 Thank you, Leith. You're probably slightly tired of seeing me by now. I'm not really... I'm not quite at a loss for words, because I've read a whole lot of stuff down, which I haven't practiced, so I'm going to read from the paper. I'm sorry. It might be short, which is a good thing. You get an early cup of tea. I also have to preface this by saying that I'm talking about one very simple idea that's linking two sites together. But they might lead into some more complicated ideas. But those ideas aren't really mine, and they're not really fully formed, and I haven't really got any answers. But they do pick up on some of the ideas we've heard at the conference already, things like small data, making connections, and then you just get into the stuff that Chris McDowell was doing, and that's off on another sort of spectrum. Having seen some of those talks, I kind of wish I wasn't giving this talk anymore, maybe if I'd given it last week it would have sounded more relevant. But there are ideas that are kind of floating around. It's what Virginia Gowell sort of refers to as these shared ideas. They're just ideas whose time has sort of come. We're all kind of thinking about them. We don't all know what the solution is, but we're kind of thinking about the same sort of stuff, so it feels like we're going to solve it at some point. So a little bit about the ministry. We've got a tiny little bit about the ministry, just that we publish a whole bunch of websites. Our two biggest websites are Teata, the Encyclopedia of New Zealand, and NZ History, New Zealand History Online. They pick up about 95% of our web traffic, so that's the ones we tend to think about. Unlike a lot of the organisations involved at NDF, we're not a collecting institution. We do a little bit of oral history. We take some photos, but we're not a collecting institution. We write text, and we write a lot of it. We have millions and millions of words, both in print and online. Where we intersect with collecting institutions is in the thousands of images that illustrate our stories. Is this going to work? Right, there's a very quick search, a Google image search for pictures on Tiara associated with Te Papa. Adrian don't look too close, because you'll probably find some anomalies. That's Google. Probably one of the country's biggest collection users. Tiara has 25,000 to 30,000 resources. There's probably someone in this room who can correct me. NZ History has maybe another 4,000 or 5,000. When it comes to collection items, like this gourd, or a kalabash, or a huay, or however you want to say it, our sites tend to explain their significance. They place them in the context of stories, which demonstrates their relevance to other items. I'll come back to that idea of relevance later, but for now I want to talk about a very simple linking project we're working on with Te Papa. We saw some images and other media from institutions like Te Papa. Most of them are available on Te Papa's website. And what we're doing in this project is not rocket science. We're just going to provide links between two websites. So if you see it on Te Papa's site, you can go and read more about it on Tiara by the link. And if you see it on Tiara, you can find the original version of it on Te Papa's site. I should just say why we're using Te Papa as a test case. A big part of it is down to people, and I'm going to blame Adrian for the whole idea. Adrian Kingston sitting over there, he not only came up with the idea, but he saw it as something that was possible, relatively easy and worth the effort. It's also partly the synergy between what the two organisations and our websites do. We both take a national view, and as of Tiara, we try to take an encyclopedic view and cover up all aspects of New Zealand culture and history, much the same way as Te Papa's collections have that sort of breadth of coverage. It's also partly the size. We don't use a lot of Te Papa's images, curiously enough. There's only about 500, so we've actually got quite a manageable set to work with on a manual basis at this stage. Te Papa also uses persistent identifiers, for items on collections online, and we can't do the stuff without persistent IDs. Knowing they'll be there forever, we hope. Adrian? Okay. So the process is currently manual. It's basically a spreadsheet that we gave to Te Papa. They're going to fill it in, they're going to send it back. They've sent half of it back already. And it's going to link our URLs to their URLs. We're manually going to put that stuff into our systems. So that's not a major hassle with 500 items. At some point, I hope we're going to have to think about how to scale it beyond 500. I'm hoping we get to the point where we think this is a good idea. Let's go and work out how to do it with 7,000 items at Turnbull. Or whatever other organization we want to work with. So what do we get? We're basically linking these two items together. You can look at it on, where are we looking at? This is from Teata. From here, you can see that... Sorry. What you're getting from Tiara is information on the item's significance. It's in the story on traditional Māori warfare. It's illustrating the page about preparations and entering into battle. From the caption, you learn about its use in this context. So Tiara provides context. It signals why this thing is important. Same picture, different website. It's even got a new name. So from here, you can find out... It's got a different name. It's also a kalabash. You can find out it was made from Harakeke, Muka, Gord, Dai. It was purchased in 1905. You can see the collection it belongs to and what it was influenced by. And all those underlined words link to more items that share those classifications. So what we're doing is helping people find the information that's relevant to them. If they're interested in the story, they can get it from one place. In detail about the item, they can get it from another place. What you get in each place is what's most relevant to you where you are. But you can easily find the information, find the other information if that's what you're also interested in. I'll just talk about some other examples that might make this slightly more interesting. Everybody knows who Abel Tasman is? Yeah, good, good. I'm pleased. Anyway, so what was I going to say about this guy? This is Abel Tasman. There was a symposium recently, Glam symposium held at Victoria University. Eric Kettler from the University of Amsterdam spoke about Abel Tasman and the various collections of, I think, ship's journals. I'm a bit hazy on the details. I think he was talking about the journals of the voyages. Different copies of these things exist all around the world. So some exist as simple scans. Others are transcripts of the Dutch, I think it was Dutch. Others are translations of that Dutch. So you've got all this duplication and that's not a major issue. As archivists tell us, lots of copies keep stuff safe. The issue is that none of these copies are linked to the others. The copy you have to stumble across directly affects your experience and how you can interact with that piece of content. If you can't read longhand scans, they're no good for you. If you can't read Dutch, the transcripts won't help. If you can't read English, you might be better off with the Dutch. It wouldn't be hard to link them together. They're on the web, so it is just hyperlinks. So, you know, link them together. If you find one copy that's not the one that works for you, you get to the one that does. Slightly closer to home example. Excuse me. This is an image from what's called the H-Series. I think. I'm pretty sure it is. It's got an H on it, so. It's a collection of World War I photographs commissioned by the New Zealand Government and taken by Henry Armitage Sanders during the First World War, I said that. So, the Turnbull holds the original glass plate negatives and other copies are held by organisations all around the country. Auckland Museum, for example, has copies and photo albums and that points to the story of how the images were originally used. They were put into albums with captions and the albums were distributed around the country so that soldiers and families could order prints. That's why they're all numbered. You knew you wanted H-391. You got H-391. So, again, making a link between those things, which doesn't currently exist, linking to the originals. Linking the originals to the albums, it gives people the chance to experience and use the images in different ways. The best place to go is Turnbull. But if you want to experience what it was like for a nation to see what the war had been like in page after page photos, you can view the albums. Just making that simple link makes that sort of thing possible. Back to Tiara. Yes, here we are. The relevance of items to other items. I've written a bit over the last few years about the way that something like Tiara, but really any publication, whether it's a book or a website, anything that uses collection items is kind of like the meat and the sandwich between collections. Where a publication uses items from different collections. It's effectively creating an inferred relationship between the items and the collections. I keep coming back to the Calabash. It's many names a part of the complexity. But as are its many uses. As we've seen, it illustrates the story about the traditional Mali warfare. But it also illustrates the story about Orono. Or the medicinal use of plants. So within that story, it's suggesting relationships with items as diverse as other plants. Other plants. Sorry, here we go. Plants. An engraving of a Mali warrior. A lindaf portrait of Tohanga, Tuhoto Aruki, and a cartoon of Maui. So those are not connections that an institution would make from its item to other things. So through that story, we also have inferred relationships forming between Topapa, Turnbull, Auckland Art Gallery, the Department for Conservation, and Godwick Publishing. Some of those organisations don't naturally see that they're related to other organisations. But through that story, they are. Some of these links even start to get a little playful. Kind of in the way that Cat Styles talked about NTF last year with the Sembel game. We can let a user make connections between items. This is from the collecting story, which is getting a little meta, but that's all right. It really picks up a really wonderful assortment of subjects. From Turnbull himself, and his book plates on the left, the four pictures. Firearms, Barbie dolls, teaspoons, the ever-popular teaspoon. They're not the sort of connections that a collecting institution tends to want to make. But once it's in the user's hands, those connections can start to form. So, where can we go with this? At a simple level, item-to-item linking opens up a few options. We can potentially share our content more easily with other organisations if they want it. That saves them the effort of writing new content about their items. So we're already looking at that with Tapapa through sharing our content back into their pool. Tiara itself could actually use the information to update copyright or other information when the institution record changes. We'll start to look at pulling in the descriptions of items for our sites to use as alt text, for screen readers to use. Where a collection item uses a more descriptive title than the way we talk about it. We're also interested in sharing our content with third parties to build new publications, websites or apps, or whatever people are going to build in the next five years. Currently, we can only share the text on our websites because that's our copyright. But if we have a direct link to an item, then it's easier for a third party to find the sources back to the image and they can go and negotiate their reuse with the holding institution. Potentially, services like Digital New Zealand could start using those links and the information to map our stories to institution records and expose those relationships through their own API. I was looking at Chris when I said that but I think he snuck away. There he is. He's hiding. I was looking at the guy. But more than that, we can go back to Adrian Kingston and start to look at his suggestion that we start to use the items to actually catalog the stories that they illustrate. And this gets a bit confusing, but that's alright. If you look at Tiara subjects, they are at a very high level. Typically, the story title and the page title is an effect on the main subject, which is fair enough because it's an encyclopedia and the headword is typically the subject. But what if we used the items to infer more specific subjects that the story might relate to? From that, we might see the story about Māori warfare. So taking this image of Ataaha on Tiara. Yep. It's also a story about wood carving. This is the information from Te Papa. So it's a story about wood carving, potentially about the use of materials like feathers, dog hair and flax in Māori society. So making that link gives us that additional story around that piece of content. So through that connection it's also then related to thousands of items in Te Papa's collection. Now, not all Te Papa subjects will be directly relevant to Tiara's story, but by being able to choose which ones are, we can start making those direct links and link our stories to much larger events for people. This is where it comes to start I'm going to name check Virginia again. So that's two each. That's good. I'm going to name check Virginia Gao again. She just threw an idea at me when I was writing this, which picks up on some work in the Netherlands. I think we're the National History Museum, which I think is an institution that doesn't have a building. Is that right, Virginia? Yeah, probably, possibly. But anyway, they built a website that basically joined up with some other websites and kind of created what was basically a trusted network of sites. So we're one site linked to another site in the network. That link got reciprocated automatically. And that's the sort of thing you could then start letting your users do for you. They think this is related to the same subject over here and they make a link that gets reciprocated and you've actually got... it's an easy thing. There's two links going in each direction. It shouldn't be impossible and it is the kind of stuff that Facebook does when it lets you tag someone in a picture or that WordPress does when you allow ping-backs to your blog posts. What they're doing is they're letting themselves and their users be aware of a much larger network that individual sites are part of or individual pages or photos are all part of. Of course that heads into the territory of Michael Asperidis talk yesterday about when you actually join two collections together. It's not A plus B, it's A times B. And you get into exponentially large numbers of potential links. But let's not get scared by large numbers just yet. I think simply linking items to items is going to take some work and it's obvious we'll need to work out ways of doing it automatically when we look at a larger set. Could we let our users do it? Could we just start sharing our data in such a way that clever people can start making those matches for us? I do think it plays into the work that Chris McDowell demoed yesterday. And I wrote that without even knowing that Chris was going to do that presentation. Well I knew it was coming but it is making those matches across collections based on people is the work he's doing. It's the sort of thing that can be done automatically by the right person with the right tools. All we need to do is let people like Chris use our websites and collections and see what they can do. People are kind of easy, ish. So are places, ish. They're kind of easy. But they're basically just hooks that connect us that can connect our websites and can connect our content. Subjects and classifications are potentially no different. They're a little more ambiguous, probably a lot harder, but they're not impossible. I was going to call this slide Historical Accidents but I didn't think that was a good title for a slide with the treaty on it. But it is the treaty. Here is the treaty. Yeah, it's not. Sorry, it's a really bad connection to make. One of the things you notice when you look at collections is that they're never as comprehensive as you'd hope. I'm going to mention Colin McCann because it's the trendy thing to do this year. It is that thing that Colin McCann has distributed collection. So collections are riddled with historical accidents. No institution has everything related to Colin McCann or any other artist. You can also think of the Treaty of Waitangi. It's held by archives New Zealand. It's soon to be housed in the building of the National Library. And arguably, it's as relevant to Papa's collection. And I've heard anecdotally that people come to Papa expecting to see the treaty. I think you can look at any significant artist and see how they're scattered across museums and galleries around the country, if not the world. That's just history. Different things get picked up by different institutions at different times and we can't change that. But for the user, it's infuriating. Why can't I see all of someone's work in one place where everything on a particular subject all together? And I have to apologise for the slide. I put it together. That's my attempt at a network. I guess the beauty of linking all that content together is that it creates a layer of meaning and use that sits above our individual collections. It lets us all play to our strengths, importantly. Organisations can maintain their own web presence. It talks to their mission, their collection, their community, their building, whatever they want to be telling as part of their institutional story. But it lets a much wider community tell their own stories that crosswalk all the separate institutions and collections. Through that, we and our users could create truly national stories using all the different parts held in different institutions around the country. So I think that's pretty much what I've got to say. It's kind of, I mean, I know it's taken us away from the simple idea of linking items to items. That network stuff is kind of hard, I mean it's very hard, but we do need to do it. I think, importantly though, at the same time, we shouldn't forget about just doing simple stuff. If the stuff you can do to connect your collection items to other collections, just do it. It's a start. It gives more use and meaning to your users than you're doing. Then you're doing something right. But we do need to keep hard stuff in mind. We need to agitate for it, remind people what it's worth doing. Do it if you can and share the results with as much of the rest of the network as you can. That way we build the richer digital ecosystem for developers and our users. Have to take questions? Have to go and have coffee? Yeah. I just had a question about how with a digital New Zealand that's in our linking network and you don't actually want the end users being links on particularly on tiara or internet? But the digital New Zealand would sense is the ability to connect them. The question is around what role could digital New Zealand play in this and there's another question about do you want people doing things on your own website. I think the second question first, potentially you do want users making changes on your own websites. My limited understanding of what happened in the Netherlands under the INL.NL brand was that it actually suggested connections. A user could start typing in and get a look ahead suggestion. So if you're on page about Rhetorangus and you start suggesting that Rhetorangus might be the subject it will look ahead into its database and say we've got Rhetorangus in these other places. I can't see why that would hurt some checks and balances. Digital New Zealand Yeah, great thing for Digital New Zealand to do. I think they're in that sort of space where they have become really critical to the future infrastructure of our network. We need to support them. We probably need to agitate them to get some more money. Chris was talking about the work he was doing yesterday and he's sitting on a train doing that. We've got some really smart minds doing the stuff. We need to get them financial backing to do more of it. Does that answer? Yeah. Chris has got his hand Chris, do you want to come up and take over? No, I just wanted to there's something I was wondering what the links are that you're linking. I mentioned that it's just a plain anchor tag going from one place to another. Is that correct when you link your Calibash from Tiara to Papa? Is that correct? Yeah. I just want to make a suggestion. I started thinking it through as you were going and it actually kind of blew my mind a bit because you could totally do this and it would be really easy. You should take a look at Schema.org. So Schema.org is the Google, Yahoo and what's that? Bing. It's a web standard. It's around microdata and HTML. I'm just going to paraphrase as you go. Chris is telling us to look at Schema.org. You've got some HTML and it's just a few extra tiny little tags that you put into your HTML but what it does is that you can define people so you can have say a heading and just also add this little microdata tag within the HTML because this is a name of a person. On a link you can put in the same as tag and the same as tag means that whatever you're pointing to, so the calabash over to Papa is the same entity as the calabash over there so it's all within the HTML. We could totally harvest that off you. That's what Google and Yahoo and Bing are doing. Google and Yahoo and Bing also recognise the same entity and so Tiara would recognise that the page represents a person and so all of us would see that it's a person page. Chris has just said that Digital New Zealand will harvest that information but good suggestions we should look into and probably need to look at your end as well. It means that the linkages between your institutions are semantic linkages and it's really so straightforward to do. Tom. We've been running Schema.org on our sites for authors and for books so it'd be great for them to do. The one thing I'm cautious about, I mean Google Books has been running Schema.org for about 18 months now and they're still only running a partial implementation so I'm sort of wondering at times if Google is still running a full implementation of it where how long it's going to take but it's working well for us. Right, I can't even paraphrase that but Project Williams Books is using Schema.org for some potential limitations from Google. I'm just, sorry I'm paraphrasing it's just the session, we don't have a microphone to pick up on the session. Sydney. We're talking about a sequence of additional objects each of which is contextualized within its D-space each of which is technically a different record and therefore visit ontologically and logically the same object and we have different users How do I paraphrase this question? It says that we're creating a generation of format line students ones who love the harvesting of content the exposing of the lengths, etc but they're actually not aware that A, they're different formats B, they exist in different places that have different ways of framing those objects and so what they assume is that the calabash that is in Teara is the calabash that is in Tecaba without recognizing that there are different stories to tell and different ways to tell those stories so can as part of the sort of paradigm reaction on the process, is there a way of incorporating some recognition that the lengths themselves have a story to tell and it's the story like Sunge Temesh says it's evidence of me that it's a view but it's not exactly the same story This could be a movie called Paraphrase and I'm just getting too tired to even think about that I honestly don't know I think it takes me into the realm of thinking for the majority of our users are they worried about that are they worried about that now is it a story we need to record but it becomes more interesting further down the line Papers pass as if they're reading the original they're not including the URL to papers passed and therefore not exposing that information I guess sorry to jump in before you get to your second example in some ways this goes back to my thought that something like Teara or into history or any kind of publication is kind of a discovery mechanism for collections where we're sort of hand picking the really picture perfect little things and using them to illustrate stories giving them context we're not saying that that is the definitive record and this is a way of getting you back to the source institution which is slightly closer to the definitive record because of course the definitive record of that colour bash is up on level 4 I think in a glass display case and by tracking it back you can actually get to that definitive record how much we need to be telling people those differences I don't know if they learn enough about how a colour bash was used and what it was made of and where they could see the original item I think we've done a pretty good job for the digital humanities in the room that might not be a good enough job but we're doing what we can that's me thank you