Conference Identifiers on Crossref

PIDs for conferences - your comments are welcome!

Aliaksandr Birukou — Thu, 19 Apr 2018 00:00:00 +0000

Aliaksandr Birukou is the Executive Editor for Computer Science at Springer Nature and is chair of the Group that has been working to establish a persistent identifier system and registry for scholarly conferences. Here Alex provides some background to the work and asks for input from the community:

Roughly one year ago, Crossref and DataCite started a working group on conference and project identifiers. With this blog post, we would like to share the specification of conference metadata and Crossmark for proceedings and are inviting the broader community to comment.

Why are conferences important?

One common misbelief is that most published research appears in journals. However, next to new ways of communication research results (blogs, presentations,…) and journals there are also other publication options, like books, very important in humanities, or conference proceedings, which are very important in computer science and a couple of related disciplines. Conference proceedings are collections of journal-like papers, often undergoing a more competitive peer review process than in journals. For instance, looking at original research in computer science in Scopus published in CS in 2012-2016, 63% of articles appeared in proceedings, while only 37% were published in journals. DBLP, one of the most important indexing services in CS, lists more than two million conference papers organized in ~5,400 conference series.

So, while it is true that CS has a significant share of conference proceedings, conferences are also relevant in many other disciplines which do not publish formal proceedings. For instance, inSPIRE contains ~23,000 conferences in high-energy physics, the American Society of Mechanical Engineers (ASME) publishes roughly 100 proceedings volumes annually.

Why do we need an open persistent ID for a conference or a conference series?

With publishers, learned societies, indexing services, libraries, conference management systems, research evaluation and funding agencies using conferences directly or indirectly in their daily work, a common vocabulary would simplify data processing, reporting and minimize errors. Right now, a publisher assigns a unique conference ID to the conference to be published, then an indexing service does it, then it is assigned in a library. Wouldn’t it be easier to do this at the very beginning of the process, when the conference planning starts, and keep the same identifier through the whole conference lifecycle?

The joint Crossref and DataCite group on conference and project identifiers has discussed this topic at half a dozen calls and various PID community meetings (PIDapalooza, FORCE conferences, AAHEP Information Provider Summit). The result of those discussions is a draft of the specification of conference metadata and Crossmark for proceedings.

The document first defines the concepts of a conference, conference series, joint and co-located conferences. It then introduces the information we want to store about those entities, e.g., the ID, name, acronym, other IDs, URL and the maintainer of the conference series, or the ID, conf series ID, number, dates, location, and URL for conferences. Such metadata can be submitted to Crossref and DataCite by conference organizers or publishers on their behalf and linked to the existing proceedings metadata, where appropriate. It can be then used for linking research outputs from a conference (beyond formal proceedings), recognizing reviewers via services such as ORCID and Publons, computing metrics of a conference series, conference disambiguation in indexing services and ratings (CORE, QUALIS, CCF), and so on.

The second part of the document introduces Crossmark for conference proceedings. Its goal is to structure and preserve the information about the peer review process of a conference as declared by the general or program chairs. Depending on how much information is available from the conference organizers, one can use the basic or extended versions of Crossmark.

In order to comment, please open the specification and leave comments using “comment” feature of Google Docs. The draft remains open for comments till the 31st of May 2018.

Next steps

After hearing from YOU, we will update the document to reflect the community comments. In parallel, we start a subgroup discussing the governance models, looking into whether we need a new membership category at Crossref, what fees should be covered, etc.

Taking the "con" out of conferences

Geoffrey Bilder — Wed, 15 Feb 2017 00:00:00 +0000

TL;DR

Crossref and DataCite are forming a working group to explore conference identifiers and project identifiers. If you are interested in joining this working group and in doing some actual work for it, please contact us at community@crossref.org and include the text conference identifiers WG in the subject heading.

All the times I could have gone to Walt Disney World…

Back around 2010 I added a filter to my email settings that automatically flagged and binned any email that contained the word “Orlando.” Back then this was a remarkably effective way of detecting and ignoring spam from the numerous fake technology conferences that all seemed to advertise the city of Orlando, Florida as the location for their non-events. I suspected they all chose Orlando as it would provide the punter that little bit of extra motivation to pay and register for the conference as they simultaneously plotted how they could tag-on some holiday time at Walt Disney World. I finally had to remove the filter last year when I realised that the scammers had moved on to advertising more realistically gritty cities in their calls for submissions and that meanwhile I had managed to miss all the mail informing me of the ALA’s summer 2016 meeting held in, you guessed it… Orlando.

Clearly we need better mechanisms to flag dubious conferences.

Late last year Crossref’s Strategic initiatives group was approached by “CounterMock,” a group of Crossref members (including major proceedings publishers like Springer Nature, Elsevier, IEEE, ACM, IET, etc) who were actively exploring the establishment of an identifier system and registry for scholarly conferences.

The long term goal of the group is to make it easier for publishers, researchers and other stakeholders to identify fraudulent and/or low-quality conferences. There has recently been a proliferation of conferences that seem to have been developed specifically to dupe international and early-career researchers into paying substantial conference and publication fees. Sometimes these conferences are intentionally named after long-standing and well-respected conferences. At worst these conferences are entirely fake - no meetings are held and no publications are issued. At best they produce subpar publications of questionable academic integrity. Members of the group are concerned that these “mock conferences” (Hence “COUNTERMOCK”) will:

Waste researcher time.
Waste publisher time.
Undermine academic trust in conferences and conference proceedings as a trustworthy means of scholarly communication.

The group understands that the “evaluation of a conference quality” and the “unambiguous identification of conferences” are separate concerns (as they are with publications, contributors, etc). But they also realise that it will be hard to address the quality issue without an infrastructure for unambiguously identifying conferences and providing meaningful provenance metadata about those conferences. Moreover, having unique identifiers for conference series would enable a number of other applications. Examples include conference-level metrics, better and more structured info about forthcoming conferences on a certain topic, and more visibility of conferences in research evaluation.

Springer Nature has built a POC prototype of a conference identifier system and shown it to a number of other parties. The feedback has been that there is interest in the project, but that the consensus is that it should be managed a run by a neutral industry group. They have approached us to form a working group and explore how this project can be advanced.

This is all good. Crossref itself doesn’t make value judgements on the quality of content registered with us. Crossref DOIs are not quality marks. But we do believe that unambiguous identification of research artifacts is a perquisite to building effective trust and reputation tools.

It is possible that the issue of conference identifiers can be folded into the work we are doing with DataCite and ORCID on organisation identifiers. For example, some have argued that organisation identifiers should include identifiers for projects or other less formal and more ephemeral corporate entities that are often included in affiliation and/or bibliographic data. It is possible to make the similar arguments in the case of conferences.

On the other hand we have also been interested in the issue of “project identifiers.” Martin Fenner and Tom Demeranville have made a strong argument that ‘projects’ can be thought of as containers for collections of project outputs, project members and project funders. Again, it seems plausible that one could make the same case for conferences.

At the very least it is important to coordinate any work that is done on conference, project and organisation identifiers. This why we have decided to form a joint Crossref/DataCite working group to specifically explore conference and project identifiers and determine how they relate both to each other and to our already ongoing work with ORCID on organisation identifiers.

Additionally, it is likely that the working group will discuss and explore how conference/project identifiers might be used for increasing the transparency of peer review at conferences, better attribution for programme chairs and program committee members, and how they might be incorporated into other services like Crossref Metadata Search, DataCite search, CrossMark, etc.

If you are interested in doing some work on this- then please indicate your interest in joining a working group by sending email to community@crossref.org and include the text conference identifiers WG in the subject heading.

We will update this blog as the group convenes and makes progress.