DOIs enable persistent linking to a piece of content, and an easy way for a publisher to update this if/when the content moves. Enabling others to persistently link to a piece of scholarly content, including in reference lists for example, makes the content easier to find, cite, link, assess, and reuse.
In this webinar we cover:
- What DOIs are and how they work in combination with other metadata you can register at Crossref.
- What DOIs are not; we will debunk some myths and misconceptions about what a DOI can do.
- Who uses persistent identifiers and scholarly metadata, many more than you might expect!
Presented in English by Vanessa Fairhurst, Crossref Outreach Manager, in collaboration with ies Research. 19th August 2020.
3. Crossref makes research outputs easy to find, cite,
link, assess, and reuse.
We’re a not-for-profit membership organization
that exists to make scholarly communications
better.
Mission
4. What can I register a DOI for?
Journals
Books
Book chapters
Conference proceedings
Datasets
Dissertations
Reports
Standards
Posted content (preprints)
Peer reviews
… and more
5. The structure of a Crossref DOI
https://doi.org/10.1006/jmbi.1995.0238
• The DOI directory: makes the DOI actionable on the web
• Prefix: assigned by Crossref
• Suffix: assigned by the publisher
Total DOI = routes through the DOI resolver to point to the registered URL
6. DOI suffix
• consistent
• simple
• short
More details: https://www.crossref.org/education/member-setup/constructing-your-dois/
7. Crossref DOI display guidelines
• Always be displayed as a full URL link
• An example of the best practice in displaying
a Crossref DOI link is:
https://doi.org/10.1629/22161
• Old format was http://dx.doi.org/
8. Your landing page
• A full bibliographic citation so that the user can verify
they have been delivered to the correct item
• The DOI displayed as a URL, per display guidelines
• A way to access full text: access to full text is
completely controlled by the publisher but the landing
page must be accessible to everyone.
13. more metadata
reference lists, funding data, ORCIDs, license data, clinical trial numbers,
errata, retractions, updates and more through our Crossmark service,
JATS-formatted abstracts, relationships between items…
14.
15. In Summary - A DOI is...
✔ an opaque, persistent, digital identifier.
✔ used to identify a unique content item and its location online.
✔ displayed as a full URL link.
✔ used in citations to enable reference linking, aiding discovery
of content.
✔ registered alongside complete, comprehensive and up-to-
date metadata.
16. A DOI is not...
Х an identifier of who published the content.
Х a mark of quality.
Х a way to get indexed into a specific database.
Х magic (unfortunately!)
17. Get help and support
• Education curriculum:
https://crossref.org/education/
• Email support@crossref.org
• New Community Forum:
https://community.crossref.org
• AMAs:
https://www.crossref.org/webina
rs/
19. How to add Crossref DOI links to references
• Use a search engine for individual articles (slow)
• Query Crossref with XML (efficient, requires skill)
• Use Crossref lookup tools (simple) -
https://search.crossref.org/references
https://apps.crossref.org/SimpleTextQuery
• Use Metadata Manager
• Third party tools such as OJS 3.1.2
https://www.crossref.org/education/reference-linking/how-do-i-create-reference-links/
Editor's Notes
What DOIs are (identification, discovery, linking including reference linking, attached to metadata like ORCID iDs)
What they’re not (magic, a way to definitely get indexed in a specific database, a quality mark, a publisher identifier - content moves)
Who uses DOIs & metadata? (metrics tools, abstracting and indexing databases, bibliographic management tools, libraries, funders etc.)
Thank you for letting us join you virtually today.
My name is Vanessa, I’ve been at Crossref for 3 years now and my role is Community Outreach Manager. I manage our LIVE local events both online and in-person when we are able to again, get feedback and input from our community around the world, webinars, and the Crossref ambassador programme. Prior to Crossref I worked in International Development with a focus on access to scholarly information and scientific research in developing countries.
With me today is Rachael Lammey, Rachael is Head of Community Outreach, she previously worked in publishing before joining Crossref in 2012. At Crossref she has worked in Product Management, managing our Similarity Check service, before moving over to the Community Outreach team in 2016.
Today we will be talking to you about what a Crossref DOI is, what it isn’t, and the importance of registering comprehensive scholarly metadata to aid the discoverability of your content.
This is our mission at Crossref (slide)
Over 16,000 member organizations
40 staff based in USA, UK, Ireland, France and Germany.
16 member board, cross section of international publishers
Metadata store of over 100 million scholarly content items
A DOI is just the start - We offer a wide array of services to ensure that scholarly research metadata is registered, linked, and distributed.
We preserve the metadata we receive and make it available via our open APIs and Search.
Today we are talking about Crossref DOIs but there are of course other registration agencies and other types of identifiers. For example ORCID provides identifiers for people, whereas if you were looking to register research data you might assign this a Datacite DOI instead. These identifiers, as part of the open infrastructure ecosystem, enable disambiguation, and links to be made between research objects, people, organizations as well as interoperability across various scholarly services.
We currently support deposits for the following types of content:
Journals and journal articles
Books and book chapters
Conference proceedings and papers
Reports / working papers
Dissertations
Standards
Posted content (preprints)
Datasets
Components (supplemental materials)
Peer reviews
You may register other types of content that don’t fit into these categories. We’ll collect some basic metadata (usually as a dataset) We’re hoping to provide support for additional content types in the future.
So a bit about the structure of a Crossref DOI.
A DOI is composed of 3 sections. The red part is the resolver address. Each DOI is an identifier but also an actionable link, which means it is resolvable in a browser.
The blue is the prefix, which is assigned to each member when they join, in the format 10.xxxxx. In some journals you may see DOIs with prefixes that only have 4 digits. Original prefixes started with 4 digits, but have been 5 digits since 2012.
Some members have one publication, some have multiple. One prefix may be used to register all of your content, even you publish different types of content (books and journals for example)
The yellow is the suffix which is the part of the DOI assigned by the publisher and is unique to each content item.
Each member has a unique publishing schedule, which could be weekly, monthly, or even yearly. DOIs can be registered at any time.
We receive many questions from new members about creating a suffix for their DOIs. A DOI is an opaque identifier, meaning the DOI itself doesn’t necessarily have any meaning, so there isn’t a prescribed formula you need to follow.
Our best advice is that your DOI suffixes should be consistent, simple, and short. Consistent and simple for easy management- you should establish a suffix pattern that’s easy to maintain, and Short so that they don't take up too much space when used in citations (I’ll talk more about this shortly). A DOI suffix doesn’t need to state anything about the item it is identifying, that’s all done within the metadata you register with us.
When creating your suffix you may use the letters a-z; the numbers 0-9, and certain characters, such as hyphens or parentheses. Some members use use the ISSN or the volume and issue numbers, others use the title abbreviation.
New display guidelines went into effect in 2017. It’s very important that all members follow these guidelines for consistency and usability.
Crossref DOIs should be displayed in the full URL wherever the bibliographic information about the content is displayed.
https is the secure protocol, but you may see older DOIs with the older format dx.doi.org.
Members are not obligated to change the format of existing DOIs but new DOIs should have the updated version.
Once your content is registered with Crossref, users will be able to retrieve identifiers and create links using them. Crossref DOIs must resolve to either the full text, if you provide open access, or to a landing page that you maintain.
The landing page must contain the full bibliographic citation of the article, the DOI displayed as URL, and instructions on how to access the full text. This may be through a login or subscription, for example. Access to the full text is controlled by the publisher but the landing page must be available to all readers.
Here is an example of a landing page, where you can see the DOI displayed as a full text link as per our DOI guidelines, it has the bibliographic citation, and a way to access the full text.
So I have mentioned that it is good to keep your DOIs short and simple for use in citations, so I’ll talk a bit more about this.
Reference linking means hyperlinking to Crossref DOIs when you create your citation list. Reference linking is an obligation for Crossref members. If you’re a member you should be linking your reference lists using DOIs (where there are DOIs available).
This makes it possible for readers to follow a DOI link from the reference list of a published work to the location of the full-text document on a member’s publishing platform, building a network infrastructure that enhances scholarly communications on the web, because DOI links don’t break over time. Publishers used to sign individual agreements between each other to agree to link to each other’s content. This wasn’t sustainable as publishing grew, so Crossref was formed to provide a central solution.
You can see this example from PeerJ - if you hover over the link in the reference list, you can see that the link is being made by the DOI.
Reference linking is accomplished by members and their production teams, with the assistance of authors and editors who add the links to each reference in their articles. You can ask your authors to add DOIs to their reference lists or add this at copyediting stage.
(There are a number of ways to add DOIs to references including searching via a search engine, which is easy but slow, querying Crossref API with XML which is very efficient but requires some skill, we also have our own look up tools.)
So back to what I said at the beginning, Crossref is not just about DOIs! When you register your content, you send us the basic citation metadata for every item you register. This includes titles, authors, publication dates, issue numbers, ISSN, ISBN - anything that describes the content you’re registering.
We also collect non-bibliographic data about the items being registered. This can include reference lists, funding data, ORCIDs, license data, clinical trial information, abstracts, and data about relationships between items. Information on errata, retractions, updates and more can be registered through our Crossmark service.
We have minimal requirements because we need to support a variety of publication practices. We ask that you send us as much metadata as possible, and that it be accurate and clean – the more comprehensive your metadata is, the more likely your content will be discovered and disseminated and We’re always adding more metadata options.
We ask that all the metadata you deposit be complete, accurate and up-to-date and this is why - because so many people use it!
Registering your content isn’t just about getting a persistent identifier for your work. It is about where your metadata goes after you register it with Crossref, and how many other organizations then use that metadata to find the content you publish. Because Crossref’s metadata is standardized and machine readable it is very useful to many organizations that make your content more discoverable from manuscript tracking services to library discovery tools, or in metrics and analytics. The uses are vast and ever growing.
So in summary a DOI is...
And it’s worth noting some of the common misconceptions as well. A DOI is not…
An identifier of who published the content. Publishing organizations can merge, or titles can be transferred to another publisher. When this happens the prefix of the DOI does not change, this is to maintain its persistence and ensure links to the content don’t break. So whilst the prefix may align with the member who published a piece of content, this is not always the case.
A mark of quality. Often people assume that having a DOI says something about the quality of the article or the publisher, however this is not the case. We do not assess quality of any content and a DOI is simply a persistent link to the location of a content item online.
Similarly it is not a way to get indexed into any specific database. It may be good practice, or even a requirement of a specific database for your work to have a DOI, but as in my previous point, a DOI is not a mark of quality of research.
Finally a DOI is not magic unfortunately! Your Crossref DOIs need to be maintained, with the URLs updated if/when content moves, and updated with additional metadata when available. Tomorrow’s webinar will cover this in more depth where we will be talking about how to manage your metadata at Crossef, including how to make corrections and additions to your records.
Blog post: https://www.crossref.org/blog/myth-busting-in-mumbai/
Thank you! I’ve added a link on the slide to tomorrow’s webinar, same time as today, there is still space to register and Rachael will be talking more about how to manage your metadata at Crossref, including how to make corrections, updates, and how to identify gaps in your metadata records.
Ok we will now stick around for any questions.
There are a number of ways to add DOIs to references including searching via a search engine, which is easy but slow, querying Crossref API with XML which is very efficient but requires some skill, we also have our own look up tools which are linked to on the slide.