Bring Back Webrings | Hacker News Reader

Bring Back Webrings

arne.me

120 pointsabahlo88 comments

I'm fairly certain that link-aggregation sites, like Hacker News, Reddit, Digg, Slashdot, and Lemmy have taken the place of Webrings.

The sad truth about webpages is that people don't want to maintain them. People will put in their weekend project, and then the webpage sits there for the rest of eternity whether or not its relevant, and then what? When do you update the webring so that they add and/or remove pages?

Here's another idea: you put up links regularly to a webpage that dynamically sorts them by popularity, relevance, and date. Oh wait, that's Reddit.

> I'm fairly certain that link-aggregation sites, like Hacker News, Reddit, Digg, Slashdot, and Lemmy have taken the place of Webrings.

Link aggregation sites serve a completely different purpose from webrings, and don't substitute for them. That's why for many years, link aggregation sites and webrings coexisted.

The US Military was still doing horse cavalry charges during WW2 in the Philippines even as battleships, carriers, and tanks charged forth.

There's a lot of "momentum" to technologies. I'm personally convinced that webrings are one of those ideas that died for good reason. There's just easier ways to organize ourselves online.

I promise you: if you want a big list of links to follow, just start a Wiki. Or share them in your own Lemmy / Mastodon. Or open up your own Subreddit. Its going to be easier.

----------

If you want to rebuild the feeling of early community-driven internet, then you should be looking up IndieWeb (https://indieweb.org/), and not just trying to resurrect random technologies from 30 years ago.

People who are arguing for webrings are not claiming they're easier. It turns out, if easy is your only metric, Everest would never have been climbed.

> I'm personally convinced that webrings are one of those ideas that died for good reason.

Except that nothing came up that replaced them.

> if you want a big list of links to follow, just start a Wiki.

What wasn't the purpose of webrings, though.

I have no special love for webrings as a mechanism. I do have a love for what they did socially, though. If something else came along that served the same purpose, that would be awesome -- but so far, nothing has. I think that the main reason for that is that the web itself has changed from a place for people to a place for commerce.

Social bonds are formed by humans, not technology. In effect, the technology used is almost a side-effect. You can build social bonds through emails, letters, walking and talking with people.

Today, kids are using Google Documents to communicate with each other in classrooms (ie: sharing links, collaborating on homework, etc. etc.), They don't need Webrings to share a set of links that happen to be on the same subject. In fact, you don't need any real tech at all, it could all just be a bulletin board, it could be a phone number that you leave somewhere with a computer hooked up that everyone connects with a 300 baud modem to. Etc. etc.

Or ya know, a Tweet, a Reddit post, a Facebook message. A google group. Etc. etc.

Yeah, not sure if you lived through webrings because I don't think anyone that used them would agree.

You click until you find a broken link in the chain and then everything goes to crap.

When you have A, B, C, D, and E, all different webmasters on different parts of the internet making a webring of A -> B -> C -> D -> E -> A, things get really messed up when C stops responding to emails.

As others have said: you need a ringmaster for this to work. When C stops responding to emails, you tell B to update their page and point to D, for example.

------------------

We have so many more technologies today than we did in the 90s. I don't know why anyone would look back at freaking Webrings when we have... I dunno... Wikis?

Now the Wiki administrator (A), could recruit B, C, D, and E to be moderators on the same Wiki. When C stops responding to emails, everything keeps going just fine.

Etc. etc.

I don't think anyone who actually lived in the 90s with directory services, Ask Jeeves, and Webrings would ever think about bringing those services back. Like, the good stuff were IRC, AIM, IRQ, USENET ?

I think you have missed the point, it's more about decentralization and increasing "locality" of the internet, get people close to each other or with shared interest to have their own communities. Webrings may not be all that great of solution technically but it does the job in that front.

Like in your wiki example, if A falls, now you don't have A B C D E F... well, the whole alphabet is gone :)

> Like in your wiki example, if A falls, now you don't have A B C D E F... well, the whole alphabet is gone :)

Its far more likely for random individuals of a webring to disappear in my experience.

Ex: A could be the starting point of the webring. When C disappears, then D, E, and F are inaccessible. (A -> B -> Broken page).

A wiki is a single point of failure: you make A more reliable so that everyone has the information. A webring is an infinite number of single points of failure: a single failure (broken link / 404, etc. etc.) messes the entire ring up. Its horrible.

Webrings don't have to be maintained, they just have to be up. They are snapshots in time that provide lots of information. Authors typically put in a lot of effort, unfiltered. The individuals voice shining through is the goal, learning more through links is a plus.

Link aggregation sites don't work well as a reference. They're just the links that are popular that day. Aside from the wiki pages for some subreddits, they don't do a good job of storing a definitive set of links on some subject.

> maintain them

Maintain what?

They’re documents. They document things. As long as there’s a tool that knows how to parse the document, which is outside the role of the author, the document remains complete. There’s nothing to maintain.

The contemporary fetish for timely ephemera is a quirk of social media feeds and a generation that grew up immersed in them, not some baseline criteria for how the internet needs to work.

As automated search engines and social media feeds continue to drown in spam and engagement-porn, expect to see a resurgence of hand-curated web rings, directories, and marginalia-like scoped search engines, that variously highlight timely, historical, and evergreen content.

You’ll be surprised what an untouched document from 5 or 15 or 25 years ago might reveal to you, once you can actually find them again.

> Maintain what?

Okay, you're kidding me right? Have you actually used Webrings?

Lets say I click on a webring, and the "next" button goes to a 404 error. Now what? How do I access all the other links?

Answer: you can't. Its basically lost information. Webrings require ALL the web-administrators in the ring to keep their previous-and-next links up-to-date, otherwise the whole ring collapses.

There's a __reason__ why we stopped doing Webrings when Geocities stopped being popular. I've lived Geocities -> Homestead -> Xenga -> Myspace -> Facebook. At no point did anyone ever care to go back to webrings.

And suddenly here we are like 20+ years later, where people who clearly never used them are suddenly pretending that it was webrings that made the early internet great. Erm, no. I guess they were a sign of the times... but they weren't good or great by any means. We have better means of sharing links with each other today.

--------

Links die. Alarmingly quickly. Even today where people try to have long-lived links for SEO purposes and have specifically programmed scripts to help make links live longer... links still die and thus webrings break.

We didn't know how bad link-rot was at the time of webrings.

> You’ll be surprised what an untouched document from 5 or 15 or 25 years ago might reveal to you, once you can actually find them again.

Good luck. Geocities literally died. Yes, there's an entire archival process was undertaken to try to save Geocities, but I'm sure we missed some info.

Any webring pointing to a Geocities site today will absolutely 404 error out. You've already got to change all the links to point to the various archives (ex: Neocities IIRC has at least the most popular pages archived).

Now I ask: were you really around for the time of webrings?

> were you really around for the time of webrings?

Yup.

> Webrings require ALL the web-administrators in the ring to keep their previous-and-next links up-to-date, otherwise the whole ring collapses.

Nope. Not then, when IFRAMEs or a pre-CORS AJAX call delegated ring maintenance upstream, and not now, when we have about 1500 more ways to delegate it. Was there a short period where they were as crude as you describe? Sure, but that's not really relevant to the forward-looking discussion of what they would look like now or the backward-looking discussion of what they looked like in the many years where they were mature and popular.

I can’t arbitrate here, but can add my anecdote to bolster what the person above is saying: I regularly remember clicking next on webrings and eventually getting a 404 with no way to progress. This was fairly regular to the extent that I sometimes just avoided clicking the webring modal.

Yeah, I remember it both ways. Sometimes you'd click on a link at the bottom of one page to hit a dead end, but you could also go to the homepage of the webring itself and you could navigate through the sites in an iframe where you'd still run into a bunch of 404s, but you'd at least still have a "next" button in a separate frame so you could roll the dice again.

Webrings require maintenance to be useful and worth using, just like search engines do. The better maintained and curated they are, the more useful they will be.

> just like search engines do

Given how easy it is to run PostgreSQL and full text search these days, I think boutique search engines for small communities makes more sense than a webring.

>Lets say I click on a webring, and the "next" button goes to a 404 error. Now what? How do I access all the other links?

An example of this from the OP link. The page links to a11y-webring.club which has prev/next/random buttons that all produce this error:

This function has crashed

An unhandled error in the function code triggered the following message:

Runtime.UserCodeSyntaxError - SyntaxError: Unexpected token ':'

Imagine if this criticism was levied at books.

James Clavell obviously doesn't care about Shogun, why ... it hasn't changed for years!?!?!

Tangential, but one thing I noticed recently is how algorithms on sites like twitter and linkedin - especially linkedin - will penalize posts that contain external links. So it's very hard to even tell people about content on your own site. HN might be one of the few places left that don't seem to do that.

Thinking more - what used to be common is a section called "other cool sites" or something similar, which would just be a list of sites to check out the author put there. Maybe that's a bit more robust than a ring.

The algorithms are probably trying to fight spam, but the end result is a closed, self-referential silo.

That's a charitable take.

I assume they're just trying to keep you on their domain for commercial reasons.

LinkedIn in some cases imposes penalties for outbound links [0].

Consider that LinkedIn has been trying to establish(/convert) itself in part as a blogging platform, and keeps trying to show you other users' blog posts, unsolicited. But not links to Twitter/X, Substack, Rumble, external blogs etc etc. Even though many LI users' blog-type posts are low-grade, it's near-impossible to get them successfully flagged as spam or self-promotional. Every time LI serve you a link to some other content on-site, and suppress serving you an outbound link (even if it was higher-quality), it presumably increases engagement on LI.

[0]: https://www.linkedin.com/advice/0/how-do-you-avoid-recover-f...

This webring isn't so much a ring as a wire, as one of the websites doesn't have the footer. This was always the problem with webrings in the first place - one broken link in the chain ruins it for everyone.

Just like token ring based LANs

Just rebrand them as WebDAGs.

I kind of like this idea, but then instead of just left and right arrows I want to be able to travel in a multi-dimensional space :)

This could be pretty basic browser extension. A little banner at the top or bottom: “websites like this one”

Or website curators all agree on a /webring.txt file (similar to a robots.txt) where they list websites they think are similar?

An important part of what made webrings good, though, was that they were curated (some more than others, but still...)

Having sites just list others the site owners think are similar is just the site providing a list of links to "friend sites" -- which also used to be a common thing. That's also fun and useful, but not what webrings are about.

I'm not sure I see the difference in curation between a webring and a "other cool sites" list, except that in the webring case there needs to be a central curator across several sites, so it's effectively just a single person's "other cool sites" list spread across several sites.

"Other cool sites" lists were quite popular back then. That's what I used for site discovery instead of webrings.

The best webrings were curated by topic rather than just being a list of sites the ringmaster likes.

Links to friend sites - does this have a name?

As a way of discovering content, I love the idea.

Or... or.... a little banner that says, "From around the Web:"!

Links between websites in a sort of hyper-dimensional space! A...hyper-link?

Last I checked, rings were usually cyclic.

And also commutative, if I remember my math right, so you can traverse them in both directions. Punning on the word ring (which is a math concept too, as in groups, rings and fields, and stretching the metaphor too, or is it a mixed metaphor? Heh.

Directed acyclic graphs?

It should be possible to periodicallt crawl the ring participants and exclude them if they don't have the footer, right?

Could do that before the user is forwarded to the website. Remove them for a day, then 2, then 4, then 8, 16, 32, 64, 128 etc it is not expensive to keep checking. Could send email too.

You'd probably want to measure click though rate and do manual review. Could also have a redirecting page so that the back button allows the visitor to flag or report something.

But definitely only remove them for powers of two days, because... geek cred.

I use it for rss feeds, it really follows how likely it is for a feed to return. Checking every day was slowing things down.

Yes, good webrings always needed an active ringmaster.

I do think webrings are badly needed again, though, to serve the same purpose they were developed for in the first place: discoverability.

It's become very difficult to find the good (according to my tastes) websites these days.

Same here. Very difficult.

Web rung?

Nice.

Let's see ...

I sprung from rung to rung

Skipping past the dung

On my wall let it be hung

That my feats remain unsung

Webrings still exist. They're not mainstream. To the extent that the internet is "ruined," it's that the mainstream internet is ruined. There's no going back from this. The only option is to ignore much of the mainstream content on the internet. It will always be there, but critical ignoring is a required skill on the modern internet. The beauty is still there, it's just buried under mountains of trash, and your friends only know about those mountains of trash. (instagram, tiktok, youtube, etc.)

"critical ignoring" love this term

Everyone has made some good points about webrings and other methods of getting personal websites noticed.

Webrings should have a place - they were great for bringing similarly themed sites together and were a way of finding smaller, personal sites. The latter is probably more important nowadays than ever.

The webring revival isn't doing so well. I write the list at https://brisray.com/web/webring-list.htm and I recently visited evey link in every webring listed. For whatever reason, the ringmasters or the site owners, only 20% of them are fully navigable. For most, clicking the next link will eventually lead to a 404 page.

These new webrings have other problems, one of which is the member subject range is much wider than that of the orginal webrings so you never know what you'll be looking at next.

I list webrings that have at least a back and forward link, to help everyone they really should have both a member list page and a random link. That would go a little way to stop some of the dead ends.

Links here, Reddit, Facebook or whatever are not the answer. They are ephemeral while webrings were supposed to be more or less permanent, at least until they were removed from the ring.

Clique listings and directory pages aren't that great either. The rate of link rot is phenomenal. Even after just a couple of months the links on my own pages start breaking. Links on pages more than a couple of years old are barely worth clicking on.

Personally, I'd like to see webrings make a comeback, but unless they get a grip the new ones are going to wither away because they aren't navigable.

Related last week:

Ask HN: Why is there no effort to bring back webrings?

https://news.ycombinator.com/item?id=38177128

And this thread with some 'rings and related threads listed: https://news.ycombinator.com/item?id=37577861

I always found plenty of interesting websites without needing to use webrings, often from sites with a list of "other sites about [whatever this site is about]" or just "other cool websites". Such personally curated lists were far more useful to me. They were less random, they could show the title, they could even put notes about the sites, I could skip over ones that I already knew about or didn't seem relevant, and they were overall less random. Also, a huge plus for implementation side of it was that it didn't rely on anyone else except you.

Are there any advantages to having webrings instead of just lists of "other cool sites"?

edit to add: Now that I'm thinking of it, a webring is a ring (duh), but the "other cool sites" concept is a decentralized network. That to me is a huge benefit. Makes it far more robust.

I was thinking a webring like solution would allow discoverability while still owning your platform. Discoverability is a major concern I've heard from many bloggists.

You could have a footer that is some simple js that allows you to go to a sampling of related articles. Could track cross-domain referrals and give them a cut of generated ad revenue (not sure how you'd track that just yet).

To keep some of the riffraff out, you can charge a small fee to join the webring. Have the rings be able to manage their members to facilitate the removal of junk links.

Might make a nice small saas; though I'm not sure what the price point would be to make sense. Each ring priced as low as a $1/mo? $5/yr? Maybe a popular ring is worth $1000/mo.

One of the appeals of webrings is they used to link sites made by people passionate about a subject.

If you link "content creators" they'll have no value to the reader.

I was thinking about this less an hour ago

Specifically I want my personal site to be a part of a “web 1.0” ring, where everyone uses technology that predates web 2.0 frameworks (Wordpress, AWS, CF etc…) and you generally roll your own everything.

Publish a post listing such sites, with a page title of "Web 1.0 Webring", accepting pull requests for additions?

If/when it grows too big, someone can add software for collaborative maintenance.

Starting point: https://old.reddit.com/r/webdev/comments/caezeo/web_10_era_w...

I like the idea, but ugh do not bring back the asinine "Web 2.0" term. Fortunately "Web3" has failed to gain the hype "Web 2.0" did.

Stupid labels should die already, starting with "Gen [whatever]."

What term would you use to differentiate?

Differentiate what?

Between types or classes or generations of websites

I’m suggesting that we link all websites that exist in a unnamed class that includes the attributes of websites published prior to 2003 (as an example criteria based on when wordpress launched).

What would you name such a class other than Web 1.0 or old web or some such

"Non-Wordpress?" Otherwise that's just an arbitrary cut-off. If you want to convey something, you need to state it in English. "Pre-Wordpress" is another possibility.

Related. Others?

Why there is no effort to bring webrings back as search quality has declined? - https://news.ycombinator.com/item?id=38177128 - Nov 2023 (12 comments)

Webring Technology - https://news.ycombinator.com/item?id=37577861 - Sept 2023 (30 comments)

Ask HN: What Modern Alternatives for WebRings are there? - https://news.ycombinator.com/item?id=36630750 - July 2023 (3 comments)

Webring History - https://news.ycombinator.com/item?id=34846719 - Feb 2023 (12 comments)

What ever happened to webrings? (2015) - https://news.ycombinator.com/item?id=33585201 - Nov 2022 (111 comments)

Homebrew Computers Web-Ring - https://news.ycombinator.com/item?id=32033833 - July 2022 (11 comments)

HomeBrew Computers Web-Ring - https://news.ycombinator.com/item?id=29272595 - Nov 2021 (14 comments)

Mischa's Cursed Webring - https://news.ycombinator.com/item?id=26358952 - March 2021 (92 comments)

Ask HN: Why are webrings not a thing anymore? - https://news.ycombinator.com/item?id=26242005 - Feb 2021 (1 comment)

Show HN: I am trying to start a webring for geeks - https://news.ycombinator.com/item?id=23549471 - June 2020 (89 comments)

Bringing Webrings Back from the 90s - https://news.ycombinator.com/item?id=19683896 - April 2019 (3 comments)

Ask HN: Anyone want to start a webring? - https://news.ycombinator.com/item?id=17618082 - July 2018 (4 comments)

I was old enough for a geocities homepage, but not old enough to get that a webring was a web-ring.

I thought it was some cool secretive club of people: “we bring you internet stuff, and we’re all friends. Maybe we’ll let you in, kid.”

Then I got a bit older and thought it was a lewd reference to circles of jerks.

It would be years later, after the blogosphere collapsed, and social media had given us all PTSD, when I first heard someone explain what it was.

Just a few minutes ago, actually! ;)

Sure.. with todays net, you'll end up with 100's of frameworks to implement the ring, probably based on the block-chain tech. Then don't forget the WRaaS, webrings as a service, paid subscription, NFTs, sponsored webrings followed by AI Web Rings. And maybe a bag of onion rings to go with it.

It is not a ring, but it can help you feel like it is: Someone shared this search site a while back: https://search.marginalia.nu/

It is a pretty great way to find smaller and more personal websites.

Marginalia has a URL parameter (&profile=vintage) for Web 1.0, https://search.marginalia.nu/search?query=hacker&profile=vin...

That put a smile on my face; thanks for sharing. #nostalgia

I don't mind it. But as others have already said, a single website can easily break a web ring. Why not a simple "I'm Feeling Lucky" button that takes you to a random website in the ring?

Do I misremember when I think that every webring footer had a button to take you to the admin's page? Which may or may not have included a list of sites or at least a random jump button?

Just in case anyone is interested, I run the https://webri.ng platform. It's a platform that lets users create their own webrings without any coding required. This might come in handy if anyone here wanted to create a webring of their own. It actually hosts one of the webrings the author lists in their article.

Links could have the webring in the query string so that the pages can put a sticky navbar at the top of the viewport.(edit: could put #webring behind the link, give the navbar id="webring" then style it to stick to the top with #webring:target{})

you could link to:

webring.com?site=my-page.com&nav=next

then visitors end up at

yourpage.com?webring=my-page.com

yourpage.com#webring

If you want it to work without a domain you could also have many next and previous buttons.

<<<<<<<< webring >>>>>>>

https://git.sr.ht/~sircmpwn/openring

I suspect that if there were a significant return to webrings we'd run into very similar problems. Shitty companies would pay to insert their garbage into popular webrings run by other people, and/or they'd start their own webrings and fill them with spam.

All nostalgia is a longing for youth.

ActivityPub potentially helps to solve this. With enough well-exposed/accessible connections to sites or posts you might be able to traverse around a network and collect similar items. There are a lot of other challenges there but it's a possibility.

Honestly what I really want is the return of DMOZ. A curated, categorized list of the best sites.

Yes, I miss DMOZ more and more as time goes on, as well.

There was a spiritual successor to dmoz: curlie.org

In order to "bring back webrings", one would first have to "bring back private websites", private as in not corporate owned, and not being a walled garden, like all the Meta services or X or Discord or Slack or or or.

https://htmx.org/webring

My partner works in corporate communications and so we frequently lament the enshitification, specifically the way social media as a concept destroyed so many things and now is experiencing its own gotterdammerung. It feels like at the moment we are in a middle place, waiting for the next trend. Perhaps it will be a return to the super personalization of the 1990s, where everyone was making content of their own and expressing themselves online with their own flavors before the "platformization" came along. Post-Covid there appears to be a resurgence in community, in "authenticity" so logically it follows that our online engagement might follow a similar pattern.

Yeah, can't wait to monetize that! Oh, darn...

I mean, your skepticism isn't wrong as it's a definite catch-22... people want tools to share and create their own content, but those tools cost money which means someone has to get paid, at which point the vendors start thinking in terms of greater lock-in and embrace-and-extend opportunities, and the cycle repeats itself.

This was supposed to be the promise of open-source software: people making tools to share with everyone for free, so people don't need to pay vendors just to do something simple.

However, while software can be made free and open-source, someone still has to pay for servers to run for any of this stuff to work. These things were easier back in the good ol' days when everyone just used university computer resources. Nowadays, university tuition has skyrocketed, while universities are no longer freely hosting internet resources like before, I guess because they need to pay their deans millions and build really fancy new dorms.

I'm honestly starting to find the 'web nostalgia' a little nauseating, I kinda like gemini, but it lacking inline images and other 'modernities' like that take it a little too far back. The web overall will never go back to how it was in whichever era you idolise (for me it was 1992-1994), so it's probably best that we stop wishing it would.

Awesome! I couldn't agree more! Thank you.

The OG distributed index.

These sort of blog posts are becoming increasingly unsettling to me.

Reply on news.ycombinator.com