New socially responsible social network

[EMnut]

Everyone will probably know about the ongoing revelations about how toxic some social media websites are. Data-Mining, lack of privacy, more addictive than crystal meth yadda yadda...

This has been a concern for years amongst a group of London academics, cognitive scientists and philosophers, long before the Cambridge Analytica revelations.
So for the last 2 years they have been developing an alternative, which has just gone live. It is called Lyra, and the link is:

https://www.hellolyra.com

I've been involved with the site design.

Unlike existing social media, this is not a commercial concern, there is no advertising, and data is private and not sold on. The site is funded by donations.
The concept behind it is to encourage civil debate, so the design is not intended to lure people in, and encourage people to fire off comments that they have had barely had any time to think about, but to be more reflective. The conversations work on a tree structure, so users can respond to any comment, and branch off. There are no flashing notifications, and no sharing stuff that often leads to abusive comments or one-up-man ship. (I get shocked at the amount of posts of the 'my bike is better than yours' 'I road X miles faster than you did' stuff that seems to feed dopamine feedback loops.

Conversations can be public, or private (you invite in who you want to talk to and no one else see's it).

As the network has only had a soft launch, we want people to check it out, and add conversation. At the moment the existing posts are quite highbrow, as they are dominated by the people that started the network, but it would be great to see some cycling conversations, or anything else. Lyra is a tool for civil conversation, about anything, so some threads from YACF users would be brilliant.

[Kim]

Sounds a lot like you've re-invented usenet/mailing lists (funded by the users for their own benefit, plain text, properly threaded), without the decentralisation or the established base of client (or for that matter, server) software.

It solves the Big Data you-are-the-product problem, but I don't see anything about moderation or an abuse policy.  What happens when things turn non-excellent?  How do you define non-excellent?

Usenet worked well while it was mostly well-intentioned academics who had access.  Eventually the flamers, the trolls, the systematic abusers, the spammers and the pr0n and warez sharers came along, as they surely will, and beyond suspending accounts and dropping entire groups, there was little that could be done other than full moderation or high-maintenance filtering at the client end.  What tools are there to prevent that sort of thing happening here?

[T42]

URL needs a C.

[fimm]

URL needs a C.

Indeed.
I read "Lyra is a tool for civil conversation..." as "Lycra is a tool for civil conversation..."

[EMnut]

The concept is a social media that encourage civil behaviour due to the design. With all the big social media networks, the posts spew constantly, time frames are minimal, users are pushed into patterns that encourage a swift response. 'My missiles are better than yours'. Image based SM sites cause all manner of issues:
‘Social media has been described as more addictive than cigarettes and alcohol, and is now so entrenched in the lives of young people that it is no longer possible to ignore it when talking about young people’s mental health issues.’
Shirley Cramer CBE, Chief Executive of Royal Society of Public Health

The idea did indeed take cues from usernet/mailing lists, but it practice it works quite differently, there is more control to ring fence privacy. It is not intended to be an obscure site known to a few. How possible it is possible to moderate the site, and stop bullying is impossible to say, but the idea is to stop people from wanting to bully by making the interface calmer avoiding the noise of constant faux-notifications.

I don't think it will stop people from consuming the junk food of current sites, Lyra is the slow food movement if you like. But the intention is to give an alternative free from the user-is-the-product model, with privacy and space to think.

[pcolbeck]

it is possible to moderate the site, and stop bullying is impossible to say, but the idea is to stop people from wanting to bully by making the interface calmer avoiding the noise of constant faux-notifications.

Good luck with that. Usenet abounded with cyber bullying and trolls and that was text only with no ability even for a fancy font never mind flashing anything. I think you are drastically underestimating the ability of some people to be complete s**ts.

[EMnut]

Gosh I have to say, what a bunch of complete cynics.
Did anyone here actually go and look at the site?

[Beardy]

I did. I’d love for it to be a success, as I’m really not that fond of Facebook, but ultimately I use social media to connect to people. If there are others on there whom I can connect with, then great I’ll be there, if there aren’t then I won’t visit that often.

I’ve seen usenet descend into chaos under the weight of trolls and I’ve also seen moderated sites collapse under bullying and hate. But some site manage to survive and even florish, indeed this place grew out of the need move on from usenet and is as strong as ever.

So good luck, and I’ll pop over again later on and have a browse.

[Kim]

Gosh I have to say, what a bunch of complete cynics.
Did anyone here actually go and look at the site?

Yes.  I saw a mail/news style interface that I liked the look of, and I think I read every available page looking for an abuse policy, without finding one.  (I had a similar issue with Mastodon^[1] when it launched.)

I've been using various online communications since the Fidonet days, and I've been the recipient of plenty of abuse over the years, generally for things like expressing a preference for a certain piece of technology, unpopular queer politics, or that perennial favourite of Being A Woman On The Internet™.  In my experience what's made it worse in recent years isn't that modern social platforms provide blingy graphics, or that always-on connectivity encourages users to seek a reaction^[2].  It's a combination of an unpleasant shift in global politics validating certain groups of systematic abusers, and simply that the more people who have access to a medium, the larger the subpopulation of arseholes.

You can seek to keep the troublemakers at bay through obscurity, but that's fundamentally in conflict with a platform being widely enough used by communities with common interests to make it useful.  YACF isn't successful because hardly anybody cares about bikes.  It's successful because it's attracted a handful of niche cycling communities, and because of the hard work of the admin and moderation team.


_{[1] Mastodon comes across as an admirable attempt at "Let's rebuild Twitter without the things that make Twitter broken!", by the sort of people who've only considered the problems that stem from Twitter themselves (user-hostile design decisions, erratic moderation policies, a moving target of an API, the fundamental flaws of a medium controlled by a for-profit company that hasn't worked out what its product is).  The use-case of J Random User confronted with an angry mob of islamophobes/gamergaters/pro-lifers/drivers/cyclists/TERFs/whatever doesn't appear to have been something they've considered at a fundamental design level.
[2] Though I will say that in the days of expensive dialup, when discourse was had over the course of couple of short dialup-and-fetch sessions per evening, and it wasn't unusual for it to take days to get a response to a given post, people did generally take greater time to read things properly, and put a bit more effort into composition.  We're never going back to dial-up, though.}

[mattc]

Kim makes some good points (and clearly has relevant experience to back it up); but I don't think Social Media is as inherently shit as made out. I do believe that the user interface, the design, the general mechanics, all make a difference to behaviour.

I don't have the answers, as human behaviour is complex (and I don't claim to have studied the issues in enough detail). I do know that there are solid peer-reviewed studies* about the effect of things like text colour and sound volume on behaviour - so there must be other effects too.


*Many of which pre-date the internet 

[Kim]

Kim makes some good points (and clearly has relevant experience to back it up); but I don't think Social Media is as inherently shit as made out. I do believe that the user interface, the design, the general mechanics, all make a difference to behaviour.

I agree, and by avoiding the engagement-driven Big Data / advertising business model, you avoid a lot of the worst usability decisions.  Facebook's UI is nasty because it's optimised to show you just as much of the content their advertisers want you to see, and the content they think will keep you coming back to Facebook to see more of it, as you'll tolerate.  Hence to see the photos of your niephews on holiday you've got to scroll past some autoplaying video advert and some friends-of-a-friend arguing about Brexit.  Twitter's rapidly going the same way, at least if you use the website.  Third-party clients avoid this nonsense, but guess what?  They're in the process of closing down the API.

With a completely open system like a web browser or email client or SIP phone, you'd seek to make money by building a better piece of software so more people would choose to pay for it.  It's a business model that still works, but it's a tricky way to get rich.

But if you build a new communications platform that people pay to use, you really want people to use the platform.  You can achieve that to some extent by providing desirable features and hoping to cultivate a dedicated community that appreciates them.  Second Life, for example, is still chugging along because it has a strong community of disabled people and arty types who appreciate it for its unique (if frequently infuriating) features.  I don't think that was what they set out to do

But the easier way to grow is to try to make your communication tool essential for people to communicate with the people they want to communicate with.  Skype would be an example: It's not the only VOIP system, and it's certainly not the best, but it had a good product early on, became the de-facto standard amongst non-business users, protects its customer base by refusing to inter-operate with anything else and now if you want to make a video call to your uncle in elbonia, you've got to use it because he's "not good with computers" and anything else is too complicated.

Pop quiz: How many instant messaging clients did you have installed in 2001?

As a general rule, success of a given system seems to depend mostly on being in the right place in the right time.

[ian]

I'm not sure you can design out human nature. People are shits on the playground, shits on the road, and shits on social media. Short of having a heavily armed nanny overseeing the process, I suspect a proportion of them will be shits.

It's not a case of a better software, it's a case of having the right software with the right cues at the right time for the right people. That's mostly not genius, it's mostly luck. There was nothing really that made Facebook better than all the other contenders that fell by the wayside, or Google the Emperor of Search engines, other than the subtle clockwork of doing all those right things at the right time.

[Beardy]

I'm not sure you can design out human nature. People are shits on the playground, shits on the road, and shits on social media. Short of having a heavily armed nanny overseeing the process, I suspect a proportion of them will be shits.

It's not a case of a better software, it's a case of having the right software with the right cues at the right time for the right people. That's mostly not genius, it's mostly luck. There was nothing really that made Facebook better than all the other contenders that fell by the wayside, or Google the Emperor of Search engines, other than the subtle clockwork of doing all those right things at the right time.

And the same goes for Bill Gates, Steve Jobs and the chap from Oracle who’s name I can never remember. Clever people who worked hard but primarily were in the right place at the right time.

[ian]

People also forget a lot of these businesspeople didn't have an 100% hit rate. Steve Jobs, for instance, was wrong far more than he was right, but he managed to be right at the right times, and thus Apple is very rich and not a footnote to the 1980s. That's not to downgrade some of the smart ideas and hard work, but a lot of people work hard and have great ideas that end up going nowhere. Obviously, history is written by (and for) the winners. That applies to business and war.

But yes, I think the only way you enforce good behaviour is moderation and that's hard work and subjective. One person's troll is someone else's entertainment. Rules arise out of communities and the ability to restrict that community (either through moderation or not capturing the interest of trolls). Of course, most social media creates echo chambers while at the same time amplifying those small signals. There's a good chapter in Jon Ronson's recent book on the poor woman who tweeted a sarcastic comment that social media decided was in fact racist, so during the period of her flights, well, read it. Basically over the space of a few hours of international flight, a sarcastic comment turned into a millions of retweet and a tsunami of hate, bile, and lies.

Alas, it's that echo chamber (people agree with me) and amplification (lots of people agree with me) that make social media so attractive and addictive.

[Kim]

Alas, it's that echo chamber (people agree with me) and amplification (lots of people agree with me) that make social media so attractive and addictive.

Interesting data point:  A friend of mine made a decision a while ago to deliberately influence Youtube's recommendation algorithm to serve up the content they wanted to see, rather than the content they'd tend to engage with.  So they'd deliberately sit all the way through the least irritating adverts, skip political videos entirely, that sort of thing.  Now they get a feed that's all special-interest fanwank, 80s nostalgia, cat videos and whatever, punctuated by arty adverts with gentle instrumental background music or those that simply show-and tell interesting-to-them products.  They hardly even get any Wix.

They've been doing the same thing with Twitter, only engaging with the algorithmically suggested tweets they like, with broadly similar results.  They've deliberately created the extreme echo-chamber all those David Bryn novels warned us about, as a source of light entertainment.  The side effect is that they're completely out of touch with what their friends are mostly tweeting about, let alone society at large.


Engagement-optimising algorithms will tend to promote outrage, because that's what gets and holds people's attention the most.  But it's not all that hard to bias them in different directions if you can somehow tweak the fitness function.

[Jaded]

Moderation is an interesting one. I think it works in small communities, but sheer number of posts leads to the overwhelming of mods. Some of the worst forums for behaviour and moderation I've been on have been the most popular.

I mod on a foofballs forum and helped write The Rules. I'm heartened when neanderthals stumble across the forum and find that they can't generally get a rise from the one, anonymous mod (one account, a few mods), and the posts are largely constructive and free from ****s. People who visit say how different it is from the run of the mill foofballs forum.

What really messes those intent on causing trouble is when threads disappear in heated moments, then get reinstated after 'a review'. (and after the whining about excessive moderation) 

[Beardy]

I had an interesting situation appear on my Facebook feed. A friend of a friend popped up on my timeline and asked to be befriended, which I duly obliged. It turns out that this person was promoting his recently published book of his PhD thesis on anti-Semitism in the labour party. OK, my mind is a broad church, so I went along for the ride. the message got a bit monotonous and at one point I asked if anti-Semitism was a uniquely left thing and got an bit of a bland and unsatisfactory answer and the feeling that such question and answers were 'off message'. I let it drop. Time passed. Then something came up about Israel doing something unsavoury (to my mind) to the Palestinians, which I questioned. I took time to point out that I recognised that the Middle East was a complicated political landscape and that I didn't fully understand the nuances, but I also pointed out that I thought that Israel was an occupying power and a bully to those it could be an to some extent did reap what it sowed. This made ma an anti-Semite in his eyes, so I naturally defended myself and pointed out that I had no problems who people worshiped, and my only 'beef' was with the leadership of the state of Israel. I was then attacked by friends of his as having condemned myself as a total Anti-Semite.
I withdrew from the discussion and didn't engage with him again, but I did leave him on my friends list so as to see a broader perspective. I have however recently given in and unfriended him as he ONLY has one topic, and it is not only monotonous and biased, it was also getting too vitriolic for my liberal sensibilities.
Sorry for the longwindedness of this, but just wanted to point out that even the most liberal of us get bored with single track thinkers from time to time

[EMnut]

OK, I'll take back that comment about cynicism. That is all very useful to hear.

Moderation is a big issue, and whether self moderation will work or not is something to be worked out. Yes everyone has a dark side, we've all typed emails in anger and then deleted them before sending them on. Research shows that particular social media sites have different problems, so sharing can allow showing off, images based sites can create body-image issues with teenagers, twitters text limit just there to generate brain farts. But it is a good point, reducing the features that can create bad behaviour is not going to sop bad behaviour completely, There are always trolls, and a reasoned reply can offend.

Its probably impossible to create a digital utopia, but a useful social network, with content users want to see must be doable.
I don't see Lyra as 'the solution' but is points a way forward.

Would be great to see some cycling conversation appear on it. Best Audax ever etc etc.
Its very early days yet, but diversity of users would be good!

[EMnut]

I've got a better response about moderation (I'd emailed the person who created it to see what they said about moderation.)
I should have thought this earlier, but the site is very hard to troll for these reasons:

'Lyra doesn't need policing - the way conversations and audiences operate have been designed from the ground up to make harassment and abuse difficult. Part of this is due to conversation ownership. Someone causing trouble in your conversation? Remove them from the audience, or set the conversation to private.

Someone making offensive conversations with you in the audience? Simply don't select "World" on the Read page, select some of your groups (into which you've placed pleasant contacts).

The main feature which helps is the absence of any public spaces (forums, hashtags, public groups...). These present very easily exploitable attack surfaces. By not using these, we make harassment much harder.

Moderation is essentially distributed among conversation owners - making people responsible for what goes on in their conversations, and giving them tools to control audiences.

This is one of the differences between existing social media. You can make a post 'private' or 'public', and allow comments from the public or just selected people. The controls can constantly be changed. It sounds simple, but the thinking behind it is very clever.

[Kim]

Ah, like Livejournal (another thing that generally worked better than twitface).  People can still post hateful stuff, but they can't make you read it.

[rogerzilla]

In real life you probably avoid people who have views or behaviours you don't like, or you tolerate them because you have to (e.g. they're family, or co-workers).  There is a kind of echo chamber effect there, too.

What seems to be different online is that discussions aren't private, or among small groups.  It's like standing up in a crowded office and shouting that Tony Blair is a warmonger, or similar.  It draws in people who don't know you very well, have no debating skills and just want to "correct" you for fear that this is what the rest of the group may think is canon.  In real life the analogy would be heckling, possibly followed by brawling, so you don't do it.

So I don't think it can ever work perfectly online, although it can be managed, to varying degrees of dissatisfaction, through moderation or karma-type scoring systems.