The current state of the web, concentrated in a few megaplatforms, is the result of compounding complexity.

We used to have a web where anyone could learn to write a webpage in HTML in an afternoon. It's just writing text, and then using tags to format the text. But over time, people, understandably, wanted the web to do more, to look better. And so the things that were possible expanded via scripting languages, that allowed for dynamic, interactive content.

Soon, the definition of what a "website" was and looked like sailed out of reach of casual users, and eventually even out of reach of all but the most dedicated hobbyists. It became the domain of specialists, so casual users, exluded by complexity, moved to templates, services, and platforms.

This process gradually concentrated a critical mass of users into a handful of social media platforms.

Line Goes Up – The Problem With NFTs, at 1:53:23, Dan Olson

in olson's video, point of this aside is to contrast the early web (simple, free, accessible) with the modern web (enclosed in corporate platforms) and cryptocurrency (also enclosed in the same ways). the point is not to dissect the history of the web, it's to explain by analogy one of many problems with cryptocurrency. the focus of this video is, as the title says, "the problem with nfts".

the point is not to dissect the history of the web. it's a quick point of comparison: the current web is already complex, which has led to centralization of power, and cryptocurrency, being more complex, can only be more centralized. his point is correct and the point includes moving on to the rest of the video.

but despite that, this quote lives in my head rent-free. it plays on repeat, taunting me. so y'know what? fine. here's my thesis:

• we have never had a "simple web"
• but it did feel simpler
• and we can have it back
• recommendations

olson describes a web where creation was accessible to laypeople, but even "self-hosting" has never been a truly solo endeavor – nor has it ever been easy, especially not for laypeople. it's easier now than it ever was!

for example, off the top of my head, here's what needed to be abstracted away from you, for your experience to be "just writing some html":

1. a webserver, to serve that webpage.
2. some way to get them that webpage. maybe ftp.
3. a network stack that supports tcp and ip.
4. the physical hardware to run those three things.
5. an unbroken chain copper and switching between you and the hardware.
6. another chain between someone looking at your site and the hardware.
7. all the networking to make that copper useful.

notably: none of this is easy! if it seemed so, it's because other people handled all the complexity – by producing templates, services, and platforms. without them things don't get simpler, and theey especially didn't at the height of the browser wars!

olson doesn't specify a timeframe in that paragraph (because again it is a brief aside unrelated to the overall point), but the "early days" have to have been between http's invention in 1991 and the explosion in popularity of social media ca. 2010. so there's about two decades he can be thinking of, and most likely the years 1995 to 2005.

back then, self-hosting was hard. you needed a reliably available public ip address, which was impossible with dial-up and unlikely with broadband, but possible if you had friends at a university or money for a business line. you needed the hardware, which could absolutely have been consumer-grade, and thus was probably the easiest part to acquire. the operating system is far harder: you could run windows, which required paying for at least one license, or the free *nixlikes, which were cli-driven and thus harder to use. in either case, once you had your os, the webserver software wasn't a huge deal. even the first-ever webserver was an okay user experience, if you're technical enough to know what a config file is.

but i don't think a layperson would even have been able to figure out "ip addresses". luckily, web hosting services like geocities were just starting to appear. geocities was undoubtedly far easier than self-hosting, since you only needed to know that each page was a "file" to "upload". sure, there's still a barrier to entry, but a much less serious one. the fact that everyone from 14-year-olds to grandmas figured it out is proof of that.

in the former case, you don't have much in the way of help. you have an internet service provider, sure, but you have to do everything yourself. and that's why it's hard!

in the latter case, you're inarguably using a platform. it's one that abstracts very little of your web design, but everything about the hosting.

and why is that? platforms like facebook are even more abstracted, so they should be easier, right?

sure. but facebook doesn't even let you make your text bold.

i think this impression of simplicity is due to a bit of excellent design, maybe accidentally. it's not that geocities was "simpler" than facebook – facebook can be, and is, used by literally almost anyone. it's the simplest experience of all.

but html isn't much more. it really is as simple as…

just writing text, and then using tags to format the text.

so it provides a friendly enough way to build webpages, even for the relatively technically illiterate, and it's the underlying format. sites can filter or restrict it, sure, but sites like geocities let you upload unrestricted html which meant being able to do anything the web was capable of.

so you have this lovely combination of immense control, and a simple enough language for anyone to learn with minimal effort. geocities didn't even have a standard structure like myspace did or dreamwidth does today. it was literally your html, exactly as you wrote it.

and i think, despite this being objectively harder to operate, that gave the old web a much simpler feel.

i doubt that was all, though. i mean, for one thing, the web used to be way more niche. these days it's as accessible as your economic status allows it to be, but in 2000, most people only knew it in vague terms. if you were online enough to want to make your own site, you were probably more ready to learn than the average joe.

on top of that, like olson says, the expectations were lower. oh, sure, there were professional websites, they existed, and some looked great, but the standard modes of interaction hadn't been established. we hadn't yet gotten used to "just type in this box", the ultra-streamlined corporate user experience design, that prioritizes smoothly returning to the ad-stream over all else, as the default mechanism of posting. we were more used to computers being fiddly and weird and bespoke.

so between everything – the free-in-both-senses services of the time, the low expectations, my uninvited psychoanalysis of a generation of strangers – it's not irrational that, in retrospect, the web of the past feels like it was simpler.

because here's the thing: it can still have that vibe. people make a lot of noise about "semantic html" or "css3", but all that means is the exact same simple, easy structure – with a few different names. <strong> instead of <b> isn't adding to the learning curve.

these days, yes, the web has gotten more complex, and there's way more functionality around. but you can ignore it. you don't need to understand flexbox or the document object model. you can write simple, unstyled html, and it's completely legible and useful for publishing information – just a bit ugly:

if you want it to be pretty, then you have a few options:

• pick any of the many prebuilt css files that make plain html look good
• write your own from scratch, which you can do incrementally and iteratively
• start with a prebuilt, then tweak it to your liking

all of these will require a bit of learning, i'll admit, but only a bit. css has a well-earned reputation for being a pain, but that's really only for relatively complex tasks. if you want to change some colors and fonts, you can do it pretty quick:

as a beginner, you'll need to memorize a handful of "magic spells" to invoke in your css which do things like centering elements or making dark mode work. and that's okay! it doesn't matter if you're a novice. it doesn't matter if you never get good at it. you don't need a comprehensive understanding! you can just build a website!

but i think this gets to the heart of why it feels more complex. it's not that the web is harder, it's insecurity. people look at youtube or facebook and go "i could never make that." they compare the effort in learning html or typing into that old familiar text box, and they go with the one with no risk of failure. smoothed-over designs got popular because they got profitable because they work. people are scared of change, scared of failing, scared of looking stupid. and when you're comparing yourself to billion-dollar corporations, failure seems inevitable.

but bluntly? fuck that. you should make a bad website. you should throw something together, and publish it, and if it's ugly, fix it later. it doesn't need to be a side hustle or a portfolio or economically justified, and it doesn't need to look professional or slick. it just needs to make you happy. i didn't style this site with the nonbinary/agender colors because it looks the best – i did it because it made me smile! or in other words:

Do not kill the part of you that is cringe. Kill the part that cringes.

Ancient internet proverb, source unknown.

the joy of being human is the act of creation. don't deny yourself for fear of creating poorly. you don't need to make a website – but a website is a great way to partake.

but a question remains: who hosts your website?

there are quite a few options. hell, i'm making one, though it won't be ready for a while. i currently know of:

• neocities is, as you'd guess, a direct geocities successor. like geocities it adds no headers or footers, and it seems to let you add whatever files you want to your site, up to some filesize limit that won't matter for personal sites.
• dreamwidth is a livejournal fork. it has a lot more structure, but it still lets you write raw html – filtered, but you can provide custom css – and heavily customize your page, while providing a lot of functionality like comments and rss built right in.
• gitlab pages has its quirks. gitlab was designed to work with git, but you don't need to worry about that – gitlab has a web editor that lets you write your html directly on the site, after a bit of a complex setup. but it's free, and if you want to make it a hobby, it gives you a pretty good starting point for that.
• github pages is a lot like gitlab pages (and came first) but it's owned by microsoft, so i'm against it by default. but some people have reported preferring its ui, so it would be wrong not to mention it.
• if you know of (or run!) a site that enables at least this degree of self-expression – the benchmark in my head is "can i upload arbitrary html" – tell me!

p.s.

dan, on the off chance you're reading this: hi! i love your work. the phrase "bespoke inefficiencies" hit me like a truck. thank you for it.