PIDapalooza 2018

With so many to choose from, sometimes deciding on which persistent identifier scheme to use can be difficult. At the Publications Office of the EU we are responsible for identifying content coming out of the EU institutions. We are a registration agency for established international identifiers such as ISBN, ISSN and DOI as well as providing support for our own EU-specific persistent URI scheme. When we received a request from one of our clients for help in identifying the content of a historical archive, we asked ourselves the question: what would be best in this specific case: the doi or our own persistent URI scheme, data.europa.eu? So we have decided to launch a small study to compare the two approaches according to the requirements of this specific case. The results will be known by the end of the year and PIDapalooza would be the ideal forum to share the conclusions.

I have the scars, here's how I got them. Working with a disparate and sometimes volatile volunteer community; soliciting and processing data donations from professional organisations; correcting the experts; Wikidata IDs as PIDs; Wikidata as the hub of all PIDS; when is a PID not a PID?

This would be a variation on a talk I gave a year ago to a PID workshop in Garching. Twenty plus years later and we are still discussing the topic. Seems easy enough, so what is left? I can take us back to the early 90s and show what has changed and what hasn't and what is left.

Like many other academic institutions, Leiden University aims to ensure that the various entities that can be distinguished within the international ecosystem for scholarly communication, or Open Science, can be identified effectively and uniquely via PIDs. Within Leiden’s repository infrastructure, all academic publications and all digitised objects from the Special Collections department and are assigned Handles. The institutional data management policy stipulates explicitly that researchers ought to deposit their data sets in trusted data repositories which assign PIDs. As presented at the first PIDapalooza, a project was also conducted in 2017 to stimulate researchers to create an ORCID id, and to associate these identifiers with their research outcomes as much as possible. Although the ORCID project did encounter a number of difficulties, the various activities that were organized to raise awareness of the benefits of ORCID eventually led to a big increase in the number of author identifiers. Leiden University is currently incorporating these identifiers within its various information systems, in such a way that these identifiers can actually create added value, both for machines and for human beings. The various services that have been built on top of ORCID, for instance, can effectively help researchers to simplify administrative tasks. A large number of researchers at Leiden use their ORCID id to ensure that the data in the CRIS is accurate and up to date. To communicate the fact that these IDs are available, Leiden University is currently developing a visual system, in which the various types of identifiers are represented using specific icons. As one of the founding members of the IIIF consortium, Leiden University is also trying to embed object IDs and author IDs logically within IIIF manifests. The various identifiers that have been created can be used, moreover, to create network visualisations which display research results within their broader context. Such graphs can clarify, for instance, that publications are based on other publications, or that some data sets are connected to specific publications. Overall, these different activities can improve the visibility of research outcomes and they can ultimately enhance the impact of academic research.

Dreams and realities of PID adoption in European research information management. A story of PID tactics.

Team members from OCLC Research, in conjunction with LIBER, have been examining the adoption and integration of PIDs in the evolving ecosystem of research information management in Finland, Germany, and the Netherlands.

In this presentation, I will share our findings regarding the dynamics around PID adoption, in particular incentives and barriers to adoption.
Are some tactics more successful than others? Come and find out.

Full report: https://doi.org/10.25333/C32K7M 

Presentation: https://doi.org/10.6084/m9.figshare.5727011.v1

We have stories in the area of molecular sciences and particularly chemistry to tell. Since the online era started around 1994, scientists publishing stories in peer reviewed journals have had the option of including “supporting information” with their articles. In its present form (PDF files), this can be characterised as Unfindable, Inaccessible, Non-interoperable and not Reusable, the very antithesis of FAIR. Our stories are about how we set about to convert this mass of information and perchance even data, into a fit-for-purpose modern scientific resource. Absolutely central to the concept of FAIR was the PID and the metadata associated with it. To fully exploit this metadata, we decided we needed to create a repository designed to utilise the DataCite V 4 schema, having tried other existing repositories and largely failed to achieve our objectives. Our stories are built on the depositions into this repository over the last 18 months or so by increasingly motivated and engaged researchers. Doing so has enabled us to devise a rich variety of searches based on our metadata and we have even assigned PIDs to these searches to make it easier for scientists to try them out. Our stories will tell how this combination of FAIR data enhanced by rich searches are creating a new way of sharing scientific results and data across a broad swath of molecular sciences. Our stories still lack some kinds of information, such as PID+metadata for the instruments and software that is used to create our rich datasets and we look forward to the day when even more types of “research objects” will carry PIDs. You can see (interact with) the presentation at PID https://doi.org/ccs4

Scholix is not one of the PID Rock Stars. We are just a bunch of people trying to do stuff with PIDS; specifically link the PID of a dataset with the PID of piece of literature and say what the relationship is. How hard can that be? Bl**dy hard. This is the backstage view of PIDs from the Blood Sweat and Tears of roadies. And we have learnt a lot about persistent identifier types; resolving identifiers, URI vs non URI forms, what metadata is useful with a PID for our purposes, multiple identifiers for the same thing, de-referencing or not de-referencing PIDS and lots more.