Thoughts on addressing the filter bubble (echo chambers & "fake news"), scalability & free speech
Hi there!
First things first, I just want to say thank you for the invite, but more importantly, thank you for taking the time to create this platform. I, as I imagine most people on here, have a love-hate relationship with reddit. Clearly the site has had a tremendous impact, in many ways positive, but with many things structurally and fundamentally holding it back. I've been a subscriber to /r/RedditAlternatives/ for a while, and there have been very few sites that have compelled me to learn more and actively take part in them, and yours is of course one of them. I just got done reading all of the articles on your docs page and was very pleased - "finally", I thought, someone who's taken into account all of the articles on the internet that have been written about designing and building communities, from both a social and technical perspective, and put it into practice. You've addressed many issues that are often ignored by the platforms themselves and done it in a brilliant way so as to ensure that our voices are heard first and foremost, and I think that's just awesome.
Okay, now that all the praise is out of the way... :P
I did notice something that was not addressed in the docs pages, so I'll be blunt and simply ask: how do you plan to address the filter bubble, or rather, do you plan to address it at all? Maximizing user freedom regarding which communities you want to see content from seems obvious, but that inevitably ends up with users being stuck in their own bubble. reddit already has an infamous reputation of being an echo chamber, and gives users tools to make it an even bigger echo chamber. A long time ago, there was a commonly held belief that the internet would bring us closer together because it would force us to expand our worldviews and interact with people as people, not knowing where they're from or who they are (the "On the Internet, nobody knows you're a dog" saying about anonymity). As reddit moves more and more toward becoming a social network like Facebook and less like the pseudonymous and anonymous internet discussion forums of old, this problem has only gotten worse, to the point of having real-world political and social consequences (especially with the increasing deluge of so-called "fake news"). I'd really like to hear your take on it.
I do have other concerns, namely: scalability, and the stance on free speech. The donation model has worked well for Wikipedia, but, well, they're Wikipedia. They're an incredibly important resource and people have clearly valued their resource so as to have sustained their model, mass donation drives with Jimbo Wales' face plastered all over the site notwithstanding. If tildes becomes the Wikipedia of internet discussion platforms, I am sure many people will find it valuable enough to donate to, though I am still not sold on how sustainable it really is.
The stance on free speech in the announcement blog post also has me concerned. As you mentioned, it is a difficult topic; that much is clear. I am mostly just curious as to where the lines are drawn in regards to how "threats, harassment, and hate speech" are defined. With an absolutist position like "we are 100% pro-free speech", things are very clear and simple, whereas any other position, I believe, comes down to the whim of the moderators/admins. Certainly most people will generally follow the golden rule and abide by basic common sense and decency (i.e. "don't be a dick"), but when discussions get heated I think it's important to not have a reasonable fear that you're going to get permabanned because you hurt someone's feelings (just as an example).
All these issues aside, I am very excited about the development of tildes and hope you & the community can come up with excellent technical and social solutions to these difficult problems.
Thanks for taking the time to read this!
(p.s. apologies for not posting this in the daily discussion topic, thought it warranted its own topic)
edit: formatting
Hey Cornyx, thanks for both the praise and the (difficult) questions. I'll say up front that I certainly don't have true solutions, and a lot of this will have to be adjusted and figured out as we go. I think that's the only way you can really do it for a community site—these are human problems as much as they are tech ones, so we can't expect to solve it by just finding the perfect technological approach.
Filter bubbles are a real problem, and I'm in contact with a few people that have done a lot of research on this topic specifically. It's a difficult one though, because again, it's a very human problem. You can't really force people to be exposed to things if they don't want to be. They'll find ways to avoid them, or (if it's an option) just downvote things they think they disagree with, without even reading them. Even if the site doesn't natively support filtering, people will make extensions that do it anyway, or find other ways to avoid seeing the content they don't like. In the most extreme case, they'll just stop visiting the site entirely if it has too much stuff they aren't interested in seeing and can't avoid it. This isn't to say that we shouldn't try to do anything to help the situation, but it's very difficult to address.
I think it's also important to recognize that, at least on some level, bubbles are kind of what people want out of a community. If I go to a community devoted to metal music, I don't want half the people in there to be constantly talking about how metal is utter garbage that nobody should ever listen to. I just want to enjoy discussing with other fans, not be in a constant debate with attackers. Banning the detractors is basically creating a bubble, but that's also how you create a community instead of a battleground.
As for sustainability from donations, I'm naturally worried about it too, but that's just kind of the nature of starting a business (even a non-profit one)—you can't really be sure if it'll actually work without trying it. I think it's definitely feasible, but nobody's really tried a non-profit community site as far as I know, so we'll have to see what happens. It's worth noting that Wikipedia is probably a lot more successful than you might realize. They're quite good at giving the impression that they're always on the brink of death, but they get about $70 million in donations per year, and have almost $100 million in the bank for "safety". They make far, far more than they truly need to run it.
Also, keeping the site running really isn't a concern at all. The server costs are quite low, are basically already covered from the donations so far, and the server can probably handle something like 100x the traffic we've got now. I'm not concerned at all about having to shut the site down, at this point it's more about whether the donations will be enough for me to be able to pay myself a reasonable amount and keep focusing on Tildes full-time, or if I'll need to find other income and run it as a side-project. Even if that's what happens, it's not a big deal. There aren't any investors, and no obligations to anyone. I can just keep it running and if it ever builds up enough for it to be feasible to work on full-time, I could change back to fully focusing on it at that point.
Finally, related to free speech, I definitely don't want to give the impression that people will get banned if they ever act like a jerk occasionally. It's going to take some really egregious behavior, or a very consistent pattern of being an asshole. The general guideline I've been thinking of is similar to one that I saw in the book The No Asshole Rule. Basically, do other people consistently feel worse after interacting with the person? It's definitely not a perfectly objective criteria or anything, but there's going to be judgment involved no matter what we do.
I hope that helped answer your questions, let me know if there's anything else I can try to clarify.
I don't envy you at all when it comes to figuring out how to deal with "free speech" vs. harassment, etc.
Would something like arbitrators work? Separate from community moderators these could be "trusted" users (and yes, then you open the can of worms of what's a "trusted" user) who volunteer to arbitrate reports of harassment, trolling, etc?
I'm fascinated by watching Tildes evolve in real time. I think y'all are already a step ahead in that respect. It feels transparent and natural.
Do you have plans to have contributions on a open source Tildes code?
Yes, ~ is going to opensource some time in the next week or two AFAIK and the plan is to focus heavily on encouraging and accepting code contributions, so long as the contributions conform to the code quality standards (which there is more written about but just not publicly available yet).
IMO, one of the things reddit did very poorly was their opensource effort (which they shut down 8 months ago); Their documentation was poor, making it hard for outside developers to get familiar with the site codebase and server architecture, and they often left outside developer contributions sitting completely ignored for months, even no-brainer ones like performance and bug fixes.
Whereas on ~ @deimos has already put a great deal of effort into comprehensive documentation, which I plan to help make even easier to understand for new devs through things like step by step guides and system diagrams, and he has also stated that accepting code contributions will be a priority for him.
See here for more on ~ plans for opensource:
https://blog.tildes.net/announcing-tildes#open-honest-and-open-source
https://docs.tildes.net/technical-goals#open-source-as-much-as-possible
https://gitlab.com/tildes
More to come
I know it's not right place to ask this question but is there any place to report bugs. Hard to use for average folks but using gitlab issues maybe? (I have problems about "remember me" check box at the login page. It doesn't remember that I checked it.)
Also, a lot of questions will be asked and answered during the first years of this project. I think something like https://ask.fm/tildesnet may help about that.
We actually already have gitlab set up, so yes, people can report bugs there:
https://gitlab.com/tildes/tildes/issues
Are you using Firefox or another Privacy-centric browser by chance? If so what are your Cookie policies set to? Firefox and many other browsers automatically clear cookies after every session so you may need to whitelist tildes.net for "remember me" to work on your particular browser beyond a single session.
I'm using Firefox but there are no problem in other sites like reddit, twitter, instagram etc... I have noticed that ~ doesn't remember after a couple hours. I'll try in another browser.
I am also using Firefox but have no issue with staying logged in. What are your
Options - Privacy & Security - Cookies and Site Datasettings?Mine are:
● Accept cookies and site data from websites (recommended)
--- Keep until: they expire
--- Accept third-party cookies and site data: never
And in Exceptions I have: https://tildes.net - Allow
That configuration already set like as you explained. I even always
accept third-party cookies and site data. Added specific exception after your previous reply. Still, there is same problem. I haven't tried in another browser yet.Hmmm okay that is an issue then. If you could give some other browsers a try and maybe even go through some of your Firefox extensions trying to isolate which is causing the issue I would greatly appreciate it so I can write a bug report. I have tested the site on Opera, Chrome, Edge and Firefox and have not had this issue, so unless I can find a way to replicate it, it will be difficult to identify the culprit and fix. :(
Thanks for bringing this up, btw. This kind of feedback is exactly what we need most during this alpha stage. :)
You're right. I'll open a ticket in gitlab as soon as I find a replication of the bug.
That would be awesome, thanks!
See also: https://tildes.net/~tildes/xv/two_bugs_keep_me_logged_in_doesnt_work_and_read_replies_still_show_as_new#comment-6dp
I genuinely appreciate the thoughtful questions and effort you put into this post, but posting at 11pm EST on a Saturday after all of us had a pretty insane weekend is probably going to mean you might have to wait until tomorrow until anyone here can answer adequately. I hope you're okay with that. :P
However, in the mean time here are, some responses @deimos gave to similar lines of questioning on HN:
"Can you expand on how you will achieve the middle-ground between safe space and hate speech?"
"What mechanisms are you planning to implement to, for lack of a better phrase, increase exposure to and improve tolerance of different points of view? Is strict moderation the only way to keep groups, or entire sites, from becoming friendly to only one point of view?"
Absolutely, I'm in no rush :)
Thanks for the link & responses!
Regarding the filter bubble. I agree that it is mostly human problem and you cannot enforce people to actively avoid it. But I still believe you can provide technology which nudges people to the intended direction. I built personal recommendation system which do the opposite that most recommendation system does and suggest 3-5 most dissimilar posts from my curated sources of articles.