Open Funder Registry on Crossref

RORing ahead: using ROR in place of the Open Funder Registry

Rachael Lammey — Tue, 30 Jan 2024 00:00:00 +0000

A few months ago we announced our plan to deprecate our support for the Open Funder Registry in favour of using the ROR Registry to support both affiliation and funder use cases. The feedback we’ve had from the community has been positive and supports our members, service providers and metadata users who are already starting to move in this direction.

We wanted to provide an update on work that’s underway to make this transition happen, and how you can get involved in working together with us on this.

Overall, we are building more comprehensive support for ROR into Crossref’s services. Some of this work is specifically to support using ROR to identify funding organisations in place of funder registry IDs. We have a number of parallel, complementary projects underway to support different elements of this work:

We are evolving our metadata schema so that we can collect ROR IDs in places where we currently support the collection of Funder IDs.
We are analysing the coverage of Funder ID to ROR ID mappings and testing the way we expose them in our APIs.
We are developing new matching strategies to match text strings to ROR IDs.

1. Schema updates

Everything flows from being able to get ROR IDs into the Crossref metadata!

We are evolving our metadata schema so that we can collect ROR IDs in places where we already support the collection of Funder IDs – for instance, in the funding section of the metadata for works and in the funder section for grants.

We’re working with members and service providers so that they can try sending us this data via a pipeline our Labs team has built to test schema updates before they go live. We are actively recruiting members to help us test our new pipeline by providing sample XML for registration. Planned metadata inputs and outputs are detailed in Including ROR as a funder identifier in your metadata (metadata prototyping instructions), we’d encourage you to provide feedback on these in the document, ideally in the next two weeks. We’re aiming to release an updated schema that supports these changes in Q1 2024.

2. Modelling ROR ID/Funder ID mappings in our metadata model

We have integrated the ROR registry into our evolving metadata model, and we have started work to integrate the Funder Registry. The aim is to create more flexibility in how Crossref’s metadata can be supplemented and queried, and give more clarity as to which party asserted or created a metadata element.

We’re working on an early iteration of how the model handles ROR IDs, funder IDs and their equivalencies. Once we have something to share, we’ll welcome community feedback on this approach and on the metadata model in general.

3. Developing new matching strategies to match text strings to ROR IDs

Ideally, everyone would always use persistent identifiers to exchange information about contributor and awardee affiliations, organisations related to works, as well as funders supporting the research. In practice, this information is often exchanged as data without identifiers, such as affiliation strings (e.g. “University of Virginia, Charlottesville, VA”), funder names, or even funding acknowledgements (e.g. “Funding and support generously provided by the Ford Foundation”). In such situations, a good metadata matching strategy can help map these to persistent identifiers.

Currently, we are focused on developing reliable strategies for matching affiliation strings to ROR IDs. In the future, we will adapt the strategies to support funder names and funding acknowledgements as well. All the strategies will be rigorously evaluated using real-life data. We will make the strategies, as well as the evaluation datasets and evaluation results, publicly available for anyone to use. If you are interested in collaborating on the development or the evaluation of the matching strategies, please get in touch!

In the future, we might also apply some of the new matching strategies at Crossref, to the metadata our members send us. This would allow us to insert matched identifiers to the metadata to better connect organisations with other items in the scholarly record. We already have a process that matches the names of funders supporting research against the Funder Registry and enriches the metadata with matched Funder Registry IDs. Developing and evaluating reliable matching strategies will allow us to modify this process to use ROR IDs instead, and extend it to support other use cases, such as contributor affiliations.

What will the transition mean for you?

We do recommend that you begin looking at what it will take to integrate ROR into your systems and workflows for identifying funders. Talk to your service providers about this to ready them for this change. To reiterate the point from the earlier post, in the short term, and even in the medium term, Funder IDs aren’t going away and the Funder IDs will continue to resolve – they are persistent, after all. Eventually, however, the Funder Registry will cease to be updated, so any new funders will only be registrable in Crossref metadata with ROR IDs. Legacy Funder IDs and their mapping to ROR IDs will be maintained, so if Crossref members submit a legacy Funder ID, it will get mapped to a ROR ID automatically. Note, too, that Crossref is committed to maintaining the current funder API endpoints until ROR IDs become the predominant identifier for newly registered content. We also know that there are questions that we’ll want to tackle with the community as we all make progress, some we know and some we don’t know. With that in mind:

Tell us what you need!

We want to hear from you! We have set up several channels of communication meant to ensure that you can tell both ROR and Crossref what will make this transition easier for you and that you can get answers to your questions.

First, we are conducting a series of Open Funder Registry user interviews designed to deepen our understanding of where Funder IDs are being used in workflows and systems. Write community@ror.org if you’d like to participate in these interviews to show and tell us how you’re using Funder IDs.

Second, in 2024, we will be running a follow-up to the funding data workshop we ran in June 2023. Please get in touch if your organisation would be interested in participating in the discussion.

Open Funder Registry to transition into Research Organization Registry (ROR)

Amanda French — Thu, 07 Sep 2023 00:00:00 +0000

There is some overlap between the Open Funder Registry and the Research Organization Registry (ROR), and funders and publishers have been asking us whether they should use Open Funder Registry IDs or ROR IDs to identify funders when they appear in both registries. We aim to merge the two registries over time. We will ensure Crossref members can use ROR to simplify persistent identifier integrations, to register better metadata, and to help connect research outputs to research funders.

Just yesterday, we published a summary of a recent workshop between funders and publishers on funding metadata workflows that we convened with the Dutch Research Council (NWO) and Sesame Open Science. As the report notes, “open funding metadata is arguably the next big thing” [in Open Science]. That being the case, we think this is the ideal time to strengthen our support of open funding metadata by beginning the transition to ROR.

Comparing the features of ROR and the Open Funder Registry

Let’s look at some of the similarities and differences between the two registries, including their history, features, scope, and usage, since there are important nuances and distinctions that are helpful to understand.

Overview

ROR	Open Funder Registry
Launched in 2019	Launched in 2013
Primary use case is contributor affiliation	Primary use case is funding acknowledgement
105k+ records	35k+ records
CC0 data	CC0 data
REST API	REST API
Free to use	Free to use
Entire registry downloadable as JSON and CSV	Entire registry downloadable as RDF; funder names and IDs downloadable as CSV
Records contain mappings to other IDs	Records do not contain mappings to other IDs
few organisation relationships and hierarchy	multiple organisation relationships and hierarchy
organisation level with no funding programs/schemes	organisation level with some funding programs/schemes
8 organisation types	2 funder types, 8 funder subtypes
Open source code and multiple open-source tools available	Open source code
Web-based registry search	Web-based search for works in Crossref associated with each Open Funder Registry ID
Web-based landing pages for each ROR record	JSON landing pages for each Open Funder Registry record
Updated monthly	Updated bimonthly
Public curation process	Private curation process
Anyone can request changes and additions	Anyone can request changes and additions
Stable financial support (Crossref, DataCite, CDL)	Stable financial support (Crossref, Elsevier)
Beginning to be supported in funding and publishing workflows	Well supported in most funding and publishing workflows
Currently used by 260+ Crossref members ¹	Currently used by 2100+ Crossref members ²

History

The Open Funder Registry was launched as FundRef over a decade ago to enable the community to cite research funding and support and assert it within the scholarly record, acknowledging the organisations granting their support. Elsevier generously donated the seed data for the Open Funder Registry and has managed its curation for the last ten years, while we have maintained the technical operations and promoted community adoption of the Open Funder Registry.

The Research Organization Registry (ROR) was introduced in 2019 by the California Digital Library, DataCite, and Crossref to enable the community to cite contributor affiliations and assert them within the scholarly record, acknowledging the organisations that housed or performed the research. Digital Science generously donated the seed data for the Research Organization Registry from its Global Research Identifier Database (GRID) initiative, and Crossref, DataCite, and the California Digital Library have contributed labour and resources to turn ROR into a mature, independent, freely available service.

Scope

One key difference between the registries is that ROR has always included funding organisations, and ROR records have always included mappings to Funder IDs where available, while the reverse is not true: the Open Funder Registry includes only funding organisations, not other kinds of organisations, and Open Funder Registry records do not currently include mappings to ROR IDs or other identifiers. It therefore makes sense to expand our initial contributor affiliation use case for ROR to include the identification of organisations that fund and support research.

Usage

More Crossref members use Funder IDs than use ROR IDs, to be sure. You can see from the table above that the number of Crossref members using Funder IDs in Crossref records is higher by almost a factor of 10 than the number of Crossref members using ROR IDs in Crossref records. But note too that the current rate of adoption is far higher for ROR than it is for the Open Funder Registry. Since January of 2022, we’ve seen a gratifying number of publishers and service providers beginning to use ROR identifiers for contributor affiliations in Crossref. In the last year, the number of Crossref members depositing ROR IDs has increased by 356%, while the number depositing Funder IDs has increased only by 12%. As evidenced by its ballooning API traffic, too, with more than 20 million requests last month,³ ROR is clearly being used by many scholarly research systems for many purposes. The more systems that use an identifier, the more valuable that identifier becomes as a vehicle for exchanging information.

ROR’s primary use case is to identify contributor affiliations and is already being used by funders. Nineteen funding organisations are depositing ROR IDs in their grant records with Crossref to denote principal investigator affiliations,⁴ and, following a meeting of the our Funder Advisory Group last month, all eighty funder members are primed to start using ROR IDs to identify themselves in grant records.

Tools and services

Both the Open Funder Registry and ROR have open data and open source code, but we think that our suite of free and open source utilities for ROR gives it an advantage. We know that publishers and their service providers have ongoing challenges in collecting and matching funding information from authors and in validating Funder IDs. With our extensive ROR toolkit, publishers and their technology providers who adopt ROR will be in a better position to improve the accuracy of funding acknowledgements in metadata, which can in turn enable the development of reliable analytics, tools, and services for funders, regulators, research facilities, and the public.

Crossref has built tools based on OpenRefine for both the Open Funder Registry and ROR: the Open Funder Registry Reconciliation Service and the ROR Reconciler are both useful ways to clean messy data. ROR, however, also offers a much-used API endpoint that helps match organisation names to ROR IDs, and several third parties have also developed and shared open source matching tools and services for ROR. Crossref is also collaborating on new strategies for affiliation matching that will improve connections for funding acknowledgements.

Community engagement models

The Open Funder Registry has been curated for over a decade through time and expertise generously donated by Elsevier and is community-governed by Crossref and it’s membership and board. ROR offers more transparent community involvement and is jointly governed by Crossref, DataCite, and the California Digital Library. ROR is openly curated and is aided by a global Curation Advisory Board of volunteers.

What will this mean for you?

The many organisations whose tools, services, and workflows have been architected to use Open Funder Registry (OFR) IDs will find this transition a challenge, and we don’t want to make light of that issue. Over the last ten years, we have encouraged the community to adopt Funder IDs, and the community has demonstrably recognized the benefits of doing so. Publishers have put a great deal of time, thought, and effort into collecting funder data and including it in Crossref metadata, and they have built internal reports and workflows around the Open Funder Registry. Crossref is committed to making the transition from the Open Funder Registry to the Research Organization Registry as simple as possible for the community.

If you are not already using the Open Funder Registry and are planning to begin standardizing funding data, we recommend that you use ROR to identify funders. If you are currently using the Open Funder Registry in your systems and workflows, don’t worry! In the medium term, Open Funder Registry IDs aren’t going away. Eventually, however, the Open Funder Registry may cease to be updated. Funder IDs and their mapping to ROR IDs will be maintained, so if Crossref members submit a Funder ID, it will get mapped to a ROR ID automatically. Note, too, that Crossref is committed to maintaining the current funder API endpoints.

In short, if you are already using Funder IDs, you can and should continue to do so. However, we do recommend that you begin looking at what it will take to integrate ROR into your systems and workflows for identifying funders as well as affiliations.

We face challenges in this transition, too. Of these, we think the largest will be (1) completing the reconciliation work involved in mapping Funder IDs to ROR IDs, and (2) updating Crossref’s schemas, APIs, and deposit tools to support ROR IDs in many the ways we currently support Funder IDs. We’ll discuss both of these challenges in future blog posts.

Tell us what you need?

We want to hear from you. You can use our Community Forum talk to us about the Crossref Open Funder Registry or contact ROR staff at Crossref via our request form. You can attend online Crossref events, including ROR-specific webinars to get updates from us and ask us your questions.

One of the major messages we’re already hearing from funders and publishers is expressed in yesterday’s post on open funding metadata: “While many concluded that there was still a long way to go to solve the many technical challenges related to funding metadata, attendees were unanimous on its importance.” We look forward to beginning this important work together.

Open funding metadata through Crossref; a workshop to discuss challenges and improving workflows

Hans de Jonge — Wed, 06 Sep 2023 00:00:00 +0000

Ten years on from the launch of the Open Funder Registry (OFR, formerly FundRef), there is renewed interest in the potential of openly available funding metadata through Crossref. And with that: calls to improve the quality and completeness of that data. Currently, about 25% of Crossref records contain some kind of funding information. Over the years, this figure has grown steadily. A number of recent publications have shown, however, that there is considerable variation in the extent to which publishers deposit these data to Crossref. Technical but also business issues seem to lie at the root of this. Crossref - in close collaboration with the Dutch Research Council NWO and Sesame Open Science - brought together a group of 26 organisations from across the ecosystem to discuss the barriers and possible solutions. This blog presents some anonymized lessons learned.

There is no Open Science without open metadata

The interest in the potential of this open-source funding metadata seems to be entering a new stage. When registering (or updating) a DOI record for a publication, publishers can include information about the funding of the research. The Open Funder Registry grew out of recommendations in the report from the US Scholarly Publishing Roundtable in 2010. During the Annual Meeting of Crossref that year, Frederick Dylla, CEO of the American Institute of Physics, argued that in order to make research funding information in publications accessible, it needed to be presented in a standard way and stored in a central location.

The benefits of having open funding metadata available, listed by Dylla in his presentation 13 years ago, are still very valid:

Researchers benefit because it increases transparency of their funding sources and supports the requirements they already have from their funders.
For funders, having this data available is essential because it allows them to identify the published outcomes of publicly funded research. Essential to monitor compliance with open access policies, but also important given the pressures funders face to account for their spending of public money.
For publishers, funding metadata provides a valuable service, as it provides insight into how the research they publish is funded.

Although Crossref has been collating funding metadata for many years, there seems to be a renewed interest in this service. Publishers have long expressed a desire to solve the challenges, meta-researchers need this information in order to analyze research on research, editors are concerned with research integrity, including funding trends, and funders themselves need to track the reach and return of their support.

Open Science seems to be an important driver: As we move to an ecosystem built on Open Science principles, not only publications, data, and software need to be openly available, but also the metadata associated with those scholarly outputs. Indeed, in an Open Science world, all meta information should be open, and academia should not be dependent anymore on data from proprietary bibliographic databases. Indicators for research assessment and policy development should be open indicators, derived from open metadata. Much has been done in this area already, in the context of Open Citations and Open Abstracts. While many in the community have focused on the bigger picture of advocating for all open metadata, e.g. Metadata 20/20, open funding metadata is arguably the next big thing. Open Research Information, including open metadata, must be a strategic priority for science and society.

Room for improvement

After ten years of collecting funding metadata, 25% of records in Crossref contain some kind of funding information, and this figure was reached by a steady growth over that time. A number of recent studies have shown, however, that there is room for improvement. A case study published by two of the present authors has shown that the extent to which publishers deposit funding information to Crossref varies considerably. Some larger society presses - American Chemical Society (ACS), American Physical Society (APS), and Royal Society of Chemistry (RSC) - perform exceptionally well, with almost 100% of publications containing funding information. But there is still a large number of publishers - among them large legacy publishers - that attain substantially lower figures or do not seem to deposit funding metadata at all. Our case study has shown that often this cannot be explained by the fact that authors have not provided any funding information, as often this information is available in the acknowledgement sections of the papers. Somehow, however, this data does not find its way to Crossref.

Workflows and challenges: collect, retain, validate, deposit

In order to chart the challenges that publishers face when collecting this information, we organized a roundtable session. 26 organisations were invited from across the ecosystem. These included: major publishers (American Chemical Society, British Medical Journal, Elsevier, IOP Publishing, PLOS, Royal Society of Chemistry, Sage, Springer Nature, Taylor & Francis, and Wiley), funders (European Research Council, Austrian Research Council, Dutch Research Council, OSTI-DOE, UKRI, and Michael J Fox Foundation) as well as service providers (Aries Editorial Manager, PKP / OJS, Scholastica, and eJournal Press).

In order to map the potential barriers and challenges publishers face, participants were presented with a workflow scheme representing a hypothetical production process.

This workflow outlined the steps in the production process at which funder information would potentially be handled, as well as some of the considerations that might be at play at each step.

collecting funder information (upon submission or acceptance)
extracting funder information from full text
retaining funder information through the production workflow
including funder information in article metadata
making metadata and/or full text available for indexing

Participants were invited to comment on this workflow and place digital dots in the scheme to identify challenges in the collection, retention, and deposit of funding information. These pain points were afterwards fleshed out in break-out groups.

Lessons learned

1. Still a lack of awareness among editors and authors

For many journals and publishers, collecting funding information starts when papers are submitted through submission systems. Many publishers use the same systems: ScholarOne and Editorial Manager, though many have multiple systems in place for different portfolios of journals. Around 25,000 journals use PKP’s Open Journal System, and Scholastica and eJournal Press are growing in popularity and importance. All of them provide the possibility for authors to enter funder information but this does not by all means mean that all journals make use of it. Submission systems are highly customizable, and publishers tend to tailor systems to the needs and wishes of their journals. Editors who do not see much value in collecting funding metadata therefore present a first ‘weak link’. Publishers and tech providers agreed that more outreach is needed about the importance of funding metadata among editors and authors.

2. Improvements are needed in submission systems

Where journals and publishers agree on asking authors to register funding information through the submission systems, many express a tension between collecting structured metadata and making it as easy as possible for authors. Many are hesitant to use mandatory input fields. Instead, funding metadata is often collected as free text, giving rise to a plethora of ambiguities. Most systems provide suggestions based on the input of the author based on the Open Funder Registry. A lot seems to go wrong at this stage. Authors often persist in the wrong spelling of their funder and do not choose predefined suggestions, making it very difficult to match input to Funder IDs. Publishers estimated the number of non-matches up to 50%. Trivial issues like “Bill & Melinda” versus “Bill and Melinda” or “Netherlands organisation” versus “Netherlands Organisation” result in errors. Here, autocomplete techniques seem to be in dire need of improvement. Based on a preliminary analysis of funder name variants used in Crossref, adding up to 3 of the most frequently used name variants to the list of ‘alternative funder names’ in the Funder registry could solve around 60% of missed matches.

3. A lot can be learned from how some publishers have changed and organized their workflows

Faced with these issues, the Royal Society of Chemistry has invested in innovative workflows to enhance the availability of funding metadata. Instead of relying solely on the free text input of the authors, RSC presented to the group the details of how they have tackled the issue. In addition to author-provided acknowledgements, they work with third-party production vendors to programmatically extract information from the acknowledgement section of papers. Data from the two sources are compared, and when differences or conflicts are being noted, the data is fixed, completed, and reformatted. The next step is crucial - the newly-cleansed funding data is fed back to the author for validation, and retained during the production phase of the paper. Implementation of this validation stage has increased the availability of funding metadata by 30%. In 2023 80% of papers published by RSC have some kind of structured funding metadata. An additional benefit of this feedback loop was its educational effect by alerting authors to the importance of correct funding information. But even RSC continues to struggle with issues of funder name ambiguity, use of acronyms, authors reporting grant or award names instead of funder names, issues with phraseology of funding acknowledgements, and frustrations with the user experience of the service provider integrations with the OFR.

Many publishers agreed that collecting funding information from full-text papers is the preferred option. Not only because it lowers the burden for authors, but also because this potentially renders better data as this is where authors are expected to include this information as part of their funder’s commitments.

4. Retaining information and submitting: no big deal

At the beginning of the workshop, it was expected that maybe the retention of funding information and the propagation through various interlinked systems might pose problems for publishers. However, this was not identified as a problem by participants. Nor was there mention of any challenges in depositing information to Crossref, nor of downstream databases having difficulties retrieving the metadata.

5. There is a genuine interest across the ecosystem to improve funding information in Crossref

While many concluded that there was still a long way to go to solve the many technical challenges related to funding metadata, attendees were unanimous on its importance. Participants agreed that these improvements would require investments from publishers. A willingness to do those was expressed, but also a sense that publishers who do should be incentivised for it, maybe as part of the agreements they have with library consortia. JISC’s recent contract with Taylor & Francis (page 164, Section 7a (iii)) is a good example of how consortia can successfully negotiate the supply of high quality metadata, including funding metadata. It was agreed that another solution could be to allow the additional deposit of the free-text acknowledgement section as a metadata field in Crossref. Instead of educating authors to enter their data correctly or relying on publishers and tech providers to improve their systems to turn free text funder acknowledgement text to structured data, text mining and machine learning could facilitate the improvement of this data.

Next steps

For this workshop, we concentrated on the collection and registration of funding metadata by publishers and did not go into the important, related, issue of the Crossref Grant Linking System (Grant IDs) nor of the plans to further align funder IDs with ROR IDs, both projects that help the community to better record funding information.

Next steps resulting from this community workshop, as

Funders are encouraged to join and register their grants with Crossref DOIs so that registered grants can in future be linked directly to publications and other outputs. About 50 funders have already created around 90,000 grant records. The more grant DOIs that are created by funders, the more likely publishers will be able to prioritize collecting them in their own publication metadata.
Publishers are encouraged to work with their service providers to prioritize the quality of the open funding metadata through Crossref, which is a source for downstream analyses and inclusion by many thousands of tools and services.
Other stakeholders are also offering opportunities to focus on funding metadata, showing a growing interest in the completeness of funder metadata. For example, OA Switchboard’s funder pilot, which also looks at the potential to feed enriched metadata back to Crossref to make them publicly available, and the Open Research Funder Group’s work to promote the improvement of tracking research output, including funding metadata, which includes an active working group in this area.
Crossref will continue to work with publishers and service providers to encourage and make it easier to include funder information in article metadata, including the use of grant identifiers and funder identifiers. Work is underway to bring the Open Funder Registry closer to ROR (Research Organization Registry), and is planning, at some point in the future, to merge the OFR into ROR, as ROR has a much wider scope and is more broadly community-governed. Crossref has also begun some work on collecting ROR IDs where we currently collect Funder IDs. More technical information is available in this ticket).

We would like to thank all the participants of the workshop for their openness and commitment to working through these issues together. It was a rare opportunity to share insights from publishers, service providers, funders, and researchers - and a useful first step in co-creating a shared understanding of the challenges and charting a path forward.

A healthy infrastructure needs healthy funding data

Kirsty Meddings — Wed, 16 Dec 2015 00:00:00 +0000

We’ve been talking a lot about infrastructure here at Crossref, and how the metadata we gather and organize is the foundation for so many services - those we provide directly - and those services that use our APIs to access that metadata, such as Kudos and CHORUS, which in turn provide the wider world of researchers, administrators, and funders with tailored information and tools.

The initiative formerly known as FundRef

Together Crossref’s funding data (previously known as FundRef – we simplified the name) and the Open Funder Registry, our taxonomy of grant-giving organisations, comprise a hub for gathering and querying metadata related to the questions:

“Who funded this research?” and “Where has the research we funded been published?”

To support the funding data initiative, three key pieces of metadata are needed from publishers:

Funder ID
Funder Name
DOI

Unfortunately only around half of the 950,000 Crossref DOIs with funding data contain funder IDs, the unique funder identifiers from the Open Funder Registry that are needed to link up all of the data. So, only half of the data is useful. (And 950,000 DOIs is only a fraction of the 77 million DOIs in our database, but more on that later).

When we looked at the funding data that was coming in without funder IDs we were a little surprised. We had expected that most of these would be names that simply aren’t in the Open Funder Registry yet, and we thought there would be a certain amount of incorrect information that had been entered into the “funder_name” field.

Instead, what we found was that many of the names were correct, and the funder IDs were just missing.

Tidying the data

To help correct this, we decided to match incoming names to funder IDs where we could do so with the highest level of confidence. After much testing to minimize false positives, we switched this on at the end of August 2015.

Throughout September and October, we inserted funder IDs for about 25% of the names that have been deposited without IDs. For October, the real numbers were 68,000 funder names with no IDs deposited, and 18,000 funder IDs inserted by Crossref.

In the same period 42,000 funder IDs were deposited by publishers. With our matching on top of this, we are achieving a little over a 50% overall success rate of “good” funding data (funder names and funder IDs together).

We have been very careful to distinguish the funder IDs that we have added from those deposited by publishers - provenance of data is an extremely important part of what we do. All funder IDs are tagged as provided either by the publisher or Crossref. Every time we insert an ID into a deposit, the publisher is notified in the deposit report.

We have also now added these tags to our REST API so that publishers can query to find out exactly which DOIs we have amended*. The ideal scenario at this point is that the publisher checks that they are happy with the matching and then redeposits the funding data for those DOIs, over-writing the <span >doi-asserted-by: “crossref”</span> tag and claiming the metadata as their own.

Setting some limits

The second largest problem with funding data was incorrectly entered funder name – e.g. concatenation of several names or authors entering overly long or vague program names instead of the official funder name.

To help weed this out, we have made a couple of changes to the funding data deposit system:

Funder_name field can no longer contain a numerical string over 4 digits
Funder_name field can no longer contain a text string over 200 characters
Funder names that that do not adhere to these two rules will now cause the funding data section of the metadata deposit (not the whole deposit) to fail and return an error message.

Getting the growth we need

As of today, 198 publishers deposit funding data with Crossref. This amounts to about 3.5% of Crossref’s membership (although it’s a larger proportion of our total deposits). We need more publishers to deposit funding data so that funding data search can become a truly useful tool for the community. There’s no sign-up process or additional fee - read about how to get started, and take a look at our best practices for depositing funding data.

Finally, we ask you: how can we get more and better funder metadata in 2016?

This is not a rhetorical question. Please tweet your thoughts @CrossrefOrg or email your replies to info@crossref.org. You will receive something special via snail mail if you reply to us – just Crossref’s way of saying thank you.

^{*At the time of posting our database is re-indexing and the “asserted-by” tags are still filtering through to the API. Check back in a day or two for the full picture.}

Best Practices for Depositing Funding Data

Kirsty Meddings — Tue, 01 Sep 2015 00:00:00 +0000

Crossref’s funding data initiative (FundRef) encourages publishers to deposit information about the funding sources of authors’ research as acknowledged in their papers. The funding data comprises funder name and identifier, and grant number or numbers. Funding data can be deposited on its own or with the rest of the metadata for an item of content.

There are two ways that publishers can collect this funding information for any given piece of content: by asking authors to input the funder name(s) and award number(s) via their submission system, or extracting the funder names and award numbers from the acknowledgements in the paper.

The funding data is only useful if it is standardised, and so it is absolutely critical that funder names are deposited with their associated funder IDs from the Funder Registry.

For publishers considering or about to start collecting and depositing funding data, and for those already doing so, we have drawn up some guidelines that will help you to ensure good quality metadata.

If you are collecting funding information from authors via your submission system:

Provide very clear instructions for your authors. Your submission system should prompt the author towards the canonical name from Crossref’s Funder Registry as they type, or guide them through a pick-list. Make it clear to authors that they should choose funder names from this list and not copy and paste from their manuscript.
Work with your submission system vendor or adapt your in-house system to make it easy for authors to select from the Funder Registry, and more difficult to paste incorrect names or ignore the suggested names. Consider a warning message if an unknown name is entered, and offer a list of close matches.
Instruct authors to look for the name of the funding body rather than a specific program or project.

If you or one of your vendors is extracting funding information from papers:

Provide the same clear instructions to your vendor(s). Stress the importance of matching the funder names in the acknowledgements to the names in the Funder Registry.
Look for common text-extraction errors such as concatenated funder names, punctuation errors, and stop words such as “of/for” that are commonly used interchangeably, or the presence or absence of “the” at the start of a funder name.

For both workflows:

Add QA into your workflow. Many of the names sent to Crossref without IDs are very obviously funders that are in the Registry, and a check by editorial or production staff could correct misspellings or fill in blanks. Check that grant numbers have been separated and are not being deposited as one long string.
Be aware that funder names deposited without IDs are not valid funding data and will be hidden from Crossref’s search tools and APIs until such time as they are updated with a funder ID.
The funding data section of a deposit (but not the rest of the deposit) will be rejected by the Crossref deposit system if
- The funder_name field contains a numerical string longer than 4 digits
- The funder_id field contains a number that is not an ID from the Funder Registry
- The funder_name contains text that exceeds 200 characters
Consider only depositing data that has funder IDs and holding the rest to re-poll against the Funder Registry at a later date when more funder names have been added. The Funder Registry is updated at approximately two-monthly intervals. You can sign up to be alerted to updates here.
If there are funders that appear regularly in your particular subject or geographical area that are not in the Registry, send a list to funder.registry@crossref.org.

♫ Researchers just wanna have funds ♫

Geoffrey Bilder — Thu, 10 Apr 2014 00:00:00 +0000

photo credit

Summary

You can use a new Crossref API to query all sorts of interesting things about who funded the research behind the content Crossref members publish.

Background

Back in May 2013 we launched Crossref’s FundRef service. It can be summarized like this:

Crossref keeps and manages a canonical list of Funder Names (ephemeral) and associated identifiers (persistent).
We encourage our members (or anybody, really- the list is available under A CC-Zero license waiver) to use this list for collecting information on who funded the research behind the content that our members publish.
We then ask that our members deposit this data in their normal Crossref metadata deposits.

And that was cool.

But then people started asking us awkward questions. Questions like “what can I do with the funder data?” and “how do I query it?”.

Stoopit people.

Can’t you just let us bask for a few minutes in the sunny glow of actually conceiving of and launching a project within a year?

But seriously, funders, were interested to see how they could use the funder metadata being collected in Crossref. In particular, some funding agencies were interested in being able to measure Key Performance Indicators (“KPIs” to management wonks) related to recent mandates such as the February 22nd 2013 OSTP memo, Public Access to the Results of Federally Funded Research. Two groups also approached us, CHORUS and SHARE. Both are interested in exploring how to build reporting tools for funders, institutions and researchers and each brought us a gigantic hairball of use-cases they were hoping we would be able to meet.

Conveniently, we were in the process of creating a revised, modern Crossref API that is entirely buzzword-compliant, and so we set to work…

We thought people might be interested in seeing what you can do with the Crossref REST API in relation to funding information and the expectations that are increasingly being attached to them. CHORUS is already using the Crossref REST API heavily and we expect that SHARE will soon start making use of it as well. The feedback from both groups has been very useful, but we are looking for broader feedback as well. The API is still in development, so now is your chance to help us shape it.

Brief Examples

Please note, the following are APIs calls, although you can copy and paste the URIs into your browser, the data is returned in a machine readable representation called JSON. If you want the results to look a little more presentable, we advise you install the JSONView plugin:

Firefox Users: JSONView
Chrome Users: JSONView

Also note that publishers have only just started to deposit the metadata needed for these APIs to work, so the data is currently sparse. We know that many of our members are working feverishly to populate more of the needed metadata, but this requires updates to the their manuscript tracking systems, production systems and hosting systems. It takes time.

But for now you can paste the relevant URIs below into your browser and see the results that we do have. Expect these numbers to increase sharply over the next few months

To start with, you might want to know how many articles in Crossref have FundRef metadata:

https://api-crossref-org.pluma.sjfc.edu/v1/works?filter=has-funder:true&rows=0

You could then be interested in knowing how many works in Crossref use FundRef to credit the United States’ National Science Foundation (NSF) for funding their research? First you need to find out what the FundRef identifier is for the NSF:

https://api-crossref-org.pluma.sjfc.edu/v1/funders?query=NSF

You can see that there are several entries that match “NSF”, and that the one we are looking for has the identifier http://dx.doi.org.pluma.sjfc.edu/10.13039/100000001. Remember, funding agency names can change frequently, the ID provides a persistent link to the funder even if their name changes.

If you are curious, you can see the details for the NSF entry, including its location, parent and child organisations:

https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001

Notice that the results also lists the work-count. This is the number of works in the Crossref metadata that list the US NSF as having funded the research.

So perhaps you would like to see the list of works. The following will list the first twenty:

https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001/works

You can page through the results with the offset argument:

https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001/works?offset=20
https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001/works?offset=40
...

How many works that have listed the NSF as a funder have license information:

https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001/works?filter=has-license:true&rows=0

Lets see the first batch that have license information:

https://api-crossref-org.pluma.sjfc.edu/v1/funders/10.13039/100000001/works?filter=has-license:true

Lets look at the metadata for one of the DOIs returned:

https://api-crossref-org.pluma.sjfc.edu/v1/works/10.1063/1.3593378

Interesting, the metadata shows an article published by AIP. It includes license information (CC-BY 3.0) as well as a link to the full text. If you follow the link to the full text, you can retrieve it:

http://link.aip.org.pluma.sjfc.edu/link/applab/v98/i21/p216101/pdf/CHORUS

Wow- A pretty short article. But you can see that it does credit the NSF and that the award number recorded in the text is the same as the award number recorded in the FundRef section of the Crossref metadata. Yay.

You can see in the brief examples above that there is a lot of other metadata you may want to query on and explore. It can include ORCIDS, information about archiving arrangements- even abstracts. It all depends on what the Crossref member has decided to provide.

You can get a simple overview of what a Crossref member has provided by looking at a member summary. Here is an example for Hindawi:

https://api-crossref-org.pluma.sjfc.edu/v1/members?query=hindawi

Note again that names are fickle, so the above query can also be accomplished using the member identifier like this:

https://api-crossref-org.pluma.sjfc.edu/v1/members/98

Groovy init?

If you want more pointers on where you can learn how to use the API, read on…

The Future of the Crossref REST API.

Our aim is for the Crossref REST API to go into production this Summer (2014). As with most of our newer APIs, there will be a free API for public use and a paid for API for professional use. The only difference between the two will be that the professional version will come with a service level agreement (SLA) covering uptime, response time and support. Naturally, this also means that the professional one will be on dedicated hosting equipment so that we can meet these SLAs, whereas the performance of the free version will be subject to the vicissitudes inherent in using a shared, constrained resource (i.e. the server and network it is running on).

Again, the basics of the API are in place. It should be fairly stable, but we do reserve the right to make changes to it over the next few months. Please send us feedback.

— The Weasel