 Ia, hi, Brody, thank you for coming along. Ia, it's a real shank, Chris can't be here. I was really looking forward to co-presenting with him, but I figured it's a pretty reasonable excuse. So this is all, this all comes from the Christchurch earthquake. I appreciate some people might be sick of hearing about Christchurch, but for some of us it's still an everyday issue. That was my old workplace at Kenton Chambers in Hereford Street in Christchurch, February 2011, about 15 minutes after the quake. And this is a presentation about some projects that are aiming to preserve materials that came out of the earthquake, in particular the skirt learning legacy project. And this is all about a project to capture material that's been developed from skirt, which is the infrastructure rebuild team. It's a consortium of multiple companies working on horizontal infrastructure. And when we say horizontal infrastructure, it's all things like the underground piping and so forth. Some of it isn't very horizontal anymore. And a lot of it is wastewater pipes, all the underground services that we rely on, but they tend to be out of sight, out of mind. But in this case we're having to rebuild half a city in a space of under a decade. So coming out of that are a lot of lessons, and they're not the kind of lessons that we learn every day. So there's an opportunity here to preserve some of that for the future. So a project was set up, a collaboration between skirt and the earthquake research centres at University of Canterbury. And skirt want to communicate their rebuild work and their objectives to capture new ideas and new approaches that they've been forced to introduce. When you're rebuilding on such a big scale and such a short time period, you've got to get really efficient and develop innovative approaches to things. And they want to share their knowledge nationally and internationally in a way that can be understood by a wide range of people. There's also seismic, which is the research consortium collecting digital content coming out of the earthquakes and creating an archive for future research. And Quake Studies is the repository that is supporting that whole seismic project. There's the Quake Centre, which is an industry-funded research centre typically focused on engineering and promoting awareness of earthquake hazards and engineering solutions. Quake Studies' research group focused on organisational resilience from many perspectives including engineering, management, psychology, sociology and business. And then Catalyst IT, and I have to point out here I've got to give due credit to Learning Media because they were the original builders of the Quake Studies archive, and Catalyst picked up the support contract after Learning Media went out of business and full credit to Jason Darwin who is the primary information architect for the system. And there's some really quite interesting themes that come out of this. This is a sculpture, a mural called, by Wayne Yule, called, I seem to have temporarily misplaced my sense of humour. And there's a lot of things that are very temporary in Christchurch these days. And the one in 2000 earthquake is another reflection on time and what might be temporary in an earth sense is not temporary by human standards. It's about lost things, lone tools, possessions, pets, lost relationships, that kind of thing. It's about the power of memory to return us to some kind of order that we want to impose on a possibly random and capricious world. Work done by agencies to rebuild the city is temporary. Most of these agencies like Skirt and Sarah and CCDU have all got very limited lifespans so part of this is preserving the work that they have done or how they did it, the why they did it. A lot of money is going into the infrastructure. We build billions of dollars and it's time to document it right now before all of these agencies disappear and the staff leave. And so the learning and legacy concept is an intervention to try to capture this kind of information and share it and preserve it. So like I say, a lot of this is underground so it's effectively invisible, especially once the work's done. I mean everybody in Christchurch will complain about the traffic and the disruption to roads and this is why they're being disrupted. But once the work's done it's out of sight and most people won't even know what was done. So it's all about, it's part cultural heritage, part research project. The whole idea is to make things visible and that's especially important when such a major programme of engineering is all underground. Skirt's learning legacy is important both to the general public and to specialists, to planners, engineers and project managers. People who are keenly interested in the knowledge that's come out of the Christchurch experience. Skirt has been recognised by the Institute of Civil Engineers in the UK for its engineering innovation and the learning legacy in some ways is a bit about PR also. But it does prompt scrutiny of what gets preserved and why and there are other organisations that could benefit from going through this kind of exercise. So the learning legacy concept itself capture information now before it disappears. Skirt is due to wind up in December 2016. Develop an archive programme that could be applied to multiple agencies. So there's already been working with other agencies like Gapfiller for example. And although seismic represents cultural heritage sector this is a more precise focus project on engineering information. But this could be useful to other cities as well. There's also some scholarships that have there are already some scholarships that are underway for currently funded at master's level including one looking at art collections management. And it's also an opportunity to develop the platform as a whole because all of the components of this are open source they're all available to other people. So what's the process that was gone through to get there? So initially the Skirt Learning Legacy project conducted some audience analysis and they interviewed a lot of stakeholders including organisations like Fulton Hogan, Fletcher's, Becker, Littleton Port Company Christchurch City Council various others to figure out what kind of information would they be looking for to take on board some of these lessons. And also some research especially around the categories that people might use to find this kind of information if they're searching for it. Then was some design work Catalyst did a few upfront designs and they chose the elements of those and came up with the design for the site which is all pretty simple because at the end of the day it's just presenting information and the presentation of it doesn't have to be that sophisticated. And out of that came a data model which included categorising the content into themes things like communication, design construction, finance all of these aspects. And the idea was that that also came out of this was the idea to reuse the Quake Studies repository that was already in place, already had a lot of data from other areas. It made a lot of sense rather than build a completely parallel infrastructure to store this information put it in Quake Studies and put the front end on it that was tuned to the specifics of the skirt learner legacy project. And then you drill down on the themes and you get to individual stories and that was basically an articulation of a problem or a challenge and how it was overcome and what solutions were applied and this is the kind of thing that people can potentially benefit from. And those stories that's just a teaser text if you click through onto that there'll be more information about the overall story and then it's supported by various attachments which are often engineering diagrams or forms PDF forms or whatever were used in the process of developing the solution and it might include flyers and brochures and public communications material all that kind of stuff there's a capacity to add video audio that kind of thing as well. The Quake Studies data structure is reasonably complex I won't bring that up now because I'm not online but there is a whole ontology behind that and if you look at the slides later you can click through to that but a very simple part of it is the whole idea of a collection and that can be a nested hierarchy of collections and then it's associated with that a collection is a part and that's typically sorry it should be an object and then associated with as part of a collection you have objects and then inside objects you have the parts and that's what we then map to over here so sorry that slide in the slide also in the Quake Studies information architecture you had collections nested collections and objects and then parts and over on the Skirtland and Legacy side you had this idea of themes and stories and attachments and what we had is a relatively straightforward mapping from one set to the other so that we could talk to the Skirt people with a language that they were understanding and map that to the existing ontology that was inside Quake Studies now I just skipped over the data model inside Fedora Commons just for those who aren't familiar with Fedora Commons as an institutional repository solution it's Java based it's very heavy on XML all of the content has persisted to disk on XML and it uses a triple store and MySQL and things as a kind of operational reporting engine but all of the data is basically in data streams that are serialized in XML so you've got your Dublin Core and your Relzex which is the relationships the external relationships from a given object to any other objects so we can get your hierarchies and your sequences, galleries and so forth and then your individual data streams which may or may not exist depending on the nature of the object, but it could be a thumbnail or an image or a sound file or a video file or whatever and those are all captured in that model so here's an example of a story with a title, text supporting metadata and it might be associated with a bunch of PDFs for example so in terms of coming up with at this stage we knew that we wanted to reuse the existing Quake Studies the Fedora Commons part of Quake Studies and fortunately it had an API but the question was how to do a front-end that was specific to SCIRT and we came up with a few options a really lightweight wrapper using a simple framework like Django or maybe Node.js because all it would have to do would be to yank data out of the API and display it we could extend the Remote Entities approach which was a Drupal approach using Drupal content management system that Quake Studies is already using for its main site as it was developed by Learning Media that was another option that was potentially more interesting to people because it kind of preserved SCIRT's options if they wanted to add social media sharing or forums or some more interactive features then having a full blown CMS kind of gave them a bit more flexibility there and then the other option was to use another Drupal based system called Islandora which unfortunately at the time that Quake Studies was built was kind of a bit too early in the game to go with but it's developed it's come a long way since then and it's definitely a viable contender for this kind of project so we had a look at all those and decided that the simplest was to extend the existing approach that Learning Media had chosen which was the Remote Entities just to give you a very quick picture of the architecture over here we've got Fedora Commons and it has an API on it it has some internal data stores obviously it's all XML under the hood but these kind of operational data stores data content in Fedora Commons is extracted using g-search and put into a solar instance for searching so that provides the search capability and it's also extracted for OAA and goes out to Digital New Zealand so it's another way to access the material and then the existing Drupal CMS for Quake Studies is up in corner and that talks to Fedora Commons to get data in and out so that's the public interface for most people in general searching they're going through Drupal and it's talking to these back-end APLAs so what we introduced is another Drupal instance but we changed a few things around so one of the first things we did was use the Chuk library from Islandora and that's apparently a fan that Chuk is Canadian for Beanie the hat and so rather than develop a whole interface wrapper we just re-used that existing capability that Islandora Projectory provided us an open source Drupal itself is open source and since the time that Quake Studies was first put together the whole remote entities concept had been progressed by the Drupal community to the point where it was really a robust solution and so we re-used that one difference at the moment is if you do a search on Quake Studies you're going to the solar instance that's fed by Fedora Commons whereas for our purposes it made more sense to the instance to the Drupal for Skirt and the idea there is that when you enter the ID of an object in Fedora Commons it gets pulled across by remote entities into Drupal and once it's in Drupal you can do all of the wonderful Drupaly things like querying and forms and tokenisation all the kind of presentational stuff and management stuff that you'd expect from a CMS so it makes it a really powerful capability as soon as you've used the remote entities approach to effectively full Drupal into thinking it's dealing with local content you then get the full spectrum of all of the content management features even though the content itself is actually coming from an external system and there's a few tricks to that so for example we pull images across into the Skirt Learning Legacy project and display them we pull PDFs across and index them in that solar instance but we don't choose to pull videos across because many cases they're huge in size so if somebody requests a video through the Skirt Learning Legacy site we just stream it behind the scenes straight out of Fedora Commons rather than duplicating it in another location so just to give you some idea of how some of these things might look if we have a look at this quick studies is in a top collection within... oops that's not going to work today not online right now thing to get across here is that all of this content is in Quake Studies and is browsable in Quake Studies and then it's been pulled through into Skirt Learning Legacy and presented in a different way but it's the same material so one of the similar lessons arising from this one of the first would be that multi-stakeholder consortiums are hard work a lot of conflicting agendas and needs and so that was a constant battle for everybody but I think there's generally a fair amount of goodwill but we're still working on this and there's still people wanting to be satisfied re-use is possible I think this is a really good demonstration of re-use both in terms of architecture and code so using open source components from pretty much from top to bottom means that we can re-use this and the wider community can benefit as well and I think one of the other things that comes out of this is the demonstration that open source communities are delivering increasingly powerful capabilities I mean from Trupul as a CMS from Solar as a search system as Fedora Commons from an institutional repository point of view the capabilities are just continually accruing at a quite a substantial rate and getting really really impressive and that big problems like earthquakes can be big opportunities to experiment with new approaches and to learn new things so I should say thanks to Skirt for backing this and of course to University of Canterbury and to Catalyst IT for getting me here what's next the collection that we're talking about is available under the Quake Studies URLs so you can go there and visit that and we're working to put some of the code that we've built on GitHub so that other people can benefit and the Skirtland and Negasi site unfortunately is not live at this moment but it should be live any day now I hope just attributing a few images there and if anybody wants to follow up on the real technical nitty gritty you can talk to me afterwards or follow some of these links around how the remote entity stuff works these slides are available on GitHub and I'm happy to take any questions thank you for your time any questions? Thanks for that Jonathan and Chris in absentia if I go to write your archiving the actual design process that leads to actual rebuild projects or are you also archiving the design thinking that goes behind projects that don't see the light but for whatever reason are still archived as potential future ways of approaching specific problems Chris would probably have a better handle on the spectrum of stories that have been developed but I think there's a few hundred in play not all two hundred are currently ingested in the system but they span a real range so some of them are public relations campaigns about handing out chocolate fish at road works just to encourage people to behave in a civil fashion and to acknowledge that they were inconvenienced but there was a reason why they were inconvenienced and then there's design diagrams and solutions for fairly substantial underground works and for siphons and for all sorts of civil engineering things that I'm not that familiar with so it's a real spectrum of artefacts that have been captured does that answer your question? I guess the question is are you archiving what hasn't worked? You'd have to ask how many of those things that are put in there are examples of the experiments that didn't work out? I don't know from a PR point of view maybe not so much but these are things that work for us but there's no guarantee that they will work in other situations because of the requirements of the earthquake rebuild and the timeframes that they're working with that might have been working on different kind of different drivers that may not always be applicable to other cases but it would make sense I'm not sure enough about the content but I could certainly find out for you. I'm curious about the learning objects how they're translating stories into learning objects for the learning legacy to implement by the ITOs for example in their apprenticeship programs and what's the trajectory to make them real tangible? It would make a huge amount of sense to try to reuse some of these real world examples but that would be up to those training organisations that they see once it's public. Question? Can you just clarify the use cases for the two different front-ends? Sure. So the Quake studies front-end well I'll go to the so the front-end for script learning legacy is very tuned just to their stories and how to find their stories and the whole idea is that the search engines will pick this up so anybody who happens to be searching and underground infrastructure is going to land here. Quake studies is a much broader collection of material and there's all sorts of stuff on there ranging from individual blog posts that have been captured to videos from the Quakebox project and newspaper pages and a whole bunch of stuff so it's a much broader research platform whereas this is very focused on justice but the beauty is that we can do both with this infrastructure and we can provide the really wide general pool of content for people who are researching across a whole range of different kind of ideas or media and here we can provide something that's really focused on a very specific use case for a pretty specific audience. I should point out that James has been a fairly major player in all of this. Yeah, I'm James Smithies. I was involved in developing and project managing Quake studies and seismic. There's an interesting thing I used to call these satellite sites or third party sites and the goal on day one with seismic or Quake studies was eventually to have hundreds of these things and in a way it was to offer content provided something extra so when we first went to Fairfax media for instance and people said I'll never give you their content. Now we'll put your content into Quake studies but you could potentially have a third party site that showcase just your content so that it wouldn't be subsumed by all the other content within seismic. Turned out that Fairfax and most of the other companies that we talked to said Quake studies is fine and we don't need it and this is really the first use case that's appeared where a company has said no, we really need our own site and that's the only way that provides the motivation to give us to gift our content to you. So it's quite a nice sustainability feature I think for projects like this. So with it being the first of its kind in that way, do you have plans to monitor use or to close that feedback loop back from users in terms of future potential? Yeah there's definitely plans in Skirt they would like to see this you know picked up around the world they've received really good feedback so far I don't think I mentioned to be honest that it was a learning legacy project that came out of the London Olympics so you know this has been done before or something along these lines and so this is kind of probably a relatively quick follow up to that kind of thing but it's a pattern that could be replicated across a whole range of industries and subject matter areas so yeah it would be really good we did talk about ways of relating this material and all that kind of stuff but it's because it tends to be mainly looked at by engineers doing research and so on it's not necessarily something that's going to show up by following say a Twitter stream or Facebook links or anything like that so it's kind of a bit of a undefined question right now but it's definitely people are thinking of that That's what I'd like to see 50 or 100 of them involved and it's interesting that it's only four years after research online that we had our first programme so this wasn't long ago last chance for questions okay so I believe we are now on afternoon break back in Oceana and then we'll have the final session of the afternoon in soundings and thanks to all our speakers this afternoon and everyone who is asking questions and participating in the audience as well