Research as Craft - Issue #5 - Ending the PDF Factory

Hey researchers! Jenny here. 👋 Recently I’ve been thinking about how we publish research to make an impact. This issue looks at the humble PDF, and whether we need an alternative.

PDFs are usually bad for accessibility, hard to read on mobile devices and harder for search engines to find. They’re best for people who like printing things. And you probably don’t own a printer.

Despite this, PDFs remain the currency of research. They offer a reassuring standard format to collect research in one place and control what it looks like. Academic article? PDF with specific formatting requirements. Think tank or nonprofit report? Snazzier PDF. Government research? More boring PDF. 

We’ve become PDF factories.

Many researchers contribute to the PDF factory by default – assuming the output of their research will be a journal paper or report, and that must be a PDF. PDFs that end up linked to from a hard-to-Google corner of a nonprofit website, or downloadable with a JSTOR login – sometimes only a handful of times.

So how do we end the PDF factory?

1. As individual researchers or research teams, we think about alternative outputs of our research – including what we share along the way

2. We call for institutions to move beyond the PDF

If we’re honest, we often don’t expect people will read our whole PDF – we expect them to skip to the chapter they’re interested in, or jump to the conclusions. Why then, don’t we support that non-linear exploration of modular content in how we present it?

In his project Screens, Research and Hypertext, Joe Miller argues the document is the wrong metaphor for research content in the digital age – instead we should think of it as a song. A song has a sequence of notes, a range of ideas and melodies, and a core refrain that’s supposed to stick in your head. But you can deliver that song in lots of ways on different instruments, experience it through live performance, recordings or written sheet music, remix it and hope others will do the same. 

As researchers explore alternative tools for thought – from digital gardening to collaborative annotation – there’s an opportunity to match that with how we externalise that thought.

While there’s potential for more video, audio or experiential content, as researchers a lot of our time is spent pouring over the written word – so it’s not a surprise it’s often the medium we want to communicate our work in. 

If writing is the right form, here are some examples of more creative not-always-linear formats, mostly from my field of technology policy:

• The Engine Room Library – a web-first way of reading the Engine Room’s research and investigations

• Branch Magazine Sustainable Interaction Design Principles – as an online magazine about creating a sustainable and just internet for all, Branch automatically changes its design based on the quantity of fossil fuels on the grid to stay inside a carbon budget at all times

• AI Justice Matrix – an online platform and collaborative authorship project on AI justice, critiquing Euro-centric knowledge processes and treating all sources and expression of knowledge as valid

• The Turing Way Handbook uses JupiterBook to make an open source, collaborative handbook for reproducible, ethical and collaborative data science

• CDEI’s AI Assurance Guide (along with some of their other projects) is an open source website hosted on their GitHub 

• Quantum Country – interactive essays on quantum computing designed to help you remember what you read

Escaping the PDF factory isn’t always in our control – grants, career progression and institutional norms (particularly in academia) often count the number of PDFs as proof of research progress. Your current referencing software probably works best with PDFs. But I’d love to see more of us pushing that where we can: if publishing on arXiv or SocArXiv, can we link to other formats? 

Can we consider what better defaults we could have? Outputs that allow referencing and attribution of knowledge, while not constraining the creation and reading experience. That are accessible to and take advantage of the full range of ways people interact with the internet.

We’d love to hear your ideas, favourite examples or challenges in escaping the PDF factory. What’s your vision for life beyond it? When don’t you want to escape? What’s stopping you? Hit reply or reach out on Twitter.

• Against the tyranny of PDFs – Wonk Watch asks why “produce and upload PDFs to their own websites” has become the default answer to what think tanks do

• A brief history & ethos of the digital garden – Maggie Appleton describes “a newly revived philosophy for publishing personal knowledge on the web”

• How can we develop transformative tools for thought? – Andy Matuschak and Michael Nielsen suggest we’ve been under-ambitious in mediums for thinking and learning on computers

• Screens, Research and Hypertext – Joe Miller’s alternative format web-first ‘book’ exploring better ways to present thought on screen

Thanks for reading Research as Craft. If you think someone else would enjoy it, we’d appreciate you forwarding it on to them or sharing it wherever you hang out on the internet.

Got an idea for an edition? Questions or feedback? Or want to nominate a thoughtful researcher in your network? Hit reply to this email or tweet at us and we’ll add them to our longlist of potential contributors.

Research as craft is co-ordinated by Jenny Brennan and Aidan Peppin. We met in 2019 when we both joined the same tech & society research institute in London, UK.* Ever since we’ve been sharing ideas, thoughts, and questions on what research is and how to do it well. 

(*Research as craft is an independent newsletter and is not affiliated with any organisation.)