By Charlie Owen on

What would it take to bring webrings back?

When I saw this tweet by the ebullient and lovely Tatiana Mac I was intrigued. What would it take to bring webrings back?

"Webrings"? WTF are you talking about grandma?

So, for the whippersnappers amongst you, back when dinosaurs roamed the Earth (the 90s) we had the concept of a "webring". A webring was a bunch of related sites listed together in a central directory. You could visit this central directory and find a link to any of the member sites from there.

This is just a tiny list of related sites (a database if you're feeling grandiose). But what made it special was that to be a member of this list you had to embed a little visual component on their page that contained a "forward", a "backward", and a "random" button.

These components and their buttons are where the magic happened: The previous and next buttons would move you to the previous or next site in that list, while the random button would, you guessed it, take you to a random site. Each site would have its own buttons and so when you arrived on any given site you had the ability to move to the next site, and so on. If you reached the end of the list you would loop around to the first entry again.

A ring of websites. A webring.

This may seem a bit bizarre to readers in 2019, but you've gotta remember that things were different back then. There were no search engines as they exist today. Google was still an academic side project at Stanford. Social media didn't exist (oh my god the bliss). You couldn't see what was trending, or see a retweet. There were no "278 friends are talking about this". You relied on word of mouth for everything.

So webrings were ENORMOUSLY important for discovering new and exciting content back in the Age of Innocence. A 50 user per month site could find itself "next" to a 5 million visitor per month site, which would then start passing visitors to it. Hitting random would take you to things that you shared an interest in, but would never have any other way of discovering.

I recall (because I am of the 90s web generation) that it was a proud badge of honour to have a Webring embed on your site. It showed what you stood for, it showed what you were interested in. If "intersectional queer anti-capitalist" had had a Webring available back in 1997, I would have been on on that Webring and I would have had the badge displayed as prominently as god dammed possible.

So it's a blogroll then?

You might recall blogrolls, and what a good memory for 90s concepts you have if you do! But a blogroll isn't quite the same thing as a Webring. A blogroll was (is! They're still around!) a way for site owners to list other sites that they found interesting and would recommend to their readers. It was maintained and run on an individual site without any external dependency. So related, but not the same.

Hmmm, sounds a bit old guard to me

This is all of course very Old Guard and should immediately be viewed with suspicion by any rollneck-wearing Frontend Engineer. Why, in the age of highly effective search engines and the connections of social media, would we want to bring back something so decrepit?

Randomness and self-discovery is the reason.

We often see sites surfaced by social media. These might be good, they might be bad, they might be nothing. All we know is that in the hyper-polarised world of social media they've caused a highly emotional reaction in bunch of people - they've been told about something new to love or to loathe. The mob has ruled on what you will find interesting.

What if we could just find random interesting sites on a particular topic instead? Not just whatever is causing an argument right now on Twitter, but something that is in a field that you're interested in and might be useful and why the fuck shouldn't you be looking at things that aren't simply being pushed into your face by The Algorithm?

Imagine if you could have a Black and PoC webring, that was curated and free of white fuckery? Imagine a webring that had only trans people on there, free of cis nonsense? Places where talked-over voices could be heard! What an amazingly precious thing that would be!

A lot of webring concepts sound very similar to the Mastodon instances model - little communities of people huddled around a shared ideal or love of a topic. The problem with Mastodon instances can be that they are so often closed, and that the suffer from the normal social media problems of emotional dopamine-hits and polarisation. Long form discussion is not - CAN not - be encouraged.

A webring would, in contrast, be an open collection of personal websites, with whatever content the author wanted on there. If you visit and you like what they produce on there then you can save it to your bookmarks, or subscribe to it via the equally 90s concept of the RSS feed!

Make the web personal again

We suffer from homogenous dirge in most of our contemporary web presences. Having a personal website has become a rarer and rarer thing in this time of social media profile pages.

However, recent months have seen a surge in personal websites and blogging amongst some members of the web tech community. This is something that we urgently need to encourage! For our ability to muse and think has been diminished by social media. It's all or nothing, the joy of cat gifs, or the existential horror of global warming. A group has angered the collective! You must join in! Want to ponder and think? No time for that, the next neurotransmitter squirt is due! (I do not mock anyone for being angry about the injustices in the world. But I fear it is killing the mental health of the people who care).

Webrings could be something that gets us back into a slower world of personal sites and personal, very human stories. None of your "personal brand" (unironic use of which needs to die in a fire, as it is simply a way of defining your worth by your employability). Bring back the human brand! Bring back the ability to find smart amazing, original, bizarre, wonderful people!

So what can we do?

I think we're in a good position to start thinking about webrings again. But lets bring them back with the current climate in mind.

One of the dangers some people fear for webrings is that they would quickly become clique circles. Get all your mates together in a Webring and circlejerk your way around their sites.

It could just as easily go the other way. A Webring for trans people would likely get infiltrated by TERF sites where they declare about how they "self-identify as an attack helicopter". Webrings for women would suffer from MRA infliltrations almost immediately. Any community that could potentially result in this new age of webrings would be quickly destroyed.

Therefore curation would have to be paramount. Because the technical implementation for webrings is trivial - it's the human side that would be the most important thing.

Luckily we've got plenty of models in that area. OSS software has been grappling with this for a long time when it comes to reviewing code, and more recently when enforcing codes of conduct.

If the central list for a webring were to be implemented somewhere like Github then a bunch of interested and appropriate admins could receive applications via Pull Requests and review each one. If they are non-problematic, then simply merge that submission into the central list (I assume something like a simple JSON file, with URL, title, and owner contact details).

It doesn't have to be via Github either. And ideally it shouldn't be, because we are trying to make this work for everyone, and not just web tech people. It would be just as easy to maintain a simple CMS that anyone could submit to via a form and a bunch of authenticated admins could review.

Either of these solutions could then publicise a JSON file that is used by navigation embeds on each site.


And what of the webring embeds on each site? We could go the original 90s way and publish a little snippet of JS that is placed into a page by the author, much like Tweets are so often embedded. The snippet could then ping the endpoint with the current URL and get back the next, previous, and a random site. Or it could just grab the entire JSON object and utilise that.

Even better, if the webring member is into static sites then they could pull the webring JSON data in at build time and embed everything as HTML on the page - no need for an extra script and API call at all!

And so much additiona data could be in the webring site entry! Author photos, last updated info, summaries. The author doesn't even need to submit all that stuff - if we encouraged more use of Microformats we could just pull that off the page and store it alongside the original submission.

Let's fucking DO THIS!

Yeah, let's do it! (er, not me, as I'm a time-limited tammy). But someone, yeah, go do it!

Just bear in mind that we probably can't use the word "webring" itself: upon searching the interwebnetwork I was amazed to see that "webring" is an actively trademarked term (go to the US Gov Trademark site and search for "webring" to see more - sorry, it won't let me link directly).

So we need a new funky name.

Something like "CircleJerks".



Okay, tell me a better name.