Let’s ROR Together! Implementing Open Identifiers for Research Organizations across the Research Landscape
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 22 | |
| Autor | ||
| Lizenz | CC-Namensnennung 3.0 Deutschland: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. | |
| Identifikatoren | 10.5446/48011 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
20
00:00
Computeranimation
00:37
Computeranimation
01:08
Computeranimation
04:06
Computeranimation
05:02
Computeranimation
05:15
Computeranimation
05:45
Computeranimation
05:59
Computeranimation
06:50
Computeranimation
08:12
Computeranimation
09:02
Computeranimation
09:38
Computeranimation
10:49
Computeranimation
11:51
Computeranimation
12:01
Computeranimation
Transkript: Englisch(automatisch erzeugt)
00:05
Hello everyone. Thanks very much for having me here today. I hope you're all doing well. I am Maria Gould from California Digital Library, and I am the project lead for the Research Organization Registry, also known as ROAR.
00:20
So in this talk, I'm going to give a very short introduction to the project and share a glimpse into how ROAR is being implemented and then happy to hear questions and feedback afterwards. So let's see. For those of you who might not be familiar with the project, ROAR's aim is to develop an open, sustainable, usable, and unique
00:44
identifier for every research organization in the world, and I'll talk a little bit more in a moment about what that means. The registry officially launched in January 19 following many years of planning and collaboration by many stakeholders across the research infrastructure landscape to develop this vision for a registry of
01:04
organization identifiers. So some of you may already be familiar with other organization identifiers and might be wondering, why do we need another one? So let me just tell you a little bit about what makes ROAR different and what kind of prompted ROAR to emerge.
01:22
So first and foremost, ROAR emerged because the ability to identify research organizations has long been this missing piece in open scholarly infrastructure for quite a while. So kind of similar to the development of ORCID identifiers for researchers and DOIs for research objects, you know, that kind of paved the way to support
01:42
broad and comprehensive and easy identification of different aspects of the research landscape. But until ROAR emerged, we didn't really have an easy or an open equivalent for affiliations. In other words, a way to identify the research organizations that are linked to those researchers and those research outputs that could really help to complete that picture.
02:05
So a few things that set ROAR apart from other types of identifiers that have been used previously and are continuing to be used to identify organizations. One is its openness. So ROAR is non-commercial registry. The data is CC0, the code is open source, and ROAR IDs are interoperable with other
02:27
identifiers. So ROAR really isn't meant to be duplicative or, you know, to repeat what other kinds of registries are doing, but the point is to be able to provide open data and infrastructure that can be easily used by various stakeholders.
02:43
Another important distinction about ROAR is that it's specifically focused on what we call the affiliation use case. And that is specifically identifying the research organizations that are affiliated with research outputs to make it easier for institutions and other kinds of research administrators and stakeholders to track outputs
03:05
associated with a specific institution. So, ROAR's scope means that we're specifically focused on building an open top-level registry of organizations and not, for example, mapping every single department within a given organization. Not that that's not important,
03:23
but that's just not the purpose of the registry in this case. And lastly, an important distinction about ROAR is that it's being developed as a community-based project and focused on really receiving and taking in input from a broad community of adopters and other stakeholders. So,
03:45
in sum, ROAR is developed to fill this missing piece in the landscape and to provide an open and trusted piece of metadata that can specifically connect organizations to outputs and to researchers, and that can be widely and easily adopted by the scholarly community.
04:00
So, let me just take you on a quick tour of what's in the registry right now. So, there are currently about 98,000 organizations in the registry. I'm showing an example of the ROAR record for my organization, California Digital Library. We have ROAR IDs for all of these organizations
04:24
and additional metadata about each organization as well, including crosswalks to other types of identifiers. So, again, really reinforcing the interoperability of the ROAR data. We launched the registry with seed data from Digital Science's grid database.
04:41
That was sort of part of the community effort that launched the first prototype for the registry, and we're now in the process of building infrastructure and workflows to support independent management and curation of the registry over the long term. So, there are a few different ways to view what's in the registry and to interact with the data.
05:01
And one of those ways is just doing a simple search on the ROAR website. You could just enter the type of the organization that you are searching for. We also have an open API and we have a data dump that we make available as well and host on Figshare. We have
05:20
everything documented and available on the ROAR GitHub. I'll make sure that the link is included in the slides that get shared out. So, anybody is welcome to to use what's on GitHub and to interact with the API and the data. We have a few tools as well for manipulating ROAR data and reusing ROAR to
05:42
clean up affiliation strings. One of those is a reconciler that uses OpenRefine to match affiliations to ROAR IDs and clean up messy or inconsistent lists of affiliations. Another tool that we have available is affiliation matching functionality in the ROAR API,
06:04
which allows you to feed a long text string for an affiliation into the API and find a match in ROAR. So, we know that for those who are working with legacy content that might have a lot of free text affiliation strings associated with it,
06:22
these tools can be helpful in disambiguating and then tidying up that data. So, let me now turn to an example of how ROAR is beginning to be implemented in different scholarly systems to use a standard and consistent ROAR ID instead of, for instance, a free text affiliation or
06:45
another kind of identifier. So, one of those examples that I will share is the example of Dryad and Dryad relaunched its platform late last year and decided to implement a ROAR lookup for researcher affiliations
07:03
when researchers submit data sets to Dryad. So, it's a very simple kind of lookup that calls the ROAR API. So, a researcher submitting a data set can start typing an affiliation as you see here. The ROAR API is called in the background and presents the researcher with a list of potential matches
07:22
and then the researcher can choose the corresponding affiliation. So, then the ROAR ID for that affiliation can be stored in the background in Dryad's database. So, something that again kind of makes ROAR a little distinct from other types of identifiers is in this case, the researcher doesn't even know that an identifier is being collected.
07:42
So, it's a little different in that sense from an ORCID ID. But it's a really, it's a simple way for Dryad's system to collect this standard piece of metadata instead of a free text affiliation. And then, another important part of this type of integration is the ability to deposit this metadata in
08:06
other parts of our research infrastructure. So, for example, the data site metadata schema has recently been updated to include support for ROAR IDs for affiliations. So, that means that when a data site member like Dryad
08:25
collects ROAR IDs, they can deposit those IDs as part of the rest of the metadata that they include when they register DOIs with data sites. So, this means that all of the research that is searchable and available in data site can be associated with a specific
08:43
affiliation using ROAR IDs. So, we'll be seeing a similar kind of implementation in the Crossref metadata scheme this year. So, that's going to be another important way in which ROAR IDs can become embedded in the scholarly landscape.
09:01
We are tracking other kinds of integrations of ROAR in various systems on the ROAR website. I might put the link there for anybody who would like to follow along. If anyone is interested in doing anything with a ROAR integration and would like to talk about it, I am
09:21
always welcome to set up a time to talk and to provide information and support. And so, I'll just close with some information about what else we're working on this year and how you can get involved in the project. So, as I mentioned, ROAR launched last year in 2019. And so,
09:45
then, you know, since launching the registry, we've been focused on a few key aspects of the project to really drive it forward. And one of those is developing infrastructure and workflows and policies to support independent curation of the registry data.
10:03
So, as part of that, developing infrastructure independent of grid is a key piece. And then, also developing appropriate community-based workflows is a core component as well. There is another key theme for
10:21
this year and beyond, which is driving adoption and integrations and encouraging ROAR IDs to be collected in various systems as widely as possible. And then, lastly, focusing on the long-term sustainability of the project and how we can keep it going while also ensuring that the registry data can
10:43
be free and open for all in perpetuity. So, if anybody is interested in getting involved with the project, we greatly encourage it. It's a community-based project and really depends on engagement
11:02
and support. So, we do have a community group that meets every other month or so for calls to discuss updates and share feedback and start, you know, work on various implementations and side projects and things like that. We also try to surface information and updates about what ROAR is working on through various communication channels, including
11:25
the ROAR website and blog. We have an open Slack group that anybody can join. I mentioned that we make all of our code and documentation available on GitHub. And then we have Twitter feed and mailing list as well.
11:42
So, a lot of different ways to stay up to date on the project and different ways to become more involved as well. So, that is my brief introduction to the ROAR project and how ROAR IDs are being implemented. And again, appreciate the opportunity to share that with all of you today.