May 27, 2019

Creative Friction

Reflections on making a new website

A photo of three Seattle shipping cranes.

I made a new website. (You are probably on it right now.) It’s been over five years since I made my last one.01

I made this new website because my old website failed.

My Old Website Failed

To make a website interesting, one must post stuff on it. My old website failed because I did not post enough stuff on it.

Now, part of this is on me for simply not posting more stuff. The rest of it is also on me, but for creating a website in which posting new stuff is annoying and hard.

Allow me to present my thesis upfront: reduce creative friction. Iron out your tools, lower whatever imaginary quality bar you have, and just get on with it. Focus on making more stuff in that “quantity builds quality” kind of way.02

This very post is the first skirmish in this crusade.

Let’s start by talking about creative friction.

Creative Friction

I’m calling “creative friction” anything that makes you less likely to produce and share creative work.03

I think that there were largely two categories of creative friction I experienced: technical factors and psychological factors.

Technical Creative Friction

I made my old website “from scratch.” By this, I mean I wrote the HTML, CSS, and web server code without any all-in-one toolkit or framework.

However, my old website made it irritating to post stuff. Let me count the ways.

So those things were the fault of my website. But actually, there were other pieces of creative friction that were just inside my head.

Psychological Creative Friction

Interlude: Why Reduce Creative Friction?

Wooden columns under a pier.

Given all of the annoyances, fears, and worries above (phew! what a mess!), maybe it’s worth asking: should I even reduce my creative friction? Is it so important to produce creatively and post your work online, if all of the above ails you?

I do not think there’s an objective answer to whether it’s best to post your work online. I would venture there’s a personal component to this. Likely some folks will enjoy keeping their creative works private, and others don’t feel the need to produce creative work at all.

For me, personally, I find all of my reasons above pretty stupid, and am eager to try out building more of an online portfolio. I find that I get a lot of personal joy and satisfaction in creating things, and I have a hunch that sharing them online will lead me to do more of it. That alone is enough for me to want to reduce my creative friction.

Reducing Creative Friction

So how will I reduce my creative friction? What’s the deal with this new website?

Reducing Technical Creative Friction

My first directive to myself is to only reinvent the wheel once. Another way of saying this is: make your own version of established things only for educational purposes.07 In this context, this means that I shouldn’t build my website from scratch again, because it’s going to cause creative friction.

The more I learn about myself, the more I realize that this—reinventing the wheel once— is my preferred mode of operation. I like to really get into the details and build something from the ground up, at least once. It’s not so much the feeling of oil on your fingers or the satisfaction of getting a screw tight—though those are enjoyable. It’s discovering what the main problems are by undergoing the design yourself. After doing this, I feel much more comfortable using someone else’s solution because I know how the underlying pieces work. I also end up with a greater appreciation for things other people have made when I’ve fumbled my way through the same problem myself. It’s humbling to see designs that cut to the essence of a problem you were dancing around and didn’t quite see.08

But once is enough. After reaping the benefits of understanding how something is built, it’s just so much more efficient to use a professionally made and maintained foundation than building on your own monstrosity all the time. Writing this down, it seems hilarious to think otherwise. But at least for me, it took several years of programming to realize that just because I could make something of my own didn’t mean that I should. You might call this technical humility, but I think a bigger aspect is learning about the time-consuming factors that go into building and maintaining a seemingly simple piece of software: you will always want more features, compatibility is always a pain, and software stacks are always evolving.

There are a couple other pieces of design that I hope will help with reducing creative friction from the website itself. One is to worry less about the minutiae of how things are displayed. Another is to stop categorizing things (posts that I make). I sat down and made a list of all the kinds of things I might like to post on my website: little digital sketches; notes about doing research or teaching; new software projects I make. The more I thought about it, a one-level hierarchy of simply “a post is an item” would be the lowest barrier for me to make and publish things. This means it’s probably the right scheme for now.

Reducing Psychological Creative Friction

What about the other worries: producing imperfect work that’s archived forever, receiving negative criticism, accidentally copying someone, and revealing yourself online?

I hope that I can overcome many of these worries by a mindset shift from something I realized recently: I like people who have personal websites with more stuff on them.

It’s fun to learn about someone. It’s fun to read about how their mind works, see what they’ve drawn or animated, and spend a few minutes in a slice of their life they’ve shared.

This is likely news only to me. I mean, see Instagram, Twitter, and their ilk. Of course people like learning about other people! People are the most interesting things around.

But there’s something warm and slow and enjoyable about someone’s personal website in particular, as opposed to a feed on a big social networking or aggregator site. I remember finding the photo journal and dream diary of an old math professor, or the book reviews of a computer science theory professor. To give two concerete examples, I enjoyed reading Jacob Steinhardt’s blogs when he was a grad student, and perusing the drawings and personal essays by Karl Stratos. (Both are now professors—wow, time flies!) Having a larger slice into their lives than a simple list of publications immediately made me feel a greater connection to them.

This made me realize that there are more likely upsides than downsides to posting work online. Perfectionism and critical paranoia have a negligible chance of panning out usefully, and there’s no grand bookkeeper finding out better related work that you didn’t mention and docking you karma points. Most of what you write and post won’t matter, so if it helps you in any way, maybe by increasing your creative output or better connecting you to others, it’s totally worth it.

Put into a set of self-directives: make imperfect things. Don’t worry about making bad things. Don’t worry about copying someone. Don’t fret over related work.

I think that the next important part of this is momentum. I hope to figure that one out soon.


  1. To give you an idea of when I made my last website, my thought process was basically, “Wow, Markdown! Wow, Node! Wow, Boostrap! Wow, Heroku! Wow, wow, wow (etc.)” ↩︎

  2. I often think of the parable about the ceramics class and students making pottery by weight. It’s apparently from the book Art and Fear and quoted here (search for “There is a famous parable”). ↩︎

  3. I am sorry if “Creative Friction” is a term already and I’m giving it another weird wrong bad definition. For whatever bizarre reason, I find the prospect of constantly looking up whether my ideas or phrases already already completely creatively paralyzing (there’s more about this in the main text), so I’m going to close my eyes and not Google this one. ↩︎

  4. One person is Leah Finnegan, whose writing for The Outline has some really great paragraphs in it (there were a few I really loved in this essay). ↩︎

  5. I’m using the term “related work” here because it’s the phrase we use in (at least my slice of) academia to describe any prior writing from others that address similar topics. Here, I’m broadly thinking about whether someone else has written about similar themes, used similar words, coined similar phrases, or expressed the same thoughts. ↩︎

  6. Perhaps it’s worth asking: why do I find the prospect of accidentally copying someone, or accidentally expressing a poorly-made derivative thought or work, to be bad? After all, it seems that in many crafts and hobbies, a great way to start is by coping others. (I think there are a lot of quotes by famous people saying this.) Maybe it’s that a blog post implicitly claims originality? Perhaps because of the expectation that the author is well-versed in their subject matter, or the low barrier to looking up related work online? Option C: this is just a detail-oriented, perfectionist-driven fear: if you can look up related work, then you must do your damndest to find a comprehensive list of them, and point out the best ones. ↩︎

  7. It’s also great to make your own version of something if you love the act of making the thing itself. I didn’t want to clutter up the text with this, but I personally think this is a great reason to make things: simply because you love doing it! Now personally, I’m OK-ish at making websites, but I really only loved doing it when I was working on a website that my friends and I would use. For my own website, it’s more of a means to an end. Just for completeness, I can also imagine other reasons you might want to make your own tools or components from scratch: because you can make materials that are higher quality, more customized, or cheaper than if you’d bought them or used onces others have made. I’m thinking of bartenders brewing their own bitters, bassoonists whittling their own reeds, or painters gessoing their own canvas. ↩︎

  8. Technical example of an elegant solution: In my old website, I wrote server code for handing routes which retrieved the posts, filtered and sorted them, and then sent them to the user. But why do all this on every request for a static website? Jekyll, which this new site uses, solves the whole thing more elegantly: just generate your website in a build step, and serve static assets at known locations. ↩︎