The Broken Link Conundrum: Three Tips to Never Lose Your (Research) Outputs

Share this article

Latest Tweets

As consultants and practitioners working in research and knowledge exchange, we regularly review and draw on published outputs and evidence to inform our work. We identify these materials through a range of methods, with the most common being targeted word searches.

Finding and accessing outputs published in an academic context (e.g. journal articles, academic books, monographs, preprints) is generally straightforward, but things sometimes get complicated when our search results include reports, policy papers and similar types of documents (sometimes referred to as “grey literature“). These types of outputs are frequently held on organisational websites, which offer no or limited guarantee on the future availability and preservation of what has been shared.

The most common challenge when sharing via organisational websites is broken links – something we are sure that many of our readers will have experienced! In many cases, links will break because of a website/brand redesign, moved resources or a website migration, so relatively common occurrences in the life of an organisation.

In this blog, we share our three top tips to help make sure that your outputs aren’t lost and are saved in the best possible location, to maximise their future reuse and potential impact.

  1. Choose a trusted repository
  2. By using an online data repository (which, in lay terms, may be seen as a database) you can ensure permanence. At Research Consulting, we use (and are financial supporters of) Zenodo, a generalist, not-for-profit, open home for authors and organisations to deposit their work. Plenty of other options exist alongside Zenodo, and the journal Scientific Data maintains a list of trusted repositories suitable for a wide range of research outputs. For organisations like us, as well as most policymakers, we’d recommend looking at the section “Generalist repositories”, which includes databases suitable for reports, presentations, posters, data and more.

    To make things easier for our readers, Research Consulting has also set up a Zenodo “community”, where we collect any of our outputs that can be shared in the public domain and are not commercially or otherwise sensitive.

    Notably, the fact that you’re sharing something via a repository doesn’t mean you can’t also upload the same document to your website. For example, based on your preferences, the repository version could potentially be seen as a fallback solution to complement your standard web presence. As a company, however, we prefer to only refer to the repository version to avoid confusion.

  3. Create and share a persistent identifier
  4. In an academic context, articles, research data and long-form outputs are typically assigned a persistent identifier, which can be described as an “unbreakable” link that will always lead the reader to the resource they are looking for. In practice, persistent identifiers are long-lasting references to (meta)data available on the web, and help people reach exactly what they”re looking for – avoiding messages like this article”s covering image. The type of identifier that most people are likely to have heard about in the past is the ISBN, which uniquely identifies books. In an academic context, ISBNs are, indeed, common, and are complemented by digital object identifiers (DOI) to describe articles, data and other outputs as well as Open Researcher and Contributor IDs (ORCID), which uniquely describe people (here’s mine as an example).

    What many don’t know is that these unbreakable links can be created by virtually anyone. By using Zenodo, all of our outputs are given a DOI, which we include in the pdf version of our final document to ensure that readers will always be able to go back to the online original if they need to (note that any repository will offer this facility – not just Zenodo).

    Using DOIs on social media
    Figure 1 – Using DOIs on social media

    The availability of a DOI is also very useful for social media sharing, as we can be absolutely certain that those following our dissemination campaigns will be taken to the exact right item, without exception. Are you starting to think that repositories and DOIs are the real deal? We hope so, and there’s one more thing to think about!

  5. Provide good (and FAIR) documentation for your work
  6. A seminal article published in 2016 introduced the idea that “all research objects should be Findable, Accessible, Interoperable and Reusable“, or FAIR (tip: check the link for this quote – it’s a DOI). This is exactly what online repositories and DOIs help you do. By uploading outputs to a repository, you aren’t only making your work more easily findable, by making sure its link doesn’t break – you are also collecting a whole lot of goodies on the way. For example, when uploading a report to an online repository, you typically need to fill in some additional information, which in technical jargon is called metadata (literally “data about data”). This includes, for example, author(s), output type (e.g. report, poster, data, etc.), title, year, funder (if applicable) and more – all entered in a simple form that conforms to a standard metadata schema aligned with sector best practices.

    Metadata entry in Zenodo
    Figure 2 – Metadata entry in Zenodo

    Although sometimes tedious, this part of the process is key: by sharing documentation on your work, you are making it hugely more visible to humans and machines alike. For example, good metadata means that your work is richly described and can be picked up by people running web searches. At the same time, metadata can be read by computers, including via REST APIs for automated retrieval, analysis or aggregation. It’s a win-win-win – the author, the reader and a broader set of digital infrastructures all benefit from your documentation efforts.

What do I do next?

Luckily, the answer is easy. Most of the above considerations may sound technical in nature, but, as an author, you just need high-level awareness of them. What we recommend is that you pick a repository you trust as soon as possible and start sharing your work through it. One thing we haven’t mentioned so far is that most generalist repositories are completely free to use – it’s a no brainer in our view!

To help you kickstart your repository journey, we have collected a few links we have found useful over time:

Still have doubts? Check out our project showcase on Kudos and you’ll see the full power of DOIs in action!

Related Posts