The Future of the Blogosphere

Two weeks ago, I was trying to finish up a blog post when I noticed a bunch of weird characters showing up in posts. Here’s a sample from Worldly, Yet Carefree:

It turned out to be a fairly complex problem, with roots in the sheer age of this blog (2007) and crud inherited from generations of updates to WordPress, MySQL, and indirectly I suppose, PHP. Turns out, the weird characters were due to UTF8 characters being tagged as latin1 characters, and then re-encoded as UTF8. This was due to MySQL’s history of dealing badly with character sets, and WordPress making janky accommodations for that. The most recent MySQL update finally did something that made an invisible problem a visible one.

My rather expensive hosting service, WPEngine (whose database migration process caused the break, since it was apparently unaware of ancient-blog problems), basically threw up their hands and said they could do nothing after a couple of unhelpful support chats. Finally, Dorian Taylor, with some help from Artem Litvinovich (both of whom have contributed here), figured out what was going on and fixed it. If you’re interested, and able to follow the detailed technical postmortem, head over to Dorian’s post-game analysis (and strident opinions, some of which I share). There are some deeper techno-historical insights there, but the big takeaway for me at a non-technical level is that the blogosphere, like any techno-social hyperobject, is showing its age at all levels. From the lowest technical level to the highest cultural level.

Problems like this will only grow, as various pieces of infrastructure age with more or less grace. The blogosphere has a lot of history at this point, and it’s getting gradually more expensive to deal with, in terms of money, time, skill, and connections (if you don’t have friends like Dorian and Artem, it would actually be non-trivial to find people capable of fixing such problems).

The question is, given these slowly mounting problems and costs, does the blogosphere have a future?

My immediate response to the incident was to apply Cory Doctorow’s enshittification framework to the blogosphere. I think Cory identified a coherent symptomology that afflicts the web, which cases like this fit, but seriously mis-diagnosed it as being due to private platform corporations and their evil capitalist politics. The thing is, enshittification afflicts everything. Private and public spaces, centralized and decentralized spaces, platform, protocol, and commons spaces. Proprietary software and open-source software.

Every flavor of technology is vulnerable to enshittificaiton.

It’s not a political economy problem though Cory chose to spotlight one particular political economy manifestation of it. It’s a problem that’s fundamental to any sufficiently complex technology, regardless of the politics that it appears to be associated with. Call it entropy, bitrot, or whatever. Enshittification is a problem that is closer to the rusting of steel than to malice in politics.

The blogosphere is a case in point. It is built on multiple open-source software projects (WordPress, the community edition of MySQL, PHP, the Apache and nginx webservers…). The deployment model is flexible and allows both multiple hosted platform players (WPEngine, Automattic’s wordpress.com offering) to exist, alongside self-hosted installs on everything from homebrew servers to leased instances on AWS or Digital Ocean. There are powerful and public mechanisms allowing integration with all sorts of other services (email being the big one), as well as the native protocol of RSS feeds (which can be used for other things but goes with blogs like PB&J). Perhaps most importantly, there are oceans of goodwill in the blogosphere, which still inspires people to contribute in a spirit of open generosity that is notably absent in most online spaces.

Yet, despite its very different political-economic DNA, the blogosphere has become enshittified as clearly as Facebook, Google, or Amazon. Not just at the level of aging software, but at the level of the aging people who inhabit it, maintain it, and continue to churn out content on it, though at a rapidly decelerating rate. And it’s hard to blame any particular party in the picture. The technical decisions that lead to the sort of messy problem that afflicted this site can’t be attributed to malice, objectionable politics, or billionaires behaving badly. They’re within the band of ordinary technology management decisions I see all over the place in my consulting work. Humans are just not good at building complex technologies that mature to a graceful immorality. The WordPress-based blogosphere is at the outer limit of complexity we are capable of getting to. It’s just a different part of the frontier than sexier things like space missions.

Immortal, complex, graceful: pick 2 of 3. And that’s at best. At worst, you’ll have a complex system that’s dying gracelessly.

And while the enshittification of the blogosphere is not a political economy problem per se, let alone a problem attributable to the particular political economy model that Doctorow blames for a lot of enshittification elsewhere, complex technologies are inseparable from the politics and economics they induce.

I tend to the view, like many engineers, that technology is upstream of both in the long term, even if in the short term it looks like economics drives technology. The blogosphere in its heyday was a kind of a left-anarcho-capitalist political economy milieu wrapped around a robust commons. Today, even though nominally a third of the web runs on WordPress, only a tiny fraction of that installed base can be considered part of the active blogosphere. So that milieu is a fast-disappearing one.

I’ll make a stronger statement: Political economy milieus are really just upper layers of technology stacks. Despite what it sounds like, a phrase like left-anarcho-capitalism wrapped around a robust commons refers to a particular historical period and its technological affordances, not an abstract ideology. It could arise only in very specific conditions and persist only while certain other conditions lasted. In fact, it would probably be most accurate if you just called the political economy of the blogosphere blogosphereism. A thing that was born with the blogosphere and will die with it someday. We want to describe it with a set of technology-agnostic humanistic labels that we want to pretend are derived from values we plucked out of the noosphere, but that’s not how any of this works. The medium is the message, so you might as well name the message after the medium. Ironically, noosphere itself is now the name of a promising young protocol. If it succeeds, I bet the noospherism that emerges on top will be shaped more by the protocol than by Teilhard de Chardin’s philosophy.

In its aged and weakened condition, blogospherism is seeing competition from younger technology stacks with their own political layers. Substackism is, arguably, a kind of center-right enclosure ideology with a vaguely benign-monarchist, as opposed to anarchist, disposition, and a tendency to capture rather than curate the commons. This tendency extends even to the word blog (I’m amused by Substack’s attempt to claim the term for itself). This is not a criticism of Substack in particular, or even really a criticism at all. It is what it is. All Web 2.0 platforms have something like a center-right monarchist enclosure ideology going on as the top layer of their stacks. It is no coincidence that the rise of the platform economy coincided with the rise of a politics, neoreaction, sympathetic to it. It is now peaking and getting enshittified along with the technology stack it rode to prominence and power. Substackism is a kind of late-modern nephew of Facebookism, Googleism, Amazonism, and so on. You could call the aggregate platformism, which is less a coherent singular ideology than a pattern of political kinship created by shared foundational technological patterns.

Similarly, you could identify an equally old and enshittifying pattern of kinship around blogospherism. There are also wikism, forumism, chanism and so on. The whole could be called commonsism. Commonism has a broader left-to-right span on the political compass, while platformism clusters towards the center, but leans authoritarian.

Arguably, one reason Old Twitter and Reddit never quite made the big leagues, despite their influence, was that they were caught in terra nullus between the kinship zones of platformism and commonsism. Elon’s Twitter feels like a reluctant, hurried, and doomed attempt to pick a family and get in the game in the twilight years of both platformism and commonsism. It feels about as significant as an 80-year-old independent politician picking a political party (platforms are like cats, they age 7 years every human year).

My participation in Substack is largely pragmatic. I’m very sympathetic to commonsism in general and blogospherism in particular. I’m neutral to mildly hostile to platformism. In the form it actually exists, it feels mostly an acceptable sort of mediocrity-friendly space to participate in, and Substackism is a flavor I can mostly live with, especially if it means making money. The monarchism tends to be latent rather than explicit (which is not the case on Elon Twitter, which is why I’ve abandoned it), and not a constraint for most things I want to do.

But both kinship groups — platformism and commonsism — feel like they’ve aged past the point of no return. It’s the technological version of what’s happening in politics worldwide, and that’s probably no accident. WordPress versus Substack is a bit like a Joe Biden vs. Ivanka Trump type choice.

Enshittification makes aging technology foundations easy to see, in the case of platforms that are as old as the ideological kinship patterns themselves (the first families?). But young embodiments of old technology patterns, like Substack, are harder to read, and their futures harder to forecast. While the platform itself is fairly new, and shows no signs of enshittification, the technology patterns underlying it are old, and reaching their limit of expressivity. Substack Notes, a post-Elonification Twitter-like feature, felt old even the day it was launched. There were no surprising features, and more revealingly, nobody seemed to expect any.

If the medium has a message, it has already been delivered. Now it mostly exists to deliver money, which works for me.

The thing is, new things cannot really be built on old technology patterns. They must either reinvent themselves at a foundational level, pioneering fragile new patterns (which are as costly to deal with as the problems of aging patterns, but there’s more upside), or enter a kind of historical cul-de-sac. I think reinvention is unlikely in the case of platformisms. It is the nature of platformism that once it emerges as a top-level stack layer, it captures the owners, who then go down with it. It’s almost the defining feature of platformisms.

Commonsism too, is doomed in the form we are familiar with, as is blogospherism, the specific flavor I’ve grown middle-aged with and somewhat attached to. But I think there’s a better chance of foundational reinvention in the space it currently occupies. Unlike platformism, commonsism doesn’t seem to be salting the earth it is dying on. There’s regenerative potential that’s coming back into circulation.

The phrase future of the blogosphere cannot point to blogospherism or anything in the contemporary kinship group of commonsism. Any attempt to revitalize the blogosphere, or spark a renaissance, that starts with the values that emerged on the underlying aging technology stack, is doomed. Values and manifestos come last, not first. If you try to start with values, you’ll be done before you can start at all.

The weird character encoding problems are the writing on the wall here. The values of blogospherism are now as dead as they are clear and articulable.

The phrase future of the blogosphere can only point to a fundamental technical innovation around which a new technology stack can grow. Whatever ism emerges on top of that will have an inheritance relationship to the blogosphere rather than one of direct kinship. Such inheritance relationships create a line of technological descent across time. In the case of the blogosphere, I think that line extends back to the first Gutenberg-era pamphlet wars. The original printing press was closer in spirit to a self-hosted WordPress blog than to the royal enclosures from which modern platforms are descended. WordPress itself seems to sense this: the block editor introduced a few versions ago is called Gutenberg (the block editor itself is a brave but clearly very late-stage incremental innovation).

So the future of the blogosphere is something we won’t (and shouldn’t) call “blogs” but will similarly bear unmistakeable marks of its line of descent. And if you want to look around for where it’s emerging, don’t bother looking at manifestos and creative writing trends. Look for technological seeds sprouting that don’t seem to fit either platformism or commonsism, but are interesting enough that people are desperately trying, but failing, to claim it for those families. AI and crypto are two obvious candidates. Platformism is trying to claim the former, and commonsism is trying to claim the latter. But the claims don’t really look credible. New things are being born in both cases. Either or both might create a thread of development that continues the Gutenberg tradition. Or it might be something else entirely.