- cross-posted to:
- opensource@programming.dev
- cross-posted to:
- opensource@programming.dev
Back in the day the best way to find cool sites when you were on a cool site was to click next in the webring. In this age of ailing search engines and confidently incorrect AI, it is time for the webring to make a comeback.
This person has given his the code to get started: Webring
@mrpalmer16 one of my favorite things back in the day was the old-school “StumbleUpon” which was like webrings on crack.
Unfortunately, advertising and profit-seeking happened.
Ah man, those times were great. Bored? Just push the button and you’ll see something new. No scrolling, just a new website with random interesting stuff to explore.
Old StumbleUpon was everything to me
Wow i got hit right in the nostalgies thinking about StumbleUpon
Oh god, I had it set as my home page for the longest time. I never got anything done but it was great having something new every time we opened our browser.
@bobdobberson @mrpalmer16 omg YES stumbleupon was incredible! I’ve asked around if people remember this and it seems that not a ton of people were on there.
That sounds amazing, I somehow missed it. Some of the other sites posted abovr seem to be trying to bring back the magic.
Stumbleupon was great. I remember having a browser plug in for it. Then I stopped using it for a little while and never went back to it.
Does it still exist?
Nope. Died like Digg and a bunch of others. There’s a run down here (which I only quickly skimmed): https://productmint.com/what-happened-to-stumbleupon/
FTA
eBay announced that it had agreed to acquire StumbleUpon for a whopping $75 million. The acquisition ultimately went through on May 30th, 2007. One of the major reasons why the team decided to sell to eBay was that it was promised complete autonomy and independence from its mother company
…sad trombone
deleted by creator
@mrpalmer16 Webrings are part of the old ‘wild west’ era of the internet that I miss. Seeing them, or something close, making a comeback would be great. So would people having webpages instead of social media accounts… but I don’t see that happening.
It will happen out of necessity once LLMs make search engines useless. Bookmarks and human-curated content will be the only way to find stuff.
It’s already affecting small businesses worldwide, who aren’t being discovered anymore by searches in their local area.
So would people having webpages instead of social media accounts
And there’s your problem… (in the voice of Jamie Hyneman, Mythbusters). To see a real return of webrings, people would need to have (make) their own pages and curate some links.
Thinking about it, with the rise of selfhosted, it’s actually really viable, cobble together a docker stack with a WYSIWYG HTML editor somewhat oriented to the task (pretty sure something out there can be repurposed), a web server, proxy, and that’s about it (probably missing a fair bit, not my bailiwick, still, once the stack is made and solid, I’m guessing many would host, I would). Set a threshold of how many people you’re willing to host, say 50 or whatever so you’re able to check for CSAM or other legal minefields, and Bob’s your uncle, stir in some solid security to keep it isolated if you’re using it at home (or VPS) and it’s golden.
OK, more complicated than I initially thought, and it’s way less friction to use something like faceplant, which is entirely their point. Still, I think, if given the opportunity, and functional tools, and low enough friction, many would prefer to have a hand curated presence on the web above a facebook page.
I’ll stop, but thanks for the interesting thought seed.
There has to be a cultural shift as well. It’s not the early 2000s anymore where a substantial portion of internet users could tinker around their desktop computers. I recently got fiber at home and we’re locked behind CGNAT. I could look for a solution for myself since I grew up opening ports on my router, but imagine someone who grew up with bubble-wrapped smartphones trying to navigate their way through that bs.
Website hosting is still a thing. Not everything needs to be self hosted.
You’re not wrong, but here we are, talking open source and GPL licences. If you can make a game portal work, or the web in general, it’s viable, your ISP is a choke point though, agreed. Was more talking about an easy stack like the 'arrs, but for webrings, just an idea…
At least now we have options like Pikapods where you can just throw a containerized server up cheap. Even people who might be overwhelmed by a VM can do that.
I also just got fiber from AT&T. I’m pretty grateful that their gateway/router can just offload all traffic to my own router and a t as just a dumb gateway. Right now I use duckdns to just public host a subsonic server for when I’m in the car or out and about but it’s been very pain free.
I read up a little on cgnat but can you tell me what issues you face? I’m curious.
Never mind… read up on it. https://en.m.wikipedia.org/wiki/Carrier-grade_NAT
I guess the alternative would be routing everything through a static ip providing vpn
deleted by creator
I love this idea, the back button on browsers feels like it exists because of webrings
It exists because web browsers used to not have tabs. Nowadays it’s useless cause with modern scripted web pages you never properly get back to the site you left
then you’re visiting websites that are badly coded
like lemmy, for example.
the back button works fine for me here
Just block JavaScript
Then the entire browser becomes useless. I couldn’t even post this comment without JavaScript.
Edit: I wish a search engine that only showed websites without JavaScript existed.
Umatrix is great, you can configure it to automatically allow first party javascript, and if sites still dont work eneable bits until they do them lock those settings so the same bits will be enabled next time you’re on that site.
You can do that with NoScript too. Is the Umatrix UI any better, or are there other benefits?
noscript is like a screwdriver. umatrix is the whole toolbox.
both have their place
I wish also that THAT search engine also made it so turning on results that have paywalls is a thing you can only have turned on if YOU turn it on
Duckduckgo has a no javascript mode.
Not useless. I don’t use JS. Yeah. Maybe 1/10 websites you have to skip.
How do you even post here without JavaScript?
I use an app, but Im sure one of the front ends doesn’t require JS
Gonna add my voice to those calling for a foss stumbleupon
Yeah! StumbleUpon was cool. Something about how it tried to engender serendipity.
Such a pity that so many other good recommendation engines died or succumbed to enshittification.
Yeah I remember very clearly — they introduced advertising and the whole thing went immediately to shit 🤷
Kagi is also experimenting with small web
Man I wanna like Kagi but I keep reading batshit things from its founder
Be like him, but don’t copy the batshit.
I’m interested in the batshit, I love weird internet lore…
Man what a trip, felt like I was hopping around the old web again.
This is like the old StumbleUpon! Thanks for this!
The idea comes up again and again on the fediverse. It feels ripe for some app/platform to kinda nail it.
I’m not sure this is it or even something that does exactly the old web ring thing. I think a simple enough system for the human curation of web pages in a standardised way that can easily be consumed and aggregated would go a long way though. The fediverse feels like its close to something.
What about https://webri.ng ?
That seems interesting!
In the end, I’m wondering if all the pieces are here on something like the fediverse but just need to be connected. I haven’t thought about this at all until now (so I’m just riffing here) … but the essence of such a system seems to me:
- Recommendations are human curated
- Recommendations come from a single human (or well defined collective)
- Reccommendations are organised in a navigable structure
Point 3 seems to be the unclear part. A “ring” is obviously a bunch of connections (not unlike a linked list). But other structures probably have a lot to provide here, especially if they’re amenable to some basic search facility.
You might be overthinking it, or I might be underthinking it.
When I hear “webring” I think of a simple list of sites, curated by the ring creator. And all members have a badge on their site, complete with a few nav buttons.
It was never broke, why fix it?
It was never broke, why fix it?
Totally fair! I don’t claim to know what I’m talking about! I’m just riffing on what I suspect would work for me, but also motivated by what I feel is a relatively urgent need to create some robust and diverse human curation of the internet. So in a way I’m not really interested in remaking web rings, but more coming from the perspective of what else can be done with the same general idea along side webrings.
Aside from 3 you are essentially creating Stumble Upon.
Sorry … I don’t know what that is
other structures probably have a lot to provide here, especially if they’re amenable to some basic search facility
I got real excited about the webring returning. This… not so much. Keep it simple.
I mean, the search doesn’t have to be centralised at all … basic search facilities could include the text search in the browser for any page, but made more user friendly and just for the webring you’re navigating or something.
So, classic webring navigation consists of arrows to the next and previous ring instances, as well as a link to the ring index. By their nature, webrings are manageable-sized communities by nature. I don’t see how that can be improved upon by a search function?
well the central site of the web ring could be searched for any particular page that’s part of the ring, and that search could be surfaced on any page that’s part of the ring.
The full set of pages could be decentralised and cached across all members for robustness, and even include each page’s own description and recommendations for every other page if they like.
And then, of course … rings of webrings with as many levels of aggregation as people are interested in maintaining, again with decentralised caches of pages, their links and descriptions (all human curated of course) that can all be searched whenever a member page or aggregating page opts into it.
Tech capabilities have advanced since the 90s enough now that basic text search in a web page over a small data set is not hard or too much to ask.
And nested rings of rings of rings are scalable because at each level the data will just be links (and descriptions or names if available) while it would be on the user to navigate the various layers however they wish until they find something they’re interested in.
Oh, man!! This looks so much like one of my old blogs!!! The layout, the colors! Brings back great memories.
There is also Gemini protocol!
I’m aware of it (and while not being super enthused about it, I can my personal interest growing over time as the internet keeps tracking the way it is).
But how does it help with a page recommendation system? Is there a strong culture of that sort of thing on Gemini?
in gemini you typically find pages using Antenna (gemini://warmedal.se/~antenna/) here you would find different blog all across gemini. also you could go to bbs(gemini://bbs.geminispace.org/) to discuss stuff of variety of range.
Man, their website is pretty off-putting. Where’s the get-started/dive-in type page? How do I use the thing?
Iirc it’s Geminispace.info
I could be wrong since I only visit their Gemini capsule.
I’m down.
You gotta be down to know whats up.
Stumbleupon was fun.
I miss old web shit.
Ninety zeros dot com was one of the Internet’s weirdest best things.
I’m in a few webrings! https://wetnoodle.org they’re under the navigation menu towards the bottom
This is a great idea. I didn’t see a Linux subway yet, but the process for requesting new lines seems pretty simple.
Maia Arson Crimew, one of my favorite hackers, is in a webring https://maia.crimew.gay
Oh man that site looks just like the internet before it started to suck.
This is so cool!
What would be really cool would be an open source, federated version of DMOZ
Yes, please!
how would you federate? it comes natural for lemmy to have each community on a seperate server, but how would you do this for a project like dmoz?
i don’t think it would be a good idea that one server could own “art” for example, and no one else could contribute. and on the other side it would not be a good idea if everyone could add sites for “art” as then it’s just a federated wiki? you still would have to fight spam? do all entries in “art” have the same priority? or should there be some voting, or verifying from other instances maybe? but then rough instances could vote for each other?!
how big is the spam problem on lemmy?
I don’t know, but it could be interesting to try. I could easily imagine topic-focussed servers that go into more depth on specific topics. Perhaps you would only federate things that are at a high level, or directly linked. Kinda like a wiki, but with each community doing it’s own decentralised curation and moderation…
I haven’t seen any spam on Lemmy yet, and only a tiny amount on mastodon (I’m much more active there).
Neocities does this right?
They do indeed
hexbear’s trans comm just hooked into one! super cool