36
votes
Tildes is awesome
I just joined, and although it’s not extremely active, I love Tildes already! Firstly, the user interface. This is what the reddit redesign should’ve been. Clean, simple and lightweight. Loading a post on new reddit takes 10 seconds or so, because of all the useless JavaScript, but posts loads instantly here. And there’s no ads here, which is a nice bonus.
I also like the fact that it’s much calmer here, people focus on through-provoking discussion instead of attacking each other.
To everyone who works on Tildes, keep up the great work! I’m sure this has been asked many times before, but do you ever plan on allowing anyone to register in order to grow Tildes?
Smaller social media sites each seem to host one of two distinct self-selected populations:
People who got banned from the big sites for extreme toxic behaviour and want a zero-moderation zone
People who are exhausted with the negativity of social media and want a responsibly-run site with a good community
Obviously Tildes is the second type. It was a real relief to me to find it, because it was not clear to me that many type 2 places existed online these days. Mastodon, at least the parts I frequent, is quite similar in some ways.
Very accurate :) I mostly just lurk around tildes, but it is the web community I enjoy the most.
Not a creator, but can tell you the closed nature of the invite system and slow/no growth helps preserve the existing culture. Letting too many people flood in at one time can easily make their existing dominant attitudes towards posting slip through. People need time to detox from toxicity and adjust to the new vibes. Plus keeping invite-only makes it harder for spammers and bots to sneak in unnoticed.
Even now I personally struggle to minimize or eliminate the harshest of my negativity that has been bred from 3 decades of the thick skin built up, neccessary to survive in the harsher communities.
Taking a positive, constructive critical attitude to some frankly abhorrent topics is difficult, if impossible...but if you look through the history of the site there are many topics that are usually a dumpsterfire elsewhere survive far longer with less moderation. There's a bit of bias when looking through the past, given the worst discussions do get locked or removed...but those are a small minority of the whole.
If we want Tildes to evolve at a faster pace, the first step is getting Deimos working on it full time. That means donations, and that means we need growth. Somewhere in the ballpark of $4k-$5k a month is where that can happen.
I'm still plugging invite links and metered open signup as the next logical step. That plus a smallish cleanup of the build documentation, making the hard-coded references to tildes.net variable and easy to change so others can more easily use the code as the basis of their own websites. Those changes and better signup tools will make the code instantly more attractive to other people who want an aggregator. It should ideally take less than an hour to set up a site based on the Tildes code in a VPS.
More sites using the code means more code contributions and more donations even if it doesn't grow the userbase of this particular tilde. I think this codebase is about 95% of the way to being for link aggregator sites what phpbb is for forums. Why bring all of reddit's communities here when they can be set free to find their own destiny? :)
Someone mentioned in an older thread that Tildes was missing its niche as an open-source powerhouse by remaining just one website, and after reflecting on that I agree. Can't find the comment or I'd link it.
Oh, and there's lots more to read about the site and features/ideas, mostly confined to the ~tildes group (and the ~tildes.official subgroup). Man, has it really been two years since the last introductions thread? Wow.
One other quirk getting this code running on dozens of other sites might bring... federation. We're always hearing about that, Deimos isn't interested and I can see why - the moderation problem is the worthier challenge, federation will instantly complicate it into insanity. People mistake federation as a silver bullet for 'sticking it to the man.' We've had federated services since the dawn of the internet, long before the web. It doesn't help.
I think it's premature until the moderation problem is solved. Trust system takes priority. If there are dozens of Tildes sites out there, we'll start to get an idea of what a tildes node actually is, what they can evolve into as they try their own experiments. The AGPL will keep all the code that diversity generates raining down like manna from heaven. So what would federation even look like?
Seems like it's destined to replicate usenet in some respects, though doing better than that should be easy. Rather than trying to federate the users across sites, perhaps it should be the content that can flow between sites. Carry that over into comment threads that allow the users to interact within that thread despite being from two different tilde nodes - just ghost the other server's users in with a username@node nomenclature, maybe even allow encrypted cross-server PMs. We can maintain a map of not-trash nodes here. Perhaps what we think of as a 'group' could also be its own server.
This sidesteps our chief problem, I think. Now we don't have to manage users and trust as part of the federation - that's internal to each node. Simple tools can block servers, not all nodes need to associate, opt-in. I wonder if activitypub would be able to pump it all around, save the work of building the plumbing.
I mean... doesn't it? I think it's a great tool when the point is for anyone on any machine to be able to talk to anyone else, and when you need to scale horizontally with that constraint; IRC and e-mail are successful federated systems from before the web that remain extremely popular today.
I do not think this would save much work. It's a very complex standard and Tildefed wouldn't need most of it.
I agree completely with this. Personally I would like to be able to block users at the very least.
While on reddit, I recommend sticking to old reddit. Some subreddits do not update their sidebar on old reddit, and some multimedia submissions will throw into new reddit, but otherwise it's just a better experience.
Yeah. Sadly, I can realistically envision them decommissioning it in the future though.
Yep: Simply start the URL with "old". So, "old.reddit.com" should send you to the old layout. I'm still on reddit to check some of the niche subreddits but the day they disable the old layout, I'm out for good.
Alternatively, if you're lazy like me, there's a handy plugin to guarantee you always land on old.reddit.com no matter the URL you follow. Saves changing it manually and it also prevents reddit's habit of making sure you click out of old.reddit into the new interface. It's called Old Reddit Redirect for both Chrome and Firefox.
I'm rarely sent to new reddit, and I vaguely recall from back when new reddit was first introduced that disabling new reddit from old reddit's settings worked better ("stuck" better) than disabling it from new reddit's settings. It's under "beta options > use new reddit as my default experience". It's always worth making sure that checkbox is unticked.
Welcome!
Thank you!
Seconded, welcome to Tildes. And if you're into high quality youtube content, check out the videos tag, a few of us (myself included) post a lot of the content from the best channels to it!
Hi, can you tell me if I'm doing this wrong? I search for videos and click the tags inside threads, is there a way to globally browse the videos tag? Thanks!
https://tildes.net/?tag=videos
If you click the tag from inside the thread you’re restricting to the group. From the global list it’s a global search. You can also see it in the url :)
That's what I was looking for! Cheers.
As much as I think it would be worth opening up Tildes to registrations, it would require a much bigger moderation team.
Ideally, one of these days, someone is going to get around to implementing deimos' ideas about distributed moderation / trusted users, which would help a lot in moderation, making growth a reasonable proposition.
One day.
Do you think that would mean trusted users would be given permission to moderate specific boards, kind of like reddit subreddits?
Generally I suspect most people who actually want to be a moderators are probably on the bottom of my list of people who should be moderators. So any system where people are politicking and angling to be given this sort of authority around a topic that they, presumably, care about and have opinions on is probably going to do a poor job of selecting ideal people.
What's more, being a moderator kind of sucks and is likely to cause you to burn out on it. I've basically taken a back seat on actually being a regular contributor in all the communities I've moderated just because it starts to feel like too much and you lose the personal connection with anonymous internet people when you're responsible for regulating behavior and breaking up arguments all the time.
I believe Deimos' idea is more like an algorithmic model for addressing many moderation issues. So it's not a single decision-maker making the call. It's more like each user is scored based on a trust network they have with other users on the site. Users and their trust networks have some sort of weights applied to their labels and votes that help take moderation action when enough of them are applied to a post. There's presumably be mods overseeing it all, but the big challenge will be how to engineer out brigading or just power-user drama where different cliques go to war with each other.
Separation of powers/responsibilities should help with that last bit. Too often when we say 'mod' we mean 'admin' and we should be angling for a wider variety of moderation roles than that. Editor, curator, bailiff, admin, publisher, bot, etc. Vote weights being based on where people spend their time interacting should blunt the effectiveness of brigades. There's no cure or fix for waves of focused user activity, but it certainly doesn't have to make such a mess when it happens.
Really, what will sink or save these ideas is where the bar lies for 'good faith' vs 'bad faith' interactions. Are 99% of the people in the good faith category, and only 1% in the ban-worthy category like we suspect? If so it'll work out well. If it's 80/20 we're going to have a much harder time washing the noise out. The ratio probably changes with the topic of a group, the nature of the users it attracts, and the size of the group too. It's a moving target.
That's definitely part of it, but I think communities can also sort of fall into a trap of becoming too inquisitorial and just spending all their time accusing each other of things or trying to deny accusations about things. Once that starts happening it becomes impossible to parse good and bad faith interaction.
Also, one of the things that was probably the downfall of old Internet forum culture was the tendency of power-user cliques to dominate conversations and intimidating newcomers out of participating because eventually the sheer volume of unspoken norms and inside jokes becomes too much for a noob to learn and they start getting shit on for violating any of them. That was, functionally, its own form of distributed moderation, though it was exercised more through culture and "soft power" instead of admin tools. You'd have to be careful not to replicate that to not just have a few groups of Regina Georges run the streets.
Not sure how you'd do that. I guess there could be some sort of 'sentiment analysis' on your participation to see what you're doing. Like, if you spend all your time being negative or shitting on people it down weights you. If you're welcoming, explaining things to people, deescalating conflicts instead of escalating them, etc. it upweights you. But that's a hard problem.
I admit I cringe a bit at the idea of turning it over to an algorithm for analysis. It's hopelessly subjective. I think there's hope, though. Comment labels are the beginning of the mechanism that can be used to identify and weight those inevitable antics, so that cruft sinks and quality gets a fair shot. Can we get that 99% to use them effectively to guard against those inevitable human behaviors? It's a bit like designing an immune system for the integrity of the discussion. Seems like no other place on the internet except this one even talks about this stuff intelligently. That's one of the things I like the most about this place. :)
More democratic IIRC, like trust is built upon your activity in the group. So not like designated caretakers, but one of 'the longer and more active you've been the more influence you have'
Yeah, kinda like that. I imagine once tildes gets there, there will be a lot of tweaking to find a working setup. Personally, I imagine that users either explicitly or implicitly (votes, agreement) trust other users, who then derive authority over the site from that. Very directly democratic then, where ultimately the user base decides who gets to moderate based upon who they trust. Though I don't think that was explicitly mentioned in the original documents on the topic, that's just my personal interpretation and opinion.
Off-topic but what's the tag under your name for?
You can label a particular comment as exemplary if you feel it deserves to be highlighted and is of particular quality.
It’s similar to the old gilding on Reddit, except no money is involved.
Lol! I just noticed and clicked the label button. Really cool!
Details about how they work and what they're supposed to mean you can see in the documentation:
https://docs.tildes.net/instructions/commenting-on-tildes#labelling-comments
If it ever does, I hope it stays invite only though, and that open registrations happen every once in awhile only temporarily. I guess I'm saying this because I've noticed and learned in my time online being part of the many different places that this is how most communities strive, aside from other things obviously, but keeping a door open at all times generally seems to be a bad idea.
Glad to have you, friend ;)
Freudian, but I'll take it! ;)
Welcome to tildes!