 I think we're going to get started. We have a good half hours worth of content for you. So without further ado, we'll get going. If I sit down, can you still hear me? OK, great. So I'm Donna Born-Tyson. It's my honor to be the Canadian Association of Research Libraries' president right now. And I'm the University Librarian at Delhousie University. And this is Lee Wilson, the Portage Service Manager, seconded from ASENET, where he's an RDM consultant. And the invisible third person here would be Corey Davis, but he had trouble getting here. So Lee is going to cover Corey's part. As best I can. Yeah, so we can go to the next slide. So I'm going to say a few very brief words about Portage in general and then turn things over to Lee. So Portage is a national coalition of the thrilling and launched by the Canadian Association of Research Libraries. Before 2015, we had a pilot project that we started with all of the regional academic consortia in Canada. So it really has been a very collaborative national effort from the beginning. So the whole point to Portage is to develop capacity for RDM in Canada. And we're doing this through a network of expertise, which is a pan-Canadian group of about 90 stellar individuals now from 35 institutions. I should say we have some of the coalition of the thrilling in the room. I would like to recognize our past president, Martha Whitehead, from Queens. And our executive director from Carl Susan Hague is here. And we also have at least one board, Carl member. Saw Melissa. Oh, did Melissa get away? There she is. And I think we probably have some experts in the room. Anybody, one of the Portage experts? I'm looking at a couple, but they're being shy. Yeah, you're being shy. There we go. And I guess we have some fans too. And then we're also developing some infrastructure platforms. And we're working with other stakeholders, such as the National CIOs group, the funding agencies, and our National Compute and Network group. So Portage, we have six different expert groups working with various other groups. So data management planning, curation, data discovery, preservation, training, and research intelligence. And then we have a number of working groups as well, including one for Dataverse. And Furder, and Lee will tell you more about Furder later. So I can't remember what it stands for. And we have two more working groups coming. So the two that are coming, the research data management strategy working group is going to be working to help our funding agencies in Canada as they roll out their new policy, which is going to gently insist on an institutional strategy. Data management plans being written by every researcher who's applying for funding. And data set deposit. And then the other group is going to be working on ethical treatment of sensitive data, which as we all know is a huge unsolved issue. So if this was a longer presentation, we would spend more time talking about funding and governance as well. But in the interest of time, I'm going to turn things over to Lena. Thank you, Donna. Yeah, as Donna said, we have a lot to cover in about 30 minutes. So this is sort of Portage from 10,000 feet up. So we'll try to sort of move through as quickly as we can so that we can respond in more detail to questions. So as Donna indicated, in addition to coordinating the network of expertise, Portage is also creating platforms and associated services for our DM in Canada. So the first I'd like to speak to you about today is the Portage DMP Assistant and the website's there. So this was launched in 2015. And it's a free web-based tool for creating data management plans. It's discipline agnostic, which means that it's sort of generic for research data management best practices in general. And institutions, funding agencies, and research groups can create custom templates. So for example, in my past life as an ocean data manager for a National Oceans Data Research Initiative, I helped to create an ocean-specific template for that group. And as you can imagine, it will be quite useful for funding agencies as they want to create policies that specifically adhere to their requirements. So right now, we have sort of DMP Assistant 1.0. 2.0 is coming down the road, and plans are in place to merge our code base with DCC and UC3 and feature the best of both systems. And also in the future, we're hoping to see a DMP repository that could be used by librarians, research services, funders, and researchers to show, for example, what an exemplary data management plan might look like for a particular discipline or funding agency. And this would be an opt-in service. So researchers could still create data management plans on this platform and keep them private. Taking a broader view in my role as the Portage Service Manager, I see the DMP Assistant as one of the first interfaces between the researcher and library RDM services as a bit of a canary in the coal mine that's going to signal future data storage needs before any data is actually collected. So the next platform I'd like to talk to you about is the Federated Research Data Repository, which is known fondly as FURTER. So this is a project that has many stakeholders, as Donna indicated. The primary stakeholders are Carl Portage and Compute Canada. Compute Canada is a national provider of research infrastructure for high performance computing in Canada and undertook much of the development on the FURTER project, where Portage has provided steering, expertise, and operational support. In addition to the Portage Compute Canada partnership, we have a third technology partner, Globus, which is an American company that develops technical solutions for RDM, with FURTER making use of its file transfer application that allows researchers to easily and quickly transfer large data sets. And we've also had collaboration with university partners. So for example, our search interface on FURTER makes use of code supplied by UBC, which supports their open collections user interface. So FURTER was created to fill what we saw as a gap in a repository that was scalable and capable of adjusting large data sets, as well as improving the overall discoverability of Canadian research data. Right now, FURTER is in what we're calling limited production, which means that anyone can use the website there to search for and download data. So that's data held across a variety of different repositories in Canada, as well as data stored natively in FURTER. But depositing is currently limited to a select number of partner research groups who are working with us to help us refine our operational and service models, while still having their data be made publicly available. So the first part is discovery. So FURTER harvests metadata records from existing Canadian repositories to make data discoverable from a single platform. Right now, we're harvesting 31 institutional government and domain-specific repositories, which translates into around 125,000 data sets. These repositories were identified by work done by one of our expert groups, the Data Discovery Expert Group, and specifically the Collections Working Group, which identified 170 repositories in existence in operation across Canada, and have made suggestion for the FURTER team for harvesting based on things like the availability of data, the use of standards by those repositories, and which technologies they're using. So over the next year or so, we're gonna be continuing to expand our harvester based on the recommendations that this group makes. So the goal of the Discovery layer is to break down silos for research data in Canada to improve discovery, but also to augment the existing ecosystem by driving traffic back to those repositories. So all the data that we have hosted, or the metadata, rather, that we have hosted onto FURTER links back to all the individual repositories. And so the idea was we didn't want FURTER to just be the 171st Canadian repository. We wanted to actually do something to help link the repositories together. So in addition to Discovery, FURTER is also a repository in its own right. It's able to ingest large data sets using the Globus file transfer technology, and it's therefore fulfilling a need that's not currently being filled by the institutional repository landscape. So we don't want to supplant existing repository operations, but we do need a home for researchers who don't have a local option, or whose local option doesn't support their data needs. So for example, the sort of massive, nationally funded research data projects that are generating more data than our campuses can currently handle. FURTER was designed with scalability in mind, so adding more storage is trivial, and storage could be managed centrally by infrastructure providers, so for example, Compute Canada, or managed locally and connected into the national platform. We're currently still exploring different models for what these kind of shared storage services are going to look like, but Portage does plan to develop a minimum standard for the type of storage hardware that can be used in FURTER. So in addition to Globus integration for data transfer, FURTER also has Archivematica integration, and this allows for preservation processing. So Archivematica is a tool used by archivists and digital preservation specialists that normalizes data files into preservation-friendly file formats, so for example, neutral or non-proprietary formats that the field hopes and we hope will be reusable in the long term, and these packages are called APES, or Archival Information Packages. Preservation processing in FURTER is scalable with automatic APE generation occurring for datasets up to 300 gigabytes or 25,000 files, and APES are expected to feed into existing preservation service providers, like for example, the Ontario Library Research Cloud and COPPL's Digital Preservation Network, and I'll talk more about our Preservation Service Provider Plan in a minute. So this is where Corey would take over if he was able to be here today, so I will do my best to speak to his slides. So I'm going to talk to you first about the Dataverse North Working Group. So Portage is working to cultivate a community of practice centered upon Dataverse, which is a web-based research data repository developed at Harvard that many people in the room will probably be familiar with, and as most of you will also likely know, Harvard makes Dataverse freely available across the world, but jurisdictional considerations of various kinds have resulted in a number of countries running their own Dataverse networks, so there's a good example in the Netherlands with Dataverse.nl, and in Canada, several research libraries also run their own Dataverse instances. However, until now, a lot of this work has been taking place in relative isolation, and the idea of Dataverse North is to change that. So this broadly representative working group nurtures a community of practice throughout the Canadian University libraries and is exploring how to best coordinate existing infrastructure at the national level. We're also building mechanisms to enable hosting institutions to offer Dataverse as a service to all University libraries and researchers in Canada, and to standardize this service across the country. We also work to support users, and these are both researchers and librarians, and we're thinking about how to best pool our collective resources to guide feature developments such as advanced data visualization techniques and capabilities, as well as integration with other services, such as the digital preservation processing and storage services from large research libraries, but also from regional academic library consortia. So we're currently focusing our efforts in three key areas. The business models working group is looking at a national service centered on one, or perhaps several large research libraries, and to provide common terms of service, SLA agreements, et cetera, regardless of the hosting institution or region. The training group is coordinating capacity building across the country by working together to create training materials and workshops, and the metadata working group is pooling resources and their collective expertise to support the creation of specific metadata templates and the implementation of recommended file structures among other things. And so all of these groups connect with the broader Dataverse community via the Dataverse North working group to ensure that the Canadian efforts we're undertaking are themselves not happening in isolation, so we're making sure that we are working with the international Dataverse communities. As we move forward in the next phase, we'll be focusing on connecting a national Dataverse service with a number of preservation service providers within the Canadian research library community. Specifically, we're gonna be looking at how to help individual institutions promote the service and advocate for further investments in our DM infrastructure and expertise not only at their campus, but regionally and nationally as well. So now to switch gears a little bit, I'm gonna say a few words about our preservation services, as mentioned in the previous slides. So this work is enabled through Portage's Preservation Expert Group, or the PEG, and it's focused on the long-term preservation of research data in Canada. The PEG is currently finalizing a series of recommendations based on the Open Archival Information System, or OAIS, reference model, pictured here. This model describes an organization of people and systems that has accepted the responsibility to preserve information and make it available for a designated community. So in this model, it's sort of an organizational model, but we envision this organization to be a lot more distributed or disaggregated, where Portage provisions some platform infrastructure, for example, like Furter and Dataverse, but in terms of long-term preservation, provides more of a coordination role for most of these functions via allied organizations, such as large research libraries or regional library consortia that are currently providing preservation services. In this way, Portage is going to act as an administrative capacity to oversee the disaggregated archive by undertaking such activities as implementing new technical organizational strategies that reflect in the external environment, coordinating activities between and among preservation service providers and prescribing basic standards for those providers. Full details will be available soon, but for those who would like a sneak peek at this innovative approach, you can access our poster abstract for the 2018 International Digital Curation Conference, or IDCC, from our slides, which is going to be available at CNI's Open Science Framework Collection. So now to bring everything together. Our vision statement for Canadian RDM services is that we want a seamless, equitable access for RDM platforms and services for Canadian institutions and researchers. And the following slide is going to show you a little bit about how we see the various services fitting together. So this vision is informed by the steps of the research data lifecycle that moves from the conceptualization of research to collection, to its eventual dissemination and preservation. So the first window that we're looking at here is planning and training. So this is really the first sort of interface with the researchers. So here we'll develop tools for data management planning, like the DMP assistant. We're currently developing through our training expert group, training modules. So down the pipeline, we have RDM 101, which is sort of a basic overview. DMP 101, which is again, a basic overview, but with a focus on DMPs. Tools and platforms related training. So for example, the Dataverse North training group is specifically focusing on training modules for Dataverse, and advanced topics for various audiences. So for example, advanced research data management training for librarians as a sort of a train the trainer sort of thing. And in addition to this, we're looking to provide on the ground training at the institutional or regional levels. So the second part is repository services. So with repository services, we're dealing with active data, as opposed to archival data. And by this, we mean published data sets possibly related to publications that have a DOI for persistent identification. So Portage will operate or support national, institutional, or domain-specific repositories, FURTER and Dataverse are both planned to operate in this space as each fills a particular need. So FURTER at the moment is the one that can handle sort of the big data sets and can be used to support large research projects or organizations who are capable of supplying their own storage infrastructure. Dataverse allows for the creation of community spaces on its platform, and allows for institutional branding and a certain degree of autonomy in terms of curation. So for example, one Dataverse instance can host multiple institutional Dataverses that can operate with a certain amount of autonomy based on how the services planned out. And Dataverse is also well-suited for smaller data sets allowing for meta-data description at the file level natively. So what we're planning for down the road between these two is development towards better integration between the two platforms, and this is ongoing. And also in terms of FURTER, the platform that we have more control over the development of, so for example, Dataverse is a broad global community effort, whereas FURTER is something more local to Canada. So we have a little bit more control over how the development will go. Is one of differentiant, differentiation, I should never put that word in here, rather than competition. So basically we want to develop towards not sort of just trying to have feature parity between the two platforms, but rather if Dataverse is filling a particular need right now. So for example, as a publication repository, and down the road, it is able to do the type of things it can't currently do. So for example, hand a large data sets and scale to a national level, then FURTER will be looking to develop in alternative spaces to continue to help fill gaps. So for example, one of the spaces we're looking into right now is this idea of a working store where we'll be working with researchers on data that's not yet ready for publication and sort of closer to the area, a point of collection, to get it into a data management system so that we're not trying to do sort of curation interventions at the end of a research data lifecycle. Which as everyone in this room can probably attest to is problematic. So we're also going to continue to expand our harvesters so that all Canadian repositories will be feeding into this national data discovery layer. And the repositories have been figured in the preservation expert group's plan as an appropriate place in the data lifecycle to produce apes through preservation processing. Although PSBs may also decide to create apes in-house. So in addition to repository services, we also have preservation service providers. So this is going to be a distributed network of preservation service providers that will provide long-term archival services and agree to work under a model of shared standards and best practices. Portage would help to define a minimum set of standards, a set of criteria that would define a PSB. With one requirement expected that PSBs expose their metadata publicly using an open archive standard that will allow us to harvest their metadata so that it's also discoverable in the national data layer. Overarching all of these services are infrastructure providers. So these could be local. For example, universities. These could be regional. So for example, regional library consortia or national. So for example, national infrastructure providers like Compute Canada. And Portage plans to operate services through agreements with infrastructure providers rather than direct ownership of infrastructure. And at the bottom, you can see the network of expertise is sort of the glue that holds all of this together. The network of expertise is the frontline staff that's going to be interfacing with researchers. And also the coordination bodies that help set standards and best practices. So as well as exploring current trends and emerging trends in research data management. So the plan is for the network of expertise to be regionally distributed but locally coordinated with parts of FTEs embedded in research universities across Canada serving both local and national needs. The upside to this approach is that we'd be retaining and leveraging existing researcher relationships as well as having access to expertise in specific subject domain areas at a national level. So not every research library needs to have the full suite of curation expertise because there will be national support and expertise. So here's an example of what the different components in each of these aspects currently look like. So we have some of the tools I mentioned today. We have the various repository services currently in operation with Archivematica providing the preservation processing layer. And then preservation service providers are represented by the four regional consortia here because we expect that in this disaggregated network of preservation service providers regional consortia will be well placed if they so desire to have their own preservation network. So in terms of the next steps, first, our first challenge we want to meet is integrating all these services onto a single user facing platform. So the federated services need to be easily accessible to an end user. We don't want them distributed across the internet on a bunch of different portals and platforms that librarians and researchers have to interface with. So we want ideally RDM start to finish on one platform. Also, we need to understand which services are best centrally managed and which ones should be regionally distributed with just national coordination. And we need to better understand the kind of agreements that are required to make this work. Right now we have a solid bedrock with Carl and with Portage with three out of the four regional consortia having MOUs with Portage. And we see this as a way forward for building for building these services off the existing frameworks. Also securing sustained funding for RDM in Canada. So the federal government has signaled that RDM is an essential component in our national research strategy. And we've been working with them as part of a consultation process to determine what a sustained federal funding stream for RDM and other research related activities would look like. In addition, the three Canadian tri-agencies are currently engaged in what they're calling a pre-consultation process for a new data management policy that could require researchers to have data management plans, institutional RDM strategies and data related publications to be disseminated through open repositories. So we're also going to be working with the funding agencies to try to prepare the ecosystem for their policy requirements. Basically laying the groundwork that'll allow institutions and researchers to conform with these forthcoming policy requirements. So you've heard a lot today and it's quite a challenge. That picture is a bit of a misnomer because there's not sort of just one group moving forward. There's a lot of groups under that canoe. But I think as long as we're moving in the same direction, we should be okay. Thank you.