Leon Furze

One of the goals of technology companies is to reduce “friction” – anything that slows down the user, gets in the way of user experience, or ultimately leads to the user putting down the product. It has obvious commercial appeal: infinite doom-scrolling on social media leads to more eyeballs on advertised products, meaning more revenue for the developers. Encountering friction in an app might lead a user to switch to a competitor, especially as technology becomes more and more efficient, fast, and intuitive.

Think about the last time you hit an issue with one of your devices. Maybe the rainbow wheel of death on a MacBook, or a glitchy Android app that froze the whole phone. Or even something less obvious, like a slow-loading webpage, or an annoying onslaught of pop-ups. Friction just gets in the way.

The term “friction” extends beyond technology and into psychology and even education. Friction might come from anything in life – not just tech breakdowns, but simple human issues. Right now, for example, I’m writing this blog post in an airport café. The chairs are made from some form of plastic that’s impervious to comfort and the basics of human anatomy. Every two minutes my thoughts are interrupted by the bing-bong of the PA. And for some reason, the café staff decided now would be the ideal time to fold three-thousand crinkly paper napkins.

To reduce the friction, I’ve jammed noise-cancelling headphones so far into my ears that I’m fairly sure I just cancelled several hundred braincells. I also downed a couple of espressos, because of course that always helps the writing process.

I could, of course, turn to Generative AI to write this article for me – the ultimate reduction in friction and especially smooth compared to my current situation. But I won’t. So, as my fingers vibrate across the keyboard and ambient music on full volume drowns out the sound of crunching paper and late arrivals, I’m left wondering about how much of the friction in the writing process is “good friction”, and why writing students might sometimes be better off choosing the path of caffeinate and plastic furniture.

Good friction

Earlier this year, I wrote an article about something I’d seen creeping into conversations about AI and writing: the use of AI writing tools to create the “first draft” of a piece. From my perspective, the “myth of the AI first draft” is that once the AI has laid down a few cursory ideas, we can pick the writing up and run with it and avoid the fear of the blank page. But it’s a myth, because as far as I’m concerned the first draft is where a lot of the magic happens. AI first drafts are an attempt to reduce friction, but I feel that getting that first something on the page is the kind of friction worth struggling thorugh.

The Myth of the AI First Draft

Since writing that article, I’ve had a lot of conversations about it with other authors, writing instructors, and students. A lot of people agree that the “fear of the blank page” can be overcome with a little perseverance, and can even be a positive force or a motivation to just get ideas down on paper. The “shitty first draft“, for many of us, is an important part of the process even if it is a point of friction.

When I spoke to Bonni Stachowiak on the Teaching in Higher Ed podcast in June, we talked a lot about those ideas.

But I also learned that not everyone feels the same way I do (shocking, I know.) For many people, the shitty first draft isn’t at all the point of writing, and they’re more than happy to reduce that particular kind of friction. I heard from lots of writers who find much more “productive struggle” in other stages of the writing process, such as editing.

Personally, I hate the editing process, and avoid it wherever possible. Since I often “write” my articles verbally, and then use transcription tools, there’s some unavoidable editing to catch transcription errors, fill in missing details, or tidy up the formatting. But honestly, if I could remove all of that friction at the end, I would. I was that kid in school who wrote the essay/story/whatever as quickly as possible, handed it in, and never wanted to look at it again. When I write books, I rely far too much on my proof-readers and editors.

And this all highlights an important aspect of the conversation around good and bad friction: it’s subjective.

What do others think about friction?

This post is my contribution to a much larger conversation about friction that seems to have progressed in the past few months around the implications of artificial intelligence and writing. I’ve been collecting as many of those viewpoints as possible, and will share them here. If you’re an author or educator with something to contribute, then by all means get in touch with the contact form at the end or leave a comment wherever you found this article.

First up, John Warner, who also spoke to Bonni Stachowiak on the TIHE podcast. While Warner doesn’t talk explicitly about friction, he does make the statement that writing is thinking, and suggests that part of learning to write involves exploring why certain writing tasks feel uncomfortable and learning to negotiate that discomfort. Listen to his thoughts here:

Teaching Writing in an Age of AI

Just as John Warner argues that writing is thinking, Marc Watkins suggests on his substack that learning is friction. He makes the point that education stands at odds with technology in the pursuit of frictionless experiences, and argues that in order to fully appreciate what we learn, we need to slow down and examine it closely.

Sarah Rose Cavanagh agrees, and also points out that – much like tech developers reducing friction for their customers – it’s often the teachers who are responsible for trying to reduce friction on behalf of our students. There’s a compulsion to reduce friction so that it feels like the students have learned: the easier they find the route from point A to point B, the more successful our instruction has been.

In Australia, we have the “TEEL” paragraph as a perfect example. Providing students with the ubiquitous Topic Evidence Explanation Link paragraph structure reduces friction and “helps” them to write paragraphs. It’s easy to assess, easy to judge progress, and often used as a means to give students who “don’t know what to write” a starting point. But I’ve argued long and hard that these kinds of frameworks rob students of the necessarily difficult parts of the writing process. If a student only ever sees a paragraph through the lens of a structure like TEEL, have they really learned how to write?

Katie Conrad has similar thoughts, going as far as to say that friction in education is a feature – not something which should be smoothed out. For me, that’s the productive struggle of the first draft. For others it’s the thoughtful editing process that makes the ideas shine. In any case, it’s not a “bug” that needs to be fixed.

On LinkedIn, where many of us seem to spend far too much time, Jason Gulya wrote about the need to resist “the button”: the growing prevalence of AI features which can summarise, explain, and simplify content at the click of a button. While Jason was talking about reading, not writing, I think it’s a useful point. Applications like Microsoft Copilot are already integrated into common writing tools like Word, and the “button” to create drafts, rewrite paragraphs, and suggest ideas is now floating nonchalantly alongside every line of text, just waiting to be clicked.

We need to find ways to encourage students to “resist the button”, and ways to avoid hitting that button ourselves. Whilst the inclination might be to rush in and “help” students – whether it’s through writing templates or technologies like Generative AI – it might be better to step back at times and allow our students to work through the friction on their own terms. Jane Rosenzweig thinks that sometimes, “friction is the point”.

And finally, when all is said and done, much of the writing we teach in schools and universities has to be assessed. To finish off this fly-through of friction, I’d like to highlight Emily Pitts Donahoe’s thoughts on friction in writing assessment. Because friction causes discomfort, particularly when there’s a grade or evaluation attached to it. But we don’t do our students any favours when we try to remove all of that friction and value the final grade over the process of actually getting there.

Introducing some good friction

Good friction can happen at any point in the writing cycle, from the initial ideas and drafting through to the final editing. Here are a few places where I feel we can encourage productive struggle:

1. Ideas and brainstorming: Pushing students to generate multiple ideas or approaches before settling on one can create valuable friction: don’t always settle for “good enough”. This might involve techniques like freewriting, mind mapping, or group brainstorming sessions – all of which can obviously be done with or without AI.
2. Research: Encouraging students to seek out diverse or conflicting sources can create productive tension as they grapple with different perspectives. For me, some of the best ideas happen in the reading before the writing – the writing is just a synthesis of everything I’ve read.
3. Outlining and structuring: Having students experiment with different organisational structures for their writing can create friction that leads to deeper understanding of their topic and more effective communication. Many students and writers have a “default” style that feels comfortable, but it can be worth pushing outside of those comfort zones.
4. Drafting: Encourage students to write “shitty first drafts” without worrying about perfection. It might create friction for those who desire perfection, but often leads to more authentic and creative writing.
5. Peer review: Structured peer feedback sessions can create productive friction as students learn to give and receive constructive criticism. It’s also a useful alternative to relying on teacher feedback, and you could encourage students to use generative AI as the “peer”.
6. Revision: Even though I personally dislike editing my own writing, I acknowledge that pushing students to make substantial revisions, not just surface-level edits, can create good friction and lead to significant improvements in their writing.
7. Reflection: Even after they’ve finished, having students critically reflect on their writing process and finished product can create internal friction that leads to growth and improvement in future writing tasks.

There’s no “right answer” to how much friction we need in education or writing instruction, but we shouldn’t be in a hurry to remove it all. Students can struggle through different parts of the writing cycle in their own ways, and with support from the teacher, their peers, or even AI they will come out the other side.

We need to find ways to convince students and early writers that the struggle – the friction – is worthwhile. With technologies that increasingly remove all of the barriers from first draft to final edit, we need to re-evaluate how and why we teach and assess writing in K-12 and Higher Education. And figuring that out might be the most important friction of all.

• LinkedIn
• X
• Facebook
• ShareCopied to clipboard

Want to learn more about GenAI professional development and advisory services, or just have questions or comments? Get in touch: