 Well good afternoon. I won't be mentioning the words blockchain or ETL, though I'm passionate about both. We're gonna be talking about the porting of FIBO ontologies over to the web via schema.org and the W3C, which is the for those who might not know is the standards body for the worldwide web. So this is where FIBO ontologies meet web publishing standards and I think that maybe in a week or two the FIBO standard for web publishing will be live through schema.org but happily we're gonna show you in a few moments some live examples of the FIBO ontology ported to schema.org live and for those of you don't know schema.org for about four and a half years ago Richard was the web the new web publishing standard declared by Google Bing, Yahoo and Yandex and a few other search engines. So that's what we're gonna we're gonna rock into we're gonna try to be as concise and as fast as possible because we have a lot of content to show you so I'll give it up to Richard. Thanks. I'm an independent consultant I spend a lot of time with the worldwide web community running community groups around bibliographic data I've got a background in bibliographic data archives data FIBO for the finance area tourism I'm even involved in some work around educational courses and how they could be described on the web. I'm currently working on the contract to Google helping them with schema.org the vocabulary itself the site that helps people share information about it extending it documenting it and encouraging organizations to involve themselves in the environment. I do some work for OCLC global library co-operative 27,000 libraries worldwide but you don't want to know about that and I'm also part of the FIBO group bringing FIBO to schema.org. So schema.org I'm quite often asked where the hell did that come from? Well it came from a blog post on Google on the 2nd of June 2011 when Google Bing and Yahoo very rapidly followed by Yandex introduced this new markup standard for structured data on the web moving on from H-card and V-card and one or two other micro formats that are around. It introduced many new types including creative works books music recordings TV series been able to describe immediate objects like audio and video events organizations people places local businesses restaurants exercise parlors products and offers on those products reviews and aggregations it the main use case they pointed it at is something called rich snippets the ability not necessarily to affect where a result appeared in the listings on the search engine but what it looked like when they showed it you in being able to bring ratings into their been able to bring reviews into their images etc etc so what is it it's a linked data vocabulary using rdf triples using URIs and strings so I quite happily put the little link data three blue balls who thought that I don't know but never mind logo in the bottom corner but notice I put the words link data in brackets the reason for this is they don't shout about it it's difficult to find those words on the schema.org site and that's because to most of the people who are going to be applying this vocabulary webmasters web designers link data that's the semantic web and scary and they'll never catch on so you know we keep away from it it's it's got types we refer to types rather than classes properties enumeration values this is where people like you who are into RDF and ontology start to get worried so this is a scary bit for you lot is that it's not domain and range it's range and domain includes this is kind of hinting at relationships not asserting relationships we will see what happens further on three serialization so if I'm a web designer and I want to put schema.org on my site I've got a choice of microdata RDFA and JSON LD and there's more column interest on web lists arguing about which one's the best of those then then goes on forever religious wars especially between the microdata and RDFA people doesn't make any difference schema that will works in all of them so the bottom line is it's a web vocabulary notice I'm using the word vocabulary not ontology it's a web vocabulary to describe stuff and any old stuff it could be anything from your lunch to where you're going on holiday to the space rocket that's just been thrown into the sky it's a general purpose vocabulary so where did it come from it's like most things on the web it's built on what went before so the schema.org family tree and this you know things merge but a family tree will do has got sort of link data semantic web and RDF at the top which is obviously based around original web standards down the bottom we've got HTML5 we've got microdata we've got RDFA we've got JSON LD micro formats going down to things like RSS HTML most of these things are under the banner of the worldwide web we've got Dublin core one of the very early vocabularies but that traces his history back to the libraries 1968 the machine readable catalogue card the heritage is in here so somebody who was saying that we needed librarians in this industry we've been here since 1968 and the worldwide web consortium has been associated or maintains the standards or runs the bodies that's behind all of this and and as was said earlier by Shannon W3C is OMG for the web I would say so where did he first come from 1989 March a little paper called information management a proposal by a certain mister Tim Berners-Lee at the time was presented to his boss and it disappeared in inside CERN he was working in CERN labs in Switzerland at the time and eventually after several months it came back with this now infamous scroll on the top that says vague but exciting and I think that's the best description of the web I've ever heard and he's still valid today he moved on from that we as the web started to establish mid 1990s we were involved with the CERN web server and Netscape appeared and all that kind of stuff 2001 this article in Scientific American appeared which introduced the semantic web the beginnings of this is still in his original paper but this is where he was expressing them and there was a bit of a problem here because I think the guy that did the cover photo read the conclusions and what this might lead to not the actual article so putting a computer with I know what you mean written on it on the front cover fired off more artificial intelligence projects in computer science departments than you could shake a stick at they were all off doing semantic web things they were all off playing with RDF producing prototypes that worked beautifully till you log the second user into your cray computer went off at a tangent it helped the standards get established it helped you arise it helped RDFA it helps sparkle and those kind of things but it was not much practical use to the wider world so Tim pops up again this time in 2009 in a TED talk called the next web well worth a watch even though it was 2009 where he explains by taking the standards that had been established and worked on by all these AI guys and applying them from practical reasons we could actually do something useful this was linked data and it led with a group of people starting to take these principles up and sharing their data on the web using linked data principles to the now infamous open linked data cloud diagram who hasn't seen this precisely so great really impressive hundreds and hundreds of data sets from different industries different sectors universities etc all shared using raw RDF principles using different vocabularies doubling core bebo all sorts of vocabularies to describe this stuff very often based using sparkle on sparkle endpoints if you want to go and get some data on one of those sources you don't need to understand an API it's open linked data and in every we got sparkle on the top we're going impressive but to the rest of the world really useful in some cases yes dbpedia giving you wikipedia in this format is handy geo names giving you a low case and yeah that's handy but the vast majority of this stuff to the vast majority of the world not really a lot of use even though it is impressive the next thing that came along was another blog post from Google 16th of May 2012 introducing something called the knowledge graph talking about things not strings which is starting to get into what we've been talking about for the last couple of days it induced the knowledge graph the first example of this we started to see on the right hand side of Google search pages when they can identify an entity that you were talking about they'd put your a page up very often entertainment or things like that or more serious things that might interest people in this room you never know describing an entity with information pulled from many places and these into entities were related together so Bart Simpson being related to Nancy Cartwright who's related to Dayton Ohio is related to a really good Air Force Museum in Dayton Ohio if you've got a spare data to while away in Ohio well worth the visit and it's free but anyway these relationships are not from a restricted set so we have Bart system who is played by Nancy Cartwright who happens to be born in Dayton Ohio and happens to have a place of interest relationship with this Air Force Museum but that set of relationships is now into the hundreds of thousands within Google they are relationships that they identify to build their knowledge graph so where do they get this data from when they started they seeded it with a data set called freebase freebase is an RDF based open data set that people have been throwing data in two for years they got it when they they purchased the organization behind it they get data from search results when they're really really sure what a page is about they'll say that's an entity and put it in their data set they pull information from Wikipedia and originally more so now they pull it from wiki data which is a data set behind all the facts in wikipedia and they pull in it from schema.org so whether they do with it well what they do is process it through their little pipeline and pour the entity descriptions into their knowledge graph yes and every time I say Google treat it as a collective name there are other search engines so what do they use that for it then drives the knowledge graph it provides the facts and the relationships in the knowledge panel display it also powers infoboxys if you search for a specific thing in one of these search engines very often you'll get a panel which talks about that thing with some links around it's a more interactive more rapid response process with the user also some of the facts are pulled out to supply to supply answers so if you ask how high Mount Everest is they will give you a box with a quick answer and some nice pictures and some navigation it's still also powering the main use case originally which was hang on we what happened there oh look my computer is telling me there's a new version of Java I think we'll we will we will skip that version right good fine so so it still happens in this rich snippets use case some people say where does this fit in with whatever's going on you may have heard of something called the Google ranking brain which is kind of artificial intelligence pulling signals from all sorts of places the knowledge graph is is a compact key component of that so this is driving the way the future goes anyway schema.org is one of these sources schema.org is a success why do I say it's a success well first off it's data is embedded in website HTML using microdata RDFA or JSON-LD it's not a challenge for people to share this data with the search engines which means it's harvested during the normal web crawls the search engines don't have to do anything special to go and get this data they just use what they've been doing before and it's under the control of the site publisher so when when a price changes you change the schema.org structured data behind the page when you want to the next crawl comes in and picks up that fact they haven't got to analyze the text on the page and kind of work out that this bit might be the price and is it different to what it was before and you can certainly give them facts and this success is now reflected in usage at the end of last year it was identified that over 12 million websites are using schema.org another way to view that is in a 10 billion page sample that Google and not a collective name this time did they analyze and identified 30% of those pages had schema.org markup in them that's a hell of a lot of the web that's adopted me so all very interesting but why is it relevant to us why are we bothering with this what why are our organizations worried about that context why should they be worried well if if I say to you website I imagine something like this gets conjured up maybe a little bit prettier but most of the people assume when we're talking about the web it looks like this well actually that view of the world is falling into disuse it's not being used as often you get this view so if I type in ATM into my phone I get this I get a map of the local ATMs I get a list of them so there's one Bank of America not far away so using information from the banks Google are providing answers to my potential users without them having to touch my website they're going directly to them they're having the conversation with my user which allows them to come use my ATM increasingly users are interacting with watches and other mobile devices directly into systems people are starting to talk to their computers you know hey Siri is what goes on here and let me a little aside here I was rehearsing this presentation because it takes a lot of rehearsing to get it into the time I've got in my room the other day and I got me iPhone on the table timing it and I came to this slide and I said people will say hey Siri where can I get some cash and this is what happened or should do come on me sounds gone no okay all right I can play that back to people later if hang on will it work no it's not gonna work so let's go back one hey Siri where can I get some cash yes please directions so that that was me in my hotel room on the seventh floor here interacting with the data that in the knowledge graph of the search engines nowhere near a website now okay that's simple data locations of branches and ATMs but hey Siri where can I get a loan to buy my house with hey Siri what bank account gives me a good overdraft level you can hear where that direction is going to go but they need the information to provide it so this is where schema.org becomes important I'm going to give you a very very brief tour through the way the library world did this a couple years ago librarians really understand their data just like you look they're good at cooperation just like you look early schema.org when they looked at it was useful but not good enough for their purposes so they will actually I formed a world war web consortium group called schema biv extend we identified public data sharing use cases knowledge graph like we talked about and unusually we didn't start off with vocabulary or ontology and try and put that into schema.org we said what can we do with schema.org now then fill in the gaps you end up with a much shorter list of things to do and you tend to fit in with the rest of the world we created real examples proposed enhancements to the core vocabulary provides a more focused extension which became bibb schema.org and it took a couple years to get there other groups that have done this the automotive group set up a w3c group minor enhancements the core now there's an auto schema.org extension the medical group put a load of terms in in fact the overwhelmed schema.org and they're coming out into an extension they're not being dropped sports news TV and radio they threw their stuff in and that's validly part of schema.org so by bow dot schema.org was to take this approach we have a great amount of information about the entities in the financial world how are we going to describe them to the outside world so that the search engines can direct our potential customers to the loans and the bank accounts and the credit cards and the other facilities that they want to take account of. We put this together in the fall last year and we've now got the beginnings of financing schema.org and this is a proposal for the core just like the other groups we've identified some generally useful terms that could go into the core of schema.org one of them's loan of credit you get a schema.org page about it you get some examples of how to use it I think that one's in RDFA but we do them in all examples and the set that we put in bank account no bank account in schema.org hey currency conversion service deposit account financial products payment services payment cards things like that have gone in and this is the real life not live but captured that Chris was talking about earlier this is Bank of California lending page loans three types of loans talked about there just to Google in its original form that's a page that just talks about loans that's all it is but by embedding schema.org markup and this time in Jason LD it identifies four entities on that page three types of loan a bank that is offering them and the relationships between them now big pieces of information but quite small here is the offered by of a particular loan type by the Bank of California for a residential loan so where are we with this core proposals which have gone into the schema.org candidate release for the next version at three minutes past two this afternoon I received the email to say it's in there and the candidate release has been released within schema.org and within one to two weeks we expect that candidate release to actually become schema.org 3.0 which includes the basis for the financial terms coming from Fiber in the background we're working on an extension to schema.org which is very financially focused building on the foundations of the core terms so we're talking about things like brokerage accounts mortgage loans repayment specifications that kind of thing that generally in schema.org that's a bit detailed so we'll put it in a Fiber extension and eventually after a few iterations of Fiber.schema.org we expect an external extension which will be it's got a running title which confuses a hell out of everybody and we must change it of schema.fiber.org which will be 12 billion users 30% of pages what the search engine is asked for if you go to Google and say how do I mark my pages up they'll tell you schema.org finance and banking was poorly represented but now through Fiber initiatives it's ensuring not only visibility of the terms but visibility to the users that can use them enabling the kind of practical applications we just showed you core proposals plus the schema.org will make financial terms based on Fiber so integral to the knowledge graphs that the search engines are building core to the new direct access web the the hey Google hey Siri environment marking King directly to where the users are so that your bank can market its financial product to the people that are looking for them and I'm going to hand over to my colleague Chris because I need to lie down a little tired huh thanks to my dear friend Richard for walking us running us through this so let's see here you made a few changes to the deck didn't you Richard so in this case BNP Paribas is a member of the EDMC and one of its subsidiaries is Bank of the West they have 800 branches west of the Mississippi here in the United States in North America they will be shortly implementing some of the code that Richard has talked about specifically there they asked me to they asked my my company they asked me caliber media group to implement JSON LD for their credit cards so pretty soon on their main credit cards webpage while the HTML along with HTML 5 embedded with JSON LD structured data as as retrieved which isn't which will be a now an outcome of the work that the EDMC has done with schema.org for porting FIBO ontologies over to schema.org vocabularies I know it sounds kind of complicated but if you could just play along without the super I'm not sure why we have this slide after the fact Richard but this is the DB pedia so DB pedia often serves as a URI within JSON LD schema which is then put into HTML any webmasters at all in the audience today okay I got one this is freebase by the way Google owns freebase and freebase is being deprecated so it's kind of no longer being updated at all instead freebase is being ported over to wiki data and Richard or I over dinner last night you gave me some fuzzy answer about when it's really going to be ported over yeah only they're in the process of transferring all the concepts that will fit into wiki data from freebase I reckon by the end of the year or okay currently we do have live examples Bank of California Bank of the West is coming up which will contain FIBO the FIBO version of schema.org currently and I know a gentleman a great friend of mine who's in attendance here hello mark Sita insurance which is a subsidiary of Brown and Brown very large insurance company they're they're starting to deploy my our code onto their pages that is FIBO schema.org compliant so they're jumping ahead of the curve but I know that our version one of FIBO for schema dot the FIBO schema.org extension is going to be going live shortly and I also have a credit union where I'll be placing the code where my team will be inserting the code it's a California federal credit union and so this is the future of the web this is the future of the web publishing standard for the financial industry so this is basically this session basically is a big FYI I highly advise you to get on this bus and get your CMO to get on this bus and by the way at the end of this at the end of this session the San Diego semantic web meetup group is having a secondary session to talk about the SEO implications search engine optimization implications of this rolling out of the FIBO extension to schema.org it'll be in the ale house which is like once you walk down hang a left here and you walk down a little bit more it's the first restaurant that's right in the lobby of the hotel and coincidentally three of the members of the EDMC FIBO extension to schema.org working group we've decided to form an international coalition because or a consortium because this is going to the implications for G-sibs everybody know what a G-sib is? All right the implications for G-sibs is going to be well just this is an overall all of all web publishing and we wanted to hit G-sibs and help G-sibs transform their web publishing to adhere to this new standard version 1 version 1.1 1.2 version 2 and three of the folks on the EDMC FIBO schema.org extension working group we've decided to come together and work together and so that would include me and this charming fellow young fellow who just spoke and also Mako lab which is also working hand-in-hand with the EDMC in order to facilitate this code coming to being and that's Mako lab unfortunately our dear friend Mira Xopec couldn't be with us today he is he seems to be he's hung up in Gainesville Florida at the moment with a client but he is based in Eastern Europe and what else do I have to say thank you and do you have any questions thank you and please ask a question no questions at all Ian Richard how long do you think what's the inside story on how long till we get released the 3.0 3.0 got me brain on backwards sorry version of schema.org one to two weeks and there's a steering group that reviews the release candidates and unless there's some major showstoppers that upsets the sensibilities of anybody on the committee we expect one to two weeks so by the end of the month I would imagine now it's live and they're harvesting it and because it's built on broader schema.org already there's a lot of terms that are in there that they're recognizing right so I I deployed extant for one client they wanted to they wanted to change their search engine optimization from being on page 7 for flexible business loans flexible commercial loans flexible residential loans because that's there that's what they do so we deployed the customary code got them from page 7 to being on page one position position one for flexible commercial loans business loans and residential loans that's a bank of California actually it at least it took a week that's not long and then what we've done is add we've had we've jumped ahead of the official FIBO release for schema.org and embedded that code anticipating that the code will get released into schema.org and will then be understood by Google bot and Bing bot which is the only two search engines I care about and it hasn't hurt their rankings they're still they're still stick sticking at number one so Google is it is it Google bot Bing bot are ignoring the the additional FIBO code but we wanted to write the code before it got released because who knows when it's gonna be released one or two weeks. One more question back in the back. The reason yes well we're coming at this problem from two ends FIBO is mapping and modeling and producing the ontology around detailed forms we're taking the schema.org way of describing things to describe those same two entities over time they will start to map together and some of the things we've described in FIBO.schema.org and not yet mapped in FIBO and vice versa so as the releases come together and we move to more in-depth extensions and eventually that external extension I was talking about that mapping when we come closer and closer but they the granularity of schema.org is far more coarse than you would expect inside FIBO so don't expect every property in your ontology to turn up in schema.org because it won't. So but we are working to align conceptually as much as we possibly can the terminology in schema.org with terminology and the concepts in FIBO so the folks in FIBO are participating and vetting the terminology so that we have that alignment eventually we will have a a bridge so that we'll be able to utilize machine capabilities to transition across. Let's get this guy out of the hand for you.