2024 May 14
Over the past few years I’ve been noticing more and more of a discourse about the “small web” in my online communities. It’s cool to see! I mentioned the concept in an earlier post and thought I might write a few more words about it today.
As far as I can tell, this 2020 essay by Parimal Satyal seems to be the origin of “the small web” as a particular term. Nowadays, I see a lot of other highly similar ideas percolating around that use slightly different language: the humane web; the indie web; the rewilding of the internet. To be honest, you probably don’t even need to click these links to guess the broad arguments these essays are making. The modern web has become corporatized, user-hostile, siloed, sterile; it used to be better, and it would be nice if it became better again. Small-web advocates generally encourage people to make their own websites and to do so in a way that consciously rejects the pressures of the attention economy.
Now, if you have an ounce of self-awareness and you find yourself fixating on how things used to be better, you start getting worried. Am I just nostalgia-tripping? I’m not going to try to construct a detailed argument but I think that we can be pretty confident that there’s more going on here than some people being having fondness for how things were when they were younger. Well within my lifetime, the web has gone from being a new invention to becoming an inescapable element of everyday life for billions of people. The modern web is a genuinely different thing than it was in the “good old days”, however you choose to define them (and however “good” you feel the old days truly were).
In particular, I see two main themes in what makes the modern-day web so disappointing compared to the early years: greed and bloat.
Greed is the less-technical side of what’s been going wrong. SEO, engagement optimization, personal-information mining, scams — they’re pretty terrible, and they’re everywhere! In my view, these phenomena are fundamentally expressions of the fact that the web is now a part of society; they’re just the web-specific manifestations of unchecked capitalism, the rot economy, at work. While the idea of enshittification was originally articulated by Cory Doctorow in reference to the web, it’s clearly resonating in other realms as well: I see it pop up in conversations about topics ranging from science to aerospace, urbanism, and Manchester United.
Bloat is the more-technical side of things. Most writers in the small-web field have the feeling that websites have become slow, flaky, and overcomplicated. This is empirically true in pretty much any metric you care to analyze. While looking for a particular quantification of this I ran into the following delightful sentence:
Vite, the build tool I’m going to introduce later, only has 34.2 MBs of dependencies […]
Only!
It’s worth pointing out that in principle, greed and bloat could be separate issues. We can imagine a web that’s flooded with sites that are full of AI-generated clickbait, but at least are clean and lightweight too. In practice, however, the two do seem to go together, which is interesting in its own right. The technology sector seems unusual in that greed and laziness generally to lead to bigger and more complicated systems, rather than smaller and simpler ones. This recalls the Kroger website case study that I wrote about earlier: it took a lot of effort to design an incredibly minimal, fast website! Such a site would surely be cheaper to operate than whatever Kroger was running instead, but I have to conclude that the savings from hiring cheaper engineers dominate. (I’m sure that insightful people have pursued this thought much more deeply, somewhere out there.)
The good news that everyone in this area likes to emphasize is that the web is still fundamentally open. There’s nothing stopping me from launching a surveillance-free, labor-of-love website that anyone with an unrestricted internet connection can enjoy, and the core technologies of the web are still pretty dang efficient: the vast majority of web bloat lives in frameworks that are stacked on top of them. (Witness the Kroger example.) It’s absolutely true that an individual with the right skills and resources can create a website that’s simply lightyears beyond anything that was possible in 1998. This is not to imply that there aren’t deep inequities at play, but it’s fair to say that “write access” to the web is as equitable as it’s ever been.
The more difficult question is whether a bunch of people making indie websites is actually going to have any effect upon the median person’s experience of the web. To be honest, I’m not too optimistic. I tend to think that the analogy with “indie scenes” in other fields is likely a good one. Whether that’s encouraging or not depends on your ambitions. If your goal is to overthrow the Googles of the world, not so much. But the platform of the web does seem well-suited to supporting an indie scene where some people can actually make a living doing what they enjoy without a boss (be it a person or The Algorithm) breathing down their necks: witness the success of outfits like Pinboard, Buttondown, or Defector. That’s not nothing!
We can’t, however, take for granted that the architecture of the web will always be quite so friendly to independent operators — protocols and expectations are always evolving. The proverbial “someone” needs to apply pressure to keep the infrastructure of the web friendly to small operators. (As well as to create supporting systems: for instance, Let’s Encrypt has surely been an incredible boon for the small-web ecosystem.) My worry here is that the small-web ethos is definitely susceptible to the tendency that you can get in environmentalism and other underdog movements: hoping if enough people just display enough personal virtue, the large-scale problem will solve itself. I doubt that many people would seriously argue that there’s no role for public policy or other forceful efforts in trying to achieve these goals, but I worry the DIY approach can easily become a trap. Small-scale effort is much easier and yields rewards on much shorter timelines than large-scale action. From what I’ve seen it easily soaks up all of people’s time and energy, leaving nothing left for the big stuff.
I did title this post “The Small Web and Science” but I haven’t mentioned science at all yet. Really, I just think that it’s important for scientists to be aware of this family of concepts and some of the discourse surrounding them. Scientists are often making indie-type websites — we invented the web, after all! — and I hope that the concept of the “small web” (or “indie web”, or whichever) gives them a way to think and talk about the choices that they make when they do so. Most of the tools and materials for building websites out there are implicitly or explicitly aimed at people that want to be like the next Facebook; it is valuable, both pragmatically and as an act of solidarity, to understand that our goals are different.