of this module the LD tool module was made in such a way that you can write your own plugin for your own endpoint for those Excuse me for those Concept so you can definitely do that. Yes. Perfect. Thanks a sort of related question I was just wondering your views of micro data compared to RDFA So that's that's very deep in the weeds here It's the question is about which format to use there's also Jason LD if you want to throw in and other formats and and Also Mike micro formats also. So we really have four formats in the HTML that we could use My take is not I don't think it's big deal or I don't think it's something that people have should have to worry about In fact in all the demos that I showed you'd never had to know anything about this Personally, I picked already FA since the beginning of triple seven and it's still there in triple eight and It's something that Google said they would support Just like just an ID and micro data so At the end of the day, it doesn't really matter which one you pick if you prefer one over the other I just think that if you bring up Jason LD in the conversation Jason LD is a bit different because it's it's like a blurb of Jason at the top of your page It's not linked to any dumb element of your page. It's not in line dry It doesn't follow the dry principle and it has downsides when you have Lots of fields with a lot of content and you end up duplicating your content at the top and loading your page But it's not a I don't think it's a big deal. It's easy to swap in and out if if we ever had to Switch to another format From switch away from already FA In core we only plan to support already FA in contrib. I'm sure there will be at least just an LD so But again You know, I think ultimately what we want to we want to be the cases for nobody had to have to even know that It's like it just happens automatically in the background, you know, and the vast majority of Drupal developers Simply, you know create, you know content models from content model templates and begin to immediately have You know linked data right away. All right, if we don't have any more questions. Oh, there's one there If you're loud enough just yell